92704 v2 Turn Down the Heat Confronting the New Climate Normal Turn Down Heat the Confronting the New Climate Normal © 2014 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved 1 2 3 4 17 16 15 14 This work was prepared for The World Bank by the Potsdam Institute for Climate Impact Research and Climate Analytics. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this commissioned work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution—NonCommercial—NoDerivatives 3.0 IGO license (CC BY-NC-ND 3.0 IGO) http://creativecommons.org/licenses/by-nc-nd/3.0/igo. Under the Creative Commons—NonCommercial—NoDerivatives license, you are free to copy, distribute, and transmit this work, for noncommercial purposes only, under the following conditions: Attribution—Please cite the work as follows: World Bank. 2014. Turn Down the Heat: Confronting the New Climate Normal. Washing- ton, DC: World Bank. License: Creative Commons Attribution—NonCommercial—NoDerivatives 3.0 IGO (CC BY-NC-ND 3.0 IGO). Noncommercial—You may not use this work for commercial purposes. No Derivative Works—You may not alter, transform, or build upon this work. Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third-party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to re-use a component of the work, it is your responsibility to determine whether permission is needed for that re-use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to the Publishing and Knowledge Division, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. The following items are used with permission and require further permission for reuse. Please refer to the caption or note corre- sponding to each item: Figures 2.2, 2.4, 2.9, 3.10, 3.14, 3.15, 3.21, 4.13, 4.14, 4.19, 4.21, 4.22, 5.11, 5.12, 5.13, 5.14, 5.15, 5.16, 5.17, 5.18, 5.19, 5.20, 5.21. ISBN: 978-1-4648-0437-3 Cover photos: photos 1, 2, 3, 5, and 7 © The World Bank Group; photo 4 (forestry), © istockphoto, used with permission, further permission for reuse; photos 6 and 8, © Erick Fernandes (floating houses in Peru and jaguar in Amazon)/The World Bank Group. Cover design: Gregory Wlosinski/General Services Department—Printing and Multimedia, The World Bank Group. Contents Acknowledgments xi Foreword xiii Executive Summary xvii Abbreviations xxxvii Glossary xxxix 1. Introduction 1 1.1 Development Narratives 2 1.2 Methodological Approach 3 1.3 Structure of the Report 3 2. The Global Picture 5 2.1 How Likely is a 4°C World? 5 2.1.1 Can Warming be Held Below 2°C? 6 2.2 Climate Sensitivity and Projected Warming 7 2.3 Patterns of Climate Change 8 2.3.1 Observed Trends in Extreme Events 8 2.3.2 El-Niño/Southern Oscillation 10 2.3.3 Projected Changes in Extreme Temperatures 10 2.3.4 Projected Changes in Extreme Precipitation 14 2.3.5 Aridity and Water Scarcity 14 2.3.6 Droughts 15 2.3.7 Agricultural Yields 16 2.3.8 Ocean Acidification 16 2.4 Sea-Level Rise 18 2.4.1 Marine Ice Sheet Instability 20 2.4.2 Regional Distribution of Sea-Level Rise 20 2.5 Social Vulnerability to Climate Change 22 2.5.1 Interaction of Key Current and Future Development Trends with Climate Change 22 2.5.2 Understanding Vulnerability, Adaptive Capacity, and Resilience 22 2.5.3 Spatial and Physical Vulnerability 22 2.5.4 Socioeconomic Vulnerability 23 2.5.5 Evidence of the Social Implications of Climate Change 26 iii TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL 3. Latin America and the Caribbean 31 3.1 Regional Summary 31 3.1.1 Regional Patterns of Climate Change 31 3.1.2 Regional Sea-Level Rise 33 3.1.3 Sector-based and Thematic Impacts 33 3.1.4 Overview of Regional Development Narratives 36 3.2 Introduction 36 3.2.1 Social, Economic and Demographic Profile of the Latin America and Caribbean Region 38 3.2.2 Vulnerabilities to Climate Change in the Latin America and the Caribbean Region 38 3.2.3 Vulnerabilities Faced by Rural Populations 38 3.2.4 Urban Settlements and Marginalized Populations 39 3.3 Regional Patterns of Climate Change 44 3.3.1 Projected Temperature Changes 44 3.3.2 Heat Extremes 44 3.3.3 Regional Precipitation Projections 47 3.3.4 Extreme Precipitation and Droughts 47 3.3.5 Aridity 49 3.3.6 Tropical Cyclones/Hurricanes 50 3.3.7 Regional Sea-level Rise 53 3.4 Regional Impacts 55 3.4.1 Glacial Retreat and Snowpack Changes 55 3.4.2 Water Resources, Water Security, and Floods 59 3.4.3 Climate Change Impacts on Agriculture 65 3.4.4 Climate Change Impacts on Biodiversity 70 3.4.5 Amazon Rainforest Dieback and Tipping Point 72 3.4.6 Fisheries and Coral Reefs 75 3.4.7 Human Health 80 3.4.8 Migration 82 3.4.9 Human Security 84 3.4.10 Coastal Infrastructure 85 3.4.11 Energy Systems 87 3.5 Regional Development Narratives 92 3.5.1 Overarching Development Narratives 92 3.5.2 Sub-regional Development Narratives 96 3.6 Synthesis Table—Latin America and the Caribbean 99 4. Middle East and North Africa 113 4.1 Regional Summary 113 4.1.1 Regional Patterns of Climate Change 113 4.1.2 Regional Sea-Level Rise 114 4.1.3 Sector-based and Thematic Impacts 115 4.1.4 Overview of Regional Development Narratives 117 4.2 Introduction 117 4.3 Regional Patterns of Climate Change 120 4.3.1 Projected Temperature Changes 120 4.3.2 Heat Extremes 122 4.3.3 Projected Precipitation Changes 124 4.3.4 Extreme Precipitation and Droughts 125 4.3.5 Aridity 125 4.3.6 Regional Sea-level Rise 127 4.4 Regional Impacts 130 4.4.1 The Agriculture-Water-Food Security Nexus 130 4.4.2 Desertification, Salinization, and Dust Storms 136 4.4.3 Human Health 140 iv CONTENTS 4.4.4 Migration and Security 141 4.4.5 Coastal Infrastructure and Tourism 147 4.4.6 Energy Systems 151 4.5 Regional Development Narratives 154 4.5.1 Changing Precipitation Patterns and an Increase in Extreme Heat Pose High Risks to Agricultural Production and Regional Food Security 155 4.5.2 Heat Extremes Will Pose a Significant Challenge for Public Health Across the Region 156 4.5.3 Climate Change Might Act as a Threat Multiplier for the Security Situation 157 4.6 Synthesis Table—Middle East and North Africa 159 5. Europe and Central Asia 169 5.1 Regional Summary 169 5.1.1 Regional Patterns of Climate Change 169 5.1.2 Regional Sea-level Rise 171 5.1.3 Sector-based and Thematic Impacts 171 5.1.4 Overview of Regional Development Narratives 173 5.2 Introduction 174 5.2.1 General Characteristics 174 5.2.2 Socioeconomic Profile of ECA 174 5.3 Regional Patterns of Climate Change 176 5.3.1 Projected Temperature Changes 176 5.3.2 Heat Extremes 177 5.3.3 Regional Precipitation Projections 179 5.3.4 Extreme Precipitation and Droughts 179 5.3.5 Aridity 180 5.3.6 Regional Sea-level Rise 182 5.4 Regional Impacts 182 5.4.1 Water Resources 182 5.4.2 Agricultural Production and Food Security 189 5.4.3 Energy Systems 192 5.4.4 Human Health 194 5.4.5 Security and Migration 195 5.4.6 Russia’s Forests: A Potential Tipping Point? 197 5.5 Regional Development Narratives 204 5.5.1 Impacts on Water Resources in Central Asia Increase the Challenge of Accommodating Competing Water Demands for Agricultural Production and Hydropower Generation 204 5.5.2 Climate Extremes in the Western Balkans Pose Major Risks to Agricultural Systems, Energy and Human Health 205 5.5.3 Responses of Permafrost and the Boreal Forests of the Russian Federation to Climate Change Have Consequences for Timber Productivity and Global Carbon Stocks 207 5.6 Synthesis Table—Europe and Central Asia 209 Appendix 217 A.1 Methods for Temperature, Precipitation, Heat Wave, and Aridity Projections 217 A.1.1 ISI-MIP Bias Correction 217 A.1.2 Heat Extreme Analysis 217 A.1.3 Aridity Index and Potential Evaporation 218 A.1.4 Spatial Averaging 218 A.2 Sea-Level Rise Projections: Methods for This Report 218 A.2.1 Individual Contributions 218 A.2.2 Comparison with Previous Reports and Expert-Elicitation Studies 219 A.3 Meta-analysis of Crop Yield Changes with Climate Change 221 A.3.1 Data Processing 221 v TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL A.3.2 Statistical Analysis 222 A.4 Warming Level Attribution and Classification 222 A.5 Summary of Evidence Concerning Social Vulnerability 223 Bibliography 239 Figures 2.1 Projections for surface-air temperature increase 6 2.2 Climate-model projections of global-mean surface-air temperature 8 2.3 ENSO and extreme events 9 2.4 Idealized schematic showing atmospheric and oceanic conditions of the tropical Pacific region and their interactions during normal conditions, El Niño conditions, and in a warmer world 11 2.5 Multi-model mean global temperature anomaly for RCP2.6 for (2°C world, left) and RCP8.5 (4°C world, right) for the boreal summer months (JJA) 12 2.6 Estimates of world population experiencing highly unusual monthly boreal summer temperatures 13 2.7 Relative change in annual water discharge for a 2°C world and a 4°C world 15 2.8 Percentile change in the occurrence of days under drought conditions 15 2.9 Median yield changes (%) for major crop types in a 4°C world 16 2.10 Global ocean acidification as expressed by a gradual decrease of ocean surface pH (indicating a higher concentration of hydrogen ions–or acidity) 17 2.11 Global mean sea-level rise projection within the 21st century 19 2.12 Patterns of regional sea-level rise 21 2.13 Regional anomaly pattern and its contributions in the median RCP8.5 scenario (4°C world) 22 2.14 Framework for understanding social vulnerability to climate change 23 3.1 Multi-model mean temperature anomaly for Latin America and the Caribbean for RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) for the austral summer months (DJF) 32 3.2 Multi-model mean of the percentage change in the aridity index 33 3.3 Temperature projections for the Latin American and Caribbean land area 44 3.4 Multi-model mean temperature anomaly for Latin America and the Caribbean 45 3.5 Multi-model mean of the percentage of austral summer months (DJF) in the time period 2071–2099 with temperatures greater than 3-sigma (top row) and 5-sigma (bottom row) 46 3.6 Multi-model mean and individual models of the percentage of Latin American and Caribbean land area warmer than 3-sigma (top) and 5-sigma (bottom) 47 3.7 Multi-model mean of the percentage change in austral summer (DJF, top), winter (JJA, middle) and annual (bottom) precipitation 48 3.8 Multi-model mean of the percentage change in the annual-mean of monthly potential evapotranspiration for RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) for Latin America and the Caribbean by 2071–2099 relative to 1951–1980 50 3.9 Multi-model mean of the percentage change in the aridity index 51 3.10 Change in average rate of occurrence of Category 4 and 5 tropical cyclones per hurricane season (August–October) at about 2.5°C warming globally above pre-industrial levels by the end of the 21st century compared to the present-day 53 3.11 Patterns of regional sea-level rise 54 3.12 Regional anomaly pattern and its contributions in the median RCP8.5 scenario 55 3.13 Sea level projections for selected cities 56 3.14 Compilation of mean annual area loss rates for different time periods for glaciated areas between Venezuela and Bolivia 57 3.15 Ice loss from outlet glaciers on the Patagonian Ice Field in southern South America since the Little Ice Age 58 3.16 Cumulative regional surface mass balance relative to the 1986–2005 mean from the model forced with CMIP5 projections up to the year 2100. SLE = Sea-level equivalent 59 3.17 Changes in seasonal total runoff in 4 IPCC climate-change scenarios with respect to the 1961–1990 mean monthly runoff 62 vi CONTENTS 3.18 Aggregate impacts on crop yields in the LAC region with adaptation, computed by the AZS-BioMA platform under 2020 and 2050 NCAR GCM for A1B scenario 68 3.19 Meta-analysis of crop yield reductions 68 3.20 Simulated precipitation changes in Eastern Amazonia from the 24 IPCC-AR4 GCMs with regional warming levels of 2–4.5 K (left panel). Simulated changes in biomass from LPJmL forced by the 24 IPCC-AR4 climate scenarios assuming strong CO2 fertilization effects (middle panel, CLIM+CO2) and no CO2 fertilization effects (CLIM only, right panel) 74 3.21 Change in maximum catch potential for Latin American and Caribbean waters 78 3.22 Sub-regional risks for development in Latin America and the Caribbean (LAC) under 4°C warming in 2100 compared to pre-industrial temperatures 93 4.1 Multi-model mean temperature anomaly for RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) for the months of June-July-August for the Middle East and North African region 114 4.2 Multi-model mean of the percentage change in the aridity index in a 2°C world (left) and a 4°C world (right) for the Middle East and North Africa by 2071–2099 relative to 1951–1980 115 4.3 Temperature projections for the Middle East and North African land area compared to the baseline (1951–1980) 121 4.4 Multi-model mean temperature anomaly for RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) for the months of JJA for the Middle East and North African region 121 4.5 Multi-model mean of the percentage of boreal summer months in the time period 2071–2099, with temperatures greater than 3-sigma (top row) and 5-sigma (bottom row) for scenarios RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) over the Middle East and North Africa 122 4.6 Multi-model mean (thick line) and individual models (thin lines) of the percentage of Middle East and North African land area warmer than 3-sigma (top) and 5-sigma (bottom) during boreal summer months (JJA) for scenarios RCP2.6 (2°C world) and RCP8.5 (4°C world) 123 4.7 Multi-model mean of the percentage change in winter (DJF, top), summer (JJA, middle) and annual (bottom) precipitation for RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) for the Middle East and North Africa by 2071–2099 relative to 1951–1980 124 4.8 Multi-model mean of the percentage change in the annual-mean (ANN) of monthly potential evapotranspiration for RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) for the Middle East and North African region by 2071–99 relative to 1951–80 126 4.9 Multi-model mean of the percentage change in the aridity index under RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) for the Middle East and North Africa by 2071–2099 relative to 1951–1980 126 4.10 Patterns of regional sea-level rise (m) 128 4.11 Regional sea-level rise anomaly pattern and its contributions to the median RCP8.5 scenario (4°C world) 128 4.12 Sea-level projections for Tangier, Tunis, and Alexandria 129 4.13 Major farming systems in the MENA region 130 4.14 Water footprints in m3 per capita and year 131 4.15 Average cereal yields (kilograms per hectare) from 1961–2010 for Northern Africa and Western Asia as compared to the world average 131 4.16 Relative change in annual water discharge in the Middle East and North Africa region in a 4°C world 132 4.17 Meta-analysis of the impact of temperature increase on crop yields 135 4.18 Meta-analysis of the impact of temperature increases on crop yields excluding adaptation and CO2 fertilization 135 4.19 The far-reaching impacts and downward spiral of desertification 137 4.20 Push factors as interrelated drivers for migration and determinants for decision making 142 4.21 The “Arc of Tension” 144 4.22 Food prices and conflict 146 4.23 Aggregated FAO Food Price Index and its sub-indices 147 4.24 Sub-regional risks for development in the Middle East and Northern Africa under 4°C warming in 2100 compared to pre-industrial temperatures 155 5.1 Multi-model mean temperature anomaly for RCP2.6 (2˚C world, left) and RCP8.5 (4˚C world, right) for the months of June-July-August for the Europe and Central Asia region 170 vii TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL 5.2 Multi-model mean of the percentage change in the aridity index (AI) for RCP2.6 (2˚C world ) (left) and RCP8.5 (4˚C world) (right) for the Europe and Central Asia region by 2071–2099 relative to 1951–1980 171 5.3 Temperature projections for the European and Central Asian region 176 5.4 Multi-model mean temperature anomaly for RCP2.6 (2˚C world, left) and RCP8.5 (4˚C world, right) for the months of JJA for the European and Central Asian region 177 5.5 Multi-model mean of the percentage of boreal summer months (JJA) in the time period 2071–2099 with temperatures greater than 3-sigma (top row) and 5-sigma (bottom row) for scenario RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) over the European and Central Asian region 178 5.6 Multi-model mean (thick line) and individual models (thin lines) of the percentage of land area in the European and Central Asian region warmer than 3-sigma (top) and 5-sigma (bottom) during boreal summer months (JJA) for scenarios RCP2.6 (2˚C world) and RCP8.5 (4˚C world) 179 5.7 Multi-model mean of the percentage change in winter (DJF, top), summer (JJA, middle), and annual (bottom) precipitation 180 5.8 Multi-model mean of the percentage change in the annual-mean of monthly potential evapotranspiration for RCP2.6 (2˚C world, left) and RCP8.5 (4˚C world, right) for the European and Central Asian region by 2071–2099 relative to 1951–1980 181 5.9 Multi-model mean of the percentage change in the aridity index (AI) for RCP2.6 (2˚C world, left) and RCP8.5 (4˚C world, right) for the ECA region by 2071–2099 relative to 1951–1980 181 5.10 Sea-level rise projection for Drini-Mati River Delta in Albania 182 5.11 Upstream parts of the Amu and Syr Darya river basin 183 5.12 Losses of glacier area in the Altai-Sayan, Pamir, and Tien Shan 184 5.13 Map of Syr Darya catchment showing mean percentage loss of glacier ice by 2049 relative to 2010 for sub-regions 185 5.14 Decrease in total glacier area in the Amu Darya and Syr Darya basins combined for 2008–2050 based on the CMIP3 (left panel) and CMIP5 (right panel) model runs for the median and extreme values of temperature and precipitation change 185 5.15 Water resources of the Aral Sea basin 186 5.16 Climate change impact on flow of large rivers in Central Asia 187 5.17 Dynamics of surface water-flow structure [in km3] for the Kyrgyz Republic (all rivers) for different temperature-rise scenarios calculated from the difference between the annual sum of atmospheric precipitation and annual evaporation; m-annual sum of precipitation compared to the baseline period 1961–1990 (climate scenario B2-MESSAGE) 188 5.18 River water discharge in the Western Balkans 189 5.19 Dynamics of total area of wildfires in Russia’s forests according to (1) GFDE3 (global fire emissions database); (2) refined data provided by the Institute of Forest, Russian Academy of Sciences (3) Space Research Institute, Russian Academy of Sciences and (4) Vivchar et al. (2010) 200 5.20 Vegetation distribution in Siberia in 2080 from HadCM3 A1FI (leading to a 4°C world) and B1 (leading to a 3°C world) climate change scenarios 201 5.21 Modeled distributions of annual number of high fire danger days across Siberia in the current climate (a) and during the 21st century (b) for HadCM3 A1FI and B1 climate change scenarios 203 5.22 Sub-regional risks for development for Europe and Central Asia at 4°C warming in 2100 compared to pre-industrial temperatures 206 A.1 Comparison of sea level projections by 2081–2100 above the present day, for the current report, previous Turn Down the Heat reports, IPCC reports, and recent subjective expert judgment assessments for a 4°C world (“Experts”: Bamber and Aspinall 2013; Horton et al. 2014) 220 A.2 Same as Figure 6.1 but for a 1.5°C world 221 Tables 2.1 Sea-level rise projections to 2081–2100 above the 1986–2005 baseline 19 2.2 Evidence Summary—Social Vulnerability to Climate Change 27 3.1 Basic Socioeconomic Indicators of LAC Countries 37 3.2 Total Population and Indigenous Population Census 2000 41 viii CONTENTS 3.3 Percentage of Latin American and Caribbean Population Living in Urban Areas and Below Five Meters of Elevation 43 3.4 Multi-model mean of the percentage of land area in Latin America and the Caribbean which is classified as hyper-arid, arid, semi-arid and sub-humid 51 3.5 Sea-level rise between 1986–2005 and 2081–2100 for the RCP2.6 (1.5°C world) and RCP8.5 (4°C world) in selected locations of the LAC region (in meters) 53 3.6 Projected Changes in Yields and Productivity Induced by Climate Change 67 3.7 Summary of Crop Yield Responses to Climate Change, Adaptation Measures, and CO2 Fertilization 68 3.8 Projected losses from sea-level rise 86 3.9 Cumulative loss for the period 2020–2025 for Latin American and Caribbean sub-regions exposed to tropical cyclones 87 3.10 Electricity production from hydroelectric and thermoelectric sources 88 3.11 Projected temperature and hydrologic changes in the Rio Lempa River 89 3.12 Maximum hydropower energy potential 89 3.13 Climate change-related stressors projected to affect hydroelectricity generation 90 3.14 Natural gas production for LAC countries in 2012 and oil production in 2013 90 3.15 Synthesis table of climate change impacts in LAC under different warming levels 99 4.1 Basic socioeconomic indicators of MENA countries 119 4.2 Mean WSDI (Warm Spell Duration Index) for capital cities in the MENA region 123 4.3 Multi-model mean of the percentage of land area in the Middle East and North African region which is classified as hyper arid, arid, semi-arid and sub-humid 127 4.4 Sea-level rise between 1986–2005 and 2080–2099 in selected MENA locations (m) 128 4.5 Rate of sea-level rise in MENA between 2080–2100 128 4.6 Summary of crop yield responses to climate change, adaptation measures, and CO2 fertilization 134 4.7 Damage and people affected by sea-level rise 149 4.8 Results for three scenarios of sea-level rise in the Nile Delta assuming no adaptation 150 4.9 Electricity production from hydroelectric and thermoelectric sources 152 4.10 Synthesis table of climate change impacts in MENA under different warming levels 159 5.1 Basic socioeconomic indicators in ECA countries 175 5.2 Multi-model mean of the percentage of land area in the European and Central Asian region which is classified as Hyper-Arid, Arid, Semi-Arid and Sub-Humid 181 5.3 Sea-level rise (SLR) projection for the Drini-Mati River Delta 182 5.4 Electricity production from hydroelectric and thermoelectric sources 192 5.5 Reduction in usable capacity (expressed in KWmax) of thermal power plants in Europe 194 5.6 Central Asia: projected number of people facing multiple risks from climate change 196 5.7 Synthesis table of climate change impacts in ECA under different warming levels 209 A.1 Climatic classification of regions according to Aridity Index (AI) 218 A.2 Summary of Evidence: Food Security and Nutrition 225 A.3 Summary of Evidence: Poverty Impacts 228 A.4 Summary of Evidence: Migration 230 A.5 Summary of Evidence: Health 232 A.6 Summary of Evidence: Conflict and Security 236 Boxes 1.1 Social Vulnerability 3 1.2 Climate Change Projections, Impacts, and Uncertainty 3 2.1 Definition of Warming Levels and Base Period in this Report 7 2.2 Mechanisms Behind the El-Niño/Southern Oscillation 11 2.3 Heat Extremes 12 2.4 The CO2 Fertilization Effect 17 3.1 Hurricane Mitch’s Impact in Urban Areas 39 3.2 The Case of Mexico City 40 3.3 Water Security in the Mexico City Metropolitan Area 60 3.4 Glacial Lake Outbursts 61 ix TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL 3.5 Water Security in Quito, La Paz, Bogotá, and Lima 62 3.6 Water from the Cordillera Blanca 62 3.7 Water Security in the Central Andes 63 3.8 Water Security and Glacial melt in La Paz and El Alto, Bolivia 63 3.9 Surface Ozone Concentrations 66 3.10 Critical Ecosystem Services of High Andean Mountain Ecosystems 70 3.11 Freshwater Fisheries—Vulnerability Factors to Climate Change 77 3.12 Distress Migration during Hurricane Mitch 84 4.1 Snow Water Storage 133 4.2 The CO2 Fertilization Effect on Drylands 138 4.3 The Nile Delta 149 4.4 Impacts of Climate Change on Tourism in Morocco 151 x Acknowledgments The report Turn Down the Heat: Confronting the New Climate Normal is a result of contributions from a wide range of experts from across the globe. The report follows Turn Down the Heat: Climate Extremes, Regional Impacts and the Case for Resilience, released in June 2013 and Turn Down the Heat: Why a 4°C Warmer World Must be Avoided, released in November 2012. We thank everyone who contributed to its richness and multidisciplinary outlook. The report has been written by a team from the Potsdam Institute for Climate Impact Research and Climate Analytics, including Hans Joachim Schellnhuber, Christopher Reyer, Bill Hare, Katharina Waha, Ilona M. Otto, Olivia Serdeczny, Michiel Schaeffer, Carl-Friedrich Schleußner, Diana Reckien, Rachel Marcus, Oleksandr Kit, Alexander Eden, Sophie Adams, Valentin Aich, Torsten Albrecht, Florent Baarsch, Alice Boit, Nella Canales Trujillo, Matti Cartsburg, Dim Coumou, Marianela Fader, Holger Hoff, Guy Jobbins, Lindsey Jones, Linda Krummenauer, Fanny Langerwisch, Virginie Le Masson, Eva Ludi, Matthias Mengel, Jacob Möhring, Beatrice Mosello, Andrew Norton, Mahé Perette, Paola Pereznieto, Anja Rammig, Julia Reinhardt, Alex Robinson, Marcia Rocha, Boris Sakschewski, Sibyll Schaphoff, Jacob Schewe, Judith Stagl, and Kirsten Thonicke. We acknowledge with gratitude the Overseas Development Institute (ODI) for their contributions to the social vulnerability analysis. The report was commissioned by the World Bank Group’s Climate Change Vice-Presidency. The Bank team, led by Kanta Kumari Rigaud and Erick Fernandes under the supervision of Jane Ebinger, worked closely with the Potsdam Institute for Climate Impact Research and Climate Analytics. The core team comprised of Philippe Ambrosi, Margaret Arnold, Robert Bisset, Charles Joseph Cormier, Stephane Hallegatte, Gabriella Izzi, Daniel Mira-Salama, Maria Sarraf, Jitendra Shah, and Meerim Shakirova. Management oversight was provided by Rachel Kyte, Junaid Ahmad, James Close, Fionna Douglas, Marianne Fay, Ede Ijjasz-Vasquez, Karin Kemper, and Laszlo Lovei. Robert Bisset, Stacy Morford, Annika Ostman, and Venkat Gopalakrishnan led outreach efforts to partners and the media. Samrawit Beyene, Patricia Braxton, Perpetual Boateng and Maria Cristina Sy provided valuable support to the team. Scientific oversight was provided throughout by Rosina Bierbaum (University of Michigan) and Michael MacCracken (Climate Institute, Washington DC). The report benefited greatly from scientific peer reviewers. We would like to thank Pramod Aggarwal, Lisa Alexander, Jens Hesselbjerg Christensen, Carolina Dubeux, Seita Emori, Andrew Friend, Jean-Christophe Gaillard, Jonathan Gregory, Richard Houghton, Jose Marengo, Anand Patwardhan, Scott Power, Venkatachalam Ramaswamy, Tan Rong, Oliver Ruppel, Anatoly Shvidenko, Thomas Stocker, Kevin Trenberth, Carol Turley, Riccardo Valentini, Katharine Vincent, and Justus Wesseler. We are grateful to colleagues from the World Bank Group for their input at key stages of this work: Bachir Abdaym, Gayatri Acharya, Sue Aimee Aguilar, Hanane Ahmed, Kazi Fateha Ahmed, Kulsum Ahmed, Angela Armstrong, Rustam Arstanov, Oscar Avalle, Mary Barton-Dock, Patricia Bliss-Guest, Livia Benavides, Raymond Bourdeaux, Carter Brandon, Adam Broadfoot, Joelle Dehasse Businger, Ludmilla Butenko, Alonso xi TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Zarzar Casis, Tuukka Castren, Térence Céreri, Diji Chandrasekharan, Adriana Damianova, Laurent Debroux, Gerhard Dieterle, Svetlana Edmeades, Ahmed Eiweida, Nathan Lee Engle, Eduardo Ferreira, Homa-Zahra Fotouhi, Luis Garcia, Carolina Diaz Giraldo, Ellen Goldstein, Christophe de Gouvello, Marianne Grosclaude, Nagaraja Rao Harshadeep, Leonard Hessling, Tomoko Hirata, Carlos Felipe Jaramillo, Rahit Khanna, Saroj Kumar Jha, Erika Jorgensen, Steen Lau Jorgensen, Angela Khaminwa, Srilata Kammila, Melanie Kappes, Sunil Khosla, Markus Kostner, Andrea Kutter, Jeffrey Lecksell, Hervé Lévite, Andrea Liverani, Kseniya Lvovsky, Pilar Maisterra, Eugenia Marinova, Benjamin McDonald, Craig Meisner, Nancy Chaarani Meza, Alan Miller, Andrew Mitchell, Nadir Mohammed, Rawleston Moore, Laurent Msellati, Farzona Mukhitdinova, Maja Murisic, John Nash, Kayly Ober, M. Yaa Pokua Afriyie Oppong, Alexandra Ortiz, Nicolas Perrin, Grzegorz Peszko, Elisa Portale, Irina Ramniceanu, Rama Reddy, Nina Rinnerberger, Sandra Lorena Rojas, Alaa Ahmed Sarhan, Daniel Sellen, Bekzod Shamsiev, Sophie Sirtaine, Marina Smetanina, Jitendra Srivastava, Vladimir Stenek, Lada Strelkova, Amal Talbi, Raul Tolmos, Xiaoping Wang, Monika Weber-Fahr, Deborah Wetzel, Gregory Wlosinski, Mei Xie, Emmy Yokoyama, Fabrizio Zarcone, and Wael Zakout. Thanks also to to the following individuals for their support: William Avis, Daniel Farinotti, Gabriel Jordà, Lara Langston, Tom Mitchell, Lena Marie Scheiffele, Xiaoxi Wang, and Emily Wilkinson. We would like to thank Gurbangeldi Allaberdiyev, Zoubeida Bargaoui, Eglantina Bruci, Shamil Iliasov, Hussien Kisswani, Artem Konstantinov, Patrick Linke, Aleksandr Merkushkin, Nasimjon Rajabov, Yelena Smirnova, and Evgeny Utkin for their participation and valuable contributions at the Capacity Building Workshop held in the spring of 2014 that helped inform the report. We acknowledge with gratitude the Climate Investment Funds (CIF), the Energy Sector Management Assistance Program (ESMAP), European Commission, the Italian Government; and the Program on Forests (PROFOR) for their contributions towards the production of this report and associated outreach materials. xii Foreword Dramatic climate changes and weather extremes are already affecting millions of people around the world, damaging crops and coastlines and putting water security at risk. Across the three regions studied in this report, record-breaking temperatures are occurring more fre- quently, rainfall has increased in intensity in some places, while drought-prone regions like the Mediter- ranean are getting dryer. A significant increase in tropical North Atlantic cyclone activity is affecting the Caribbean and Central America. There is growing evidence that warming close to 1.5°C above pre-industrial levels is locked-in to the Earth’s atmospheric system due to past and predicted emissions of greenhouse gases, and climate change impacts such as extreme heat events may now be unavoidable. As the planet warms, climatic conditions, heat and other weather extremes which occur once in hundreds of years, if ever, and considered highly unusual or unprecedented today would become the “new climate normal” as we approach 4°C—a frightening world of increased risks and global instability. The consequences for development would be severe as crop yields decline, water resources change, diseases move into new ranges, and sea levels rise. Ending poverty, increasing global prosperity and reduc- ing global inequality, already difficult, will be much harder with 2°C warming, but at 4°C there is serious doubt whether these goals can be achieved at all. For this report, the third in the Turn Down the Heat series, we turned again to the scientists at the Potsdam Institute for Climate Impact Research and Climate Analytics. We asked them to look at the likely impacts of present day (0.8°C), 2°C and 4°C warming on agricultural production, water resources, cities and ecosystems across Latin America and the Caribbean, Middle East and North Africa, and parts of Europe and Central Asia. Their findings are alarming. In Latin America and the Caribbean, heat extremes and changing precipitation patterns will have adverse effects on agricultural productivity, hydrological regimes and biodiversity. In Brazil, at 2°C warming, crop yields could decrease by up to 70 percent for soybean and up to 50 percent for wheat. Ocean acidification, sea level rise, tropical cyclones and temperature changes will negatively impact coastal livelihoods, tour- ism, health and food and water security, particularly in the Caribbean. Melting glaciers would be a hazard for Andean cities. In the Middle East and North Africa, a large increase in heat-waves combined with warmer average tem- peratures will put intense pressure on already scarce water resources with major consequences for regional food security. Crop yields could decrease by up to 30 percent at 1.5–2°C and by almost 60 percent at 3–4°C. At the same time, migration and climate-related pressure on resources might increase the risk of conflict. In the Western Balkans and Central Asia, reduced water availability in some places becomes a threat as temperatures rise toward 4°C. Melting glaciers in Central Asia and shifts in the timing of water flows xiii TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL will lead to less water resources in summer months and high risks of torrential floods. In the Balkans, a higher risk of drought results in potential declines for crop yields, urban health, and energy generation. In Macedonia, yield losses are projected of up to 50 percent for maize, wheat, vegetables and grapes at 2°C warming. In northern Russia, forest dieback and thawing of permafrost threaten to amplify global warming as stored carbon and methane are released into the atmosphere, giving rise to a self-amplifying feedback loop. Turn Down the Heat: Confronting the New Climate Normal builds on our 2012 report, which concluded the world would warm by 4°C by the end of this century with devastating consequences if we did not take concerted action now. It complements our 2013 report that looked at the potential risks to development under different warming scenarios in Sub-Saharan Africa, South East Asia and South Asia, and which warned that we could experience a 2°C world in our lifetime. Many of the worst projected climate impacts outlined in this latest report could still be avoided by holding warming below 2°C. But, this will require substantial technological, economic, institutional and behavioral change. It will require leadership at every level of society. Today the scientific evidence is overwhelming, and it’s clear that we cannot continue down the current path of unchecked, growing emissions. The good news is that there is a growing consensus on what it will take to make changes to the unsustainable path we are currently on. More and more voices are arguing that is possible to grow greener without necessarily growing slower. Today, we know that action is urgently needed on climate change, but it does not have to come at the expense of economic growth. We need smart policy choices that stimulate a shift to clean public transport and energy efficiency in factories, buildings and appliances can achieve both growth and climate benefits. This last report in the Turn Down the Heat series comes at a critical moment. Earlier this year, the UN Secretary General’s Climate Summit unleased a new wave of optimism. But our reports make clear that time is of the essence. Governments will gather first in Lima and then Paris for critical negotiations on a new climate treaty. Inside and outside of the conference halls, global leaders will need to take difficult decisions that will require, in some instances, short term sacrifice but ultimately lead to long term gains for all. At the World Bank Group we will use our financial capacity to help tackle climate change. We will innovate and bring forward new financial instruments. We will use our knowledge and our convening power. We will use our evidence and data to advocate and persuade. In short, we will do everything we can to help countries and communities build resilience and adapt to the climate impacts already being felt today and ensure that finance flows to where it is most needed. Our response to the challenge of climate change will define the legacy of our generation. The stakes have never been higher. Dr. Jim Yong Kim President, World Bank Group xiv Executive Summary Executive Summary The data show that dramatic climate changes, heat and weather extremes are already impacting people, damaging crops and coastlines and putting food, water, and energy security at risk. Across the three regions studied in this report, record-breaking temperatures are occurring more frequently, rainfall has increased in intensity in some places, while drought-prone regions are getting dryer. In an overview of social vulnerability, the poor and underprivileged, as well as the elderly and children, are found to be often hit the hardest. There is growing evidence, that even with very ambitious mitigation action, warming close to 1.5°C above pre-industrial levels by mid-century is already locked-in to the Earth’s atmospheric system and climate change impacts such as extreme heat events may now be unavoidable.1 If the planet continues warming to 4°C, climatic conditions, heat and other weather extremes considered highly unusual or unprecedented today would become the new climate normal—a world of increased risks and instability. The consequences for development would be severe as crop yields decline, water resources change, diseases move into new ranges, and sea levels rise. The task of promoting human development, of ending poverty, increasing global prosperity, and reducing global inequality will be very challenging in a 2°C world, but in a 4°C world there is serious doubt whether this can be achieved at all. Immediate steps are needed to help countries adapt to the climate impacts being felt today and the unavoidable consequences of a rapidly warming world. The benefits of strong, early action on climate change, action that follows clean, low carbon pathways and avoids locking in unsustainable growth strategies, far outweigh the costs. Many of the worst projected climate impacts could still be avoided by holding warming to below 2°C. But, the time to act is now. This report focuses on the risks of climate change to development in Latin America and the Caribbean, the Middle East and North Africa, and parts of Europe and Central Asia. Building on earlier Turn Down the Heat reports this new scientific analysis examines the likely impacts of present day (0.8°C), 2°C and 4°C warming above pre-industrial temperatures on agricultural production, water resources, ecosystem services and coastal vulnerability for affected populations. Scope of the Report Central Asia (ECA).3 The focus is on the risks of climate change to development. While covering a range of sectors, special attention This third report in the Turn Down the Heat series2 covers three is paid to projected impacts on food and energy systems, water World Bank regions: Latin America and the Caribbean (LAC); the resources, and ecosystem services. The report also considers the Middle East and North Africa (MENA); and parts of Europe and social vulnerability that could magnify or moderate the climate 1 Holding warming to below 2°C and bringing warming back to 1.5°C by 2100 is technically and economically feasible but implies stringent mitigation over the short term. While IPCC AR5 WGIII identified many mitigation options to hold warming below 2°C with a likely chance, and with central estimates of 1.5–1.7°C by 2100, only “a limited number of studies have explored scenarios that are more likely than not to bring temperature change back to below 1.5°C by 2100”. The scenarios in these studies are “charac- terized by (1) immediate mitigation action; (2) the rapid upscaling of the full portfolio of mitigation technologies; and (3) development along a low-energy demand trajectory”. 2 Turn Down the Heat: Why a 4°C Warmer World Must be Avoided, launched by the World Bank in November 2012; and Turn Down the Heat: Climate Extremes, Regional Impacts, and the Case for Resilience, launched by the World Bank in June 2013 constitute the first two reports. 3 The World Bank Europe and Central Asia region in this report includes only the following countries: Albania, Bosnia and Herzegovina, Kazakhstan, Kosovo, the Kyrgyz Republic, the former Yugoslav Republic of Macedonia, Montenegro, the Russian Federation, Serbia, Tajikistan, Turkmenistan, and Uzbekistan. xvii TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL change repercussions for human well-being. The report comple- ments the first Turn Down the Heat report (2012) that offered a Box 1: The Case for Immediate global overview of climate change and its impacts in a 4°C world4 Action and concluded that impacts are expected to be felt disproportion- ately in developing countries around the equatorial regions. Also, CO2 emissions continue unabated. Current warming is at 0.8°C it extends the analysis in the second report (2013) that focused above pre-industrial levels. CO2 emissions are now 60 percent on the consequences of climate change for present day, 2°C, and higher than in 1990, growing at about 2.5 percent per year. If emis- 4°C warming above pre-industrial levels in Sub-Saharan Africa, sions continue at this rate, atmospheric CO2 concentrations in line South Asia, and South East Asia and demonstrated the potential with a likely chance of limiting warming to 2°C would be exceeded within just three decades. of early onset of impacts at lower levels of warming. Observed impacts and damages. Widespread, recently This analysis draws on the Intergovernmental Panel on Cli- observed impacts on natural and human systems confirm the high mate Change (IPCC) Fifth Assessment Report (AR5) Working sensitivity of many of these systems to warming and the potential Group reports released in 2013 and 2014, as well as peer-reviewed for substantial damage to occur at even low levels of warming. literature published after the cutoff dates for AR5. The few cases Examples include negative impacts on crop yields, the accelerating where there are significant differences in interpretation of projected loss of ice from Antarctica and Greenland, and widespread bleach- impacts from the IPCC assessments (such as for sea-level rise and ing of coral reefs. The physical effects of warming to 1.5°C, such as El Niño) are highlighted and explained. extreme heat events, may be unavoidable. 21st-century projected impacts. The projected impacts The Global Picture for the 21st century confirm the scale of the risk to development at 2°C—and the severe consequences of exceeding this level of This report reaffirms earlier assessments, including the IPCC AR5 warming. Even at warming of 1.5°C–2°C, significant, adverse risks are projected for a number of regions and systems, such as the and previous Turn Down the Heat reports, that in the absence of potential for the complete loss of existing long-lived coral reefs, near-term mitigation actions and further commitments to reduce associated marine biodiversity and the livelihoods from tourism and emissions the likelihood of 4°C warming being reached or exceeded fishing. this century has increased. Under current policies there is about Multi-century consequences of 21st-century emissions. a 40 percent chance of exceeding 4°C by 2100 and a 10 percent Scientific evidence is growing of the multi-century consequences chance of exceeding 5°C.5 However, many of the worst projected of CO2 and other greenhouse gas emissions. Examples include: climate impacts in this report could still be avoided by holding ‘locking-in’ a long-term sea-level rise of about two meters per warming below 2°C. degree Celsius of sustained global mean warming and a multi-cen- tury ocean acidification with wide-ranging adverse consequences on Selected Key Findings from Across coral reefs, marine ecology, and ultimately the planet. Risk of large-scale, irreversible changes in the Earth’s the Regions biomes and ecosystems. Large scale, irreversible changes in the Earth’s systems have the potential to transform whole regions. At the current level of 0.8°C warming above pre-industrial levels, Examples of risks that are increasing rapidly with warming include adverse impacts of climate change have already been observed. degradation of the Amazon rainforest with the potential for large Examples include: emissions of CO2 due to self-amplifying feedbacks, disintegration of • Extreme heat events are occurring more frequently. The occur- the Greenland and Antarctic ice sheets with multi-meter sea-level rence of record-breaking monthly mean temperatures has rise over centuries to millennia, and large-scale releases of methane been attributed to climate change with 80 percent probability. from melting permafrost substantially amplifying warming. Recent peer reviewed science shows that a substantial part of the West Antarctic ice sheet, containing about one meter of sea-level rise 4 In this report, as in the previous two reports, “a 4°C world” and “a 2°C world” is equivalent in ice, is now in irreversible, unstable retreat. used as shorthand for warming reaching 4°C or 2°C above pre-industrial levels by Rapidly closing window for action. The buildup of carbon the end of the century. It is important to note that, in the case of 4°C warming, this intensive, fossil-fuel-based infrastructure is locking us into a future of does not imply a stabilization of temperatures nor that the magnitude of impacts is CO2 emissions. The International Energy Agency (IEA) has warned, expected to peak at this level. Because of the slow response of the climate system, the greenhouse gas emissions and concentrations that would lead to warming of 4°C and numerous energy system modelling exercises have confirmed, by 2100 and associated higher risk of thresholds in the climate system being crossed, that unless urgent action is taken very soon, it will become extremely would actually commit the world to much higher warming, exceeding 6°C or more costly to reduce emissions fast enough to hold warming below 2°C. in the long term with several meters of sea-level rise ultimately associated with this warming. A 2°C world implies stabilization at this level beyond 2100. 5 IEA (2012) World Energy Outlook 2012. This was reported in the second Turn Down the Heat report. xviii E X ECU TI VE S U MMA RY • Extreme precipitation has increased in frequency and intensity 4°C. With earlier glacier melt in Central Asia shifting in many places. the timing of water flows, and a higher risk of drought in the Balkans, this carries consequences for crop yields, • A robust drying trend has been observed for already drought- urban health, and energy generation. In Macedonia, for prone regions such as the Mediterranean. example, there could be yield losses of up to 50 percent • A significant increase in tropical North Atlantic cyclone activity for maize, wheat, vegetables and grapes at 2°C warm- has been observed and is affecting the Caribbean and Central ing. Flood risk is expected to increase slightly along the America. Danube, Sava and Tisza rivers. Under future climate change scenarios projected impacts include: 3. Agricultural yields and food security: Significant crop yield impacts are already being felt at 0.8°C warming, and as 1. Highly unusual and unprecedented heat extremes: State- temperatures rise from 2°C to 4°C, climate change will add of-the-art climate modeling shows that extreme heat events further pressure on agricultural systems. increase not only in frequency but also impact a larger area • The risks of reduced crop yields and production losses of land under unabated warming. The prevalence of highly increase rapidly above 1.5°–2°C warming. In the Middle unusual and unprecedented heat extremes increases rapidly East and North Africa and the Latin America and the under an emissions pathway associated with a 4°C world.6 Caribbean regions, without further adaptation actions, Highly unusual heat extremes are similar to those experienced strong reductions in potential yield are projected for in Russia and Central Asia in 2010 and the United States in 2012 around 2°C warming. For example, a 30–70 percent and unprecedented heat extremes refer to events essentially decline in yield for soybeans and up to 50 percent decline absent under present day climate conditions. Unprecedented for wheat in Brazil, a 50 percent decrease for wheat in heat extremes would likely remain largely absent in a 2°C world Central America and the Caribbean, and 10–50 percent but in a 4°C world, could affect 70–80 percent of the land area reduction for wheat in Tunisia. Projected changes in in the Middle East and North Africa and Latin America and potential crop yields in Central Asia are uncertain at the Caribbean and approximately 55 percent of the land area around 2°C warming. Increasing droughts and flood- in the parts of Europe and Central Asia assessed in this report. ing events represent a major risk for agriculture in the 2. Rainfall regime changes and water availability: Precipitation Western Balkans. changes are projected under continued warming with sub- • While adaptation interventions and CO2 fertilization may stantial, adverse consequences for water availability. Central compensate for some of the adverse effects of climate America, the Caribbean, the Western Balkans, and the Middle change below 2°C warming, this report reaffirms the East and North Africa stand out as hotspots where precipitation findings of the IPCC AR5 that under 3–4°C warming is projected to decline 20–50 percent in a 4°C world. Conversely, large negative impacts on agricultural productivity can heavy precipitation events are projected to intensify in Central be expected. There is some empirical evidence that, and Eastern Siberia and northwestern South America with despite possible positive CO2 fertilization effects lead- precipitation intensity increasing by around 30 percent and ing to increased productivity, higher atmospheric levels flooding risks increasing substantially in a 4°C world. of carbon dioxide could result in lowered protein and • In the Western Balkans and Central Asia, water avail- micronutrient (iron and zinc) levels of some major grain ability becomes a threat as temperatures rise toward crops (e.g., wheat and rice). • The projected impacts on subsistence and export crops production systems (e.g., soybeans, maize, wheat, and 6 In this report, highly unusual heat extremes refer to 3-sigma events and unprec- rice) would be felt at the local, national, and global levels. edented heat extremes to 5-sigma events. In general, the standard deviation (sigma) shows how far a variable tends to deviate from its mean value, which in this report While global trade can improve food security and pro- refers to the possible year-to-year changes in local monthly temperature because of tect against local shocks, there is a possibility for some natural variability. For a normal distribution, 3 sigma events have a return time of regions to become over dependent on food imports and 740 years. Monthly temperature data do not necessarily follow a normal distribution thus more vulnerable to weather events in other world (for example, the distribution can have long tails, making warm events more likely) and the return times can be different but will be at least 100 years. Nevertheless, regions and to the interruption of imports because of 3-sigma events are extremely unlikely and 4-sigma events almost certainly have not export bans in those regions. occurred over the lifetime of key infrastructure. A warming of 5 sigma means that the average change in the climate is 5 times larger than the normal year-to-year 4. Terrestrial Ecosystems: Ecosystem shifts are projected with variation experienced today, and has a return period of several million years. These events, which have almost certainly never occurred to date, are projected for the increasing temperatures and changes in precipitation patterns coming decades. significantly diminishing ecosystem services. This would xix TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL have major repercussions on, for example, the global carbon 2081–2100 compared to the reference period 1986–2005.8 Due cycle. For example: to the time lag in the oceans’ response and the long response • Projected increases in heat and drought stress, together time of the Greenland and Antarctic ice sheets to atmospheric with continuing deforestation, substantially increase the temperatures (thermal inertia) sea levels will continue to rise risk of large-scale forest degradation (reduction in forest for many centuries beyond 2100. biomass and area) in the Amazon rainforest. This could • Sea-level rise poses a particular threat to urban communi- turn this carbon sink of global importance into a source of ties in the Middle East and North Africa and Latin America carbon; this has already been observed as a consequence and the Caribbean, where large urban settlements and of the severe droughts in 2005 and 2010 when scientists important infrastructure are situated along coastlines. estimated that the Amazon faced a decrease in carbon The impact of rising sea levels will be particularly severe storage of approximately 1.6 Pg carbon (2005) and 2.2 Pg for the Caribbean island communities as possibilities carbon (2010) compared to non-drought years.7 for retreat are extremely limited. Rising sea levels will • Russia’s permafrost regions and boreal forests are sensitive substantially increase the risk posed by storm surges and to changes in temperature that could lead to productivity tropical cyclones, in particular for highly exposed small increases. But there is a risk of increasing disturbances, island states and low-lying coastal zones. In addition, such as fires and pests, leading to widespread tree mortal- rising sea levels could contribute to increased salt-water ity. Forest dieback and thawing of permafrost threaten to intrusion in freshwater aquifers (particularly in the Middle amplify global warming as stored carbon and methane East and North Africa), a process made worse by other are released into the atmosphere, giving rise to a self- climate impacts (e.g., reduction in water availability) and amplifying feedback loop. With a 2°C warming, methane other human-induced drivers (e.g., resource overuse). emissions from permafrost thawing could increase by 7. Glaciers: A substantial loss of glacier volume and extent has 20–30 percent across boreal Russia. been observed under current levels of warming in the Andes 5. Marine ecosystems: Substantial, adverse effects on marine and Central Asia. Increasing glacial melt poses a high risk of ecosystems and their productivity are expected with rising flooding and severely reduces freshwater resources during temperatures, increases in ocean acidity, and likely reductions crop growing seasons. It can also have negative impacts on in available oxygen due to their combined effects. Observed hydropower supply. rates of ocean acidification are already the highest in 300 million • Tropical glaciers in the Central Andes have lost large years and rates of sea level rise are the highest for 6,000 years. amounts of ice volume throughout the 20th century and Projections of coral bleaching indicate that preserving complete deglaciation is projected in a 4°C world. In more than 10 percent of these unique ecosystems calls for Peru it is estimated that a 50 percent reduction in glacier limiting global warming to 1.5°C. Reef-building corals are runoff would result in a decrease in annual power output critical for beach formation, coastal protection, fisheries, of approximately 10 percent, from 1540 gigawatt hours and tourism. (GWh) to 1250 GWh. Physiological changes to fish and fish larvae have been • Since the 1960s Central Asian glaciers have reduced observed and are expected with future ocean acidification. in area by 3–14 percent depending on their location. Below 2°C warming and without taking into account changes Further substantial losses of around 50 percent and up in ocean acidity, fishery catches in a number of locations are to 80 percent are projected for a 2°C and a 4°C world projected to markedly decrease by 2050 as fish populations respectively. As a result, river flows are expected to shrink migrate towards cooler waters. 6. Sea-level rise: In a 1.5°C world sea level rise is projected to increase by 0.36 m (range of 0.20 m to 0.60 m) and by 0.58 m 8 The sea-level projections presented here follow the methodology adopted in the IPCC (range of 0.40 m to 1.01 m) in a 4°C world for the period AR5 WGI with the important update that more realistic scenario-dependent contribu- tions from Antarctica based on post-IPCC literature are included. Recent publications suggest that IPCC estimates are conservative given the observed destabilization of parts of the West Antarctic Ice Sheet. Note that the regional projections given in this report are also based on this adjustment to the AR5 WGI methodology and do not include land subsidence. Sea-level rise projections presented in this report are based 7 The change in carbon sequestration is caused by the combined effects of reduced on a larger model ensemble with an ensemble mean warming of less than 1.75°C; uptake of carbon resulting from suppressed tree growth due to the drought, and loss as a result, end-of-century sea-level rise in RCP2.6 is classified as 1.5° warming. See of carbon due to drought induced tree mortality and decomposition over several years. Box 2.1 and Section 6.2, Sea-Level Rise Projections for further explanation. xx E X ECU TI VE S U MMA RY Figure 1: Water resources: Relative change in annual discharge for a 2°C and a 4°C world in the 2080s relative to the 1986–2005 period based on an ISI-MIP model inter-comparison. Colors indicate the multi-model mean change; the saturation of colors indicates the agreement across the model ensemble. More saturated colors indicate higher model agreement. Source: Adjusted from Schewe et al. (2013). by 25 percent at around 3°C warming during the summer 8. Social Vulnerability to Climate Change. The social impacts months when water demand for agriculture is highest. of climate change are hard to predict with certainty as they • In Central Asia hydropower generation has the potential depend on climatic factors and their interaction with wider to play a major role in the future energy mix however the development trends. However, there is clear evidence that predicted changes in runoff distribution will mean that climate change is already affecting livelihoods and wellbe- there will be less water available for energy generation ing in parts of the three regions and is likely to do so to a in summer months when it will compete with demands significantly greater extent if more extensive climate change from agriculture. occurs (Box 2). Where governance is weak, or infrastructure xxi TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Box 2: Social Vulnerability Impacts of Climate Change Shocks and stresses related to climate change can undermine poverty reduction and push new groups into poverty. Informal settlements on flood plains and steep hillsides in many Latin American cities and the Western Balkans, for example, have been severely affected by floods and landslides in recent years. While many poor people will be living in isolated, rural areas, continued urban expansion into hazard-prone areas means that a growing proportion of urban populations will be at risk of climate-related extreme events and rising food prices, and thus of increasing poverty levels among urban groups. The impacts of climate change will often be most severely felt by poor and socially excluded groups, whose capacity to adapt to both rapid- and slow-onset climate change is more limited. These include indigenous people and ethnic minorities, migrant workers, women, girls, older people, and children. Although these groups—like their more advantaged counterparts—are already adapting to climatic and other changes, these efforts are often undermined by their limited assets, lack of voice, and discriminatory social norms. For example, increasing water stress, projected for parts of Latin America and low-income Middle East and North Africa countries, can dramatically increase the labor burden asso- ciated with fetching water in rural and poor urban environments; and child malnutrition linked to climate change reducing protein and micronutrient contents of staple foods (wheat, rice) could have irreversible, negative life-time consequences for affected children. Climate change may lead to displacement and also affect patterns and rates of migrations. Most displacement related to extreme weather events has, to date, been temporary. However, if climate change renders certain areas uninhabitable (for example, if they become too hot, too dry, or too frequently affected by extreme events—or sea-level rise) such migration may increase in scale and more often lead to permanent resettlement (as already being seen in some water-scarce parts of MENA). Large-scale migration may pose significant challenges for family relations, health, and human security. There is a risk of disadvantaged groups being trapped in adversely affected rural areas as they lack the funds and/or social connections to move. outdated or insufficient (as in parts of all three regions), Changes to the hydrological cycle could endanger this is likely to amplify the social challenges associated with the stability of freshwater supplies and ecosystem adapting to further climate change. services. Altered precipitation cycles characterized by more intense down- pours followed by longer droughts, loss of glaciers, degradation of Latin America and the Caribbean key ecosystems, and the loss of critical ecosystem services (e.g., water supplies, water buffering, retention, regulation, and soil The Latin American and the Caribbean (LAC) region is highly protection) will impact freshwater supplies regionally and poten- heterogeneous in terms of economic development and social and tially generate upstream-downstream tradeoffs and synergies. A indigenous history with a population of 588 million (2013), of range of impacts are expected to increase in intensity and severity which almost 80 percent is urban. The current GDP is estimated as global mean temperatures rise from 2oC to 4oC. at $5.655 trillion (2013) with a per capita GNI of $9,314 in 2013. • Projections indicate that most dry regions get drier and wet In 2012, approximately 25 percent of the population was living regions get wetter. Reductions in precipitation are as high in poverty and 12 percent in extreme poverty, representing a as 20–40 percent for the Caribbean, Central America, central clear decrease compared to earlier years. Undernourishment in Brazil, and Patagonia in a 4°C world. Drought conditions are the region, for example, declined from 14.6 percent in 1990 to projected to increase by more than 20 percent. Limiting warming 8.3 percent in 2012. Despite considerable economic and social to 2°C is projected to reduce the risk of drought significantly: development progress in past decades, income inequality in the to a one percent increase of days with drought conditions in region remains high. the Caribbean and a nine percent increase for South America. At the current 0.8oC warming significant impacts of climate At the same time, an increase in frequency and intensity of change are being felt throughout the LAC region’s terrestrial extreme precipitation events is projected particularly for the (e.g. Andean mountains and rainforests) and marine (especially tropical and subtropical Pacific coastline and southern Brazil. the coral reefs) biomes. As global mean temperatures rise to 2oC and beyond the projected intensity and severity of impacts will • Massive loss of glaciers is projected in the Andes in a 2°C increase across the entire region (three significant impacts are world (up to 90 percent) and almost complete glacier loss described below). beyond 4°C. Changes to glacial melt, in response to land Figure 2 shows the occurrence of highly unusual summer surface warming, alter the timing and magnitude of river temperatures in a 2oC and 4oC world. Box 3 gives an overview of flows and result in a higher risk of flooding and freshwater the climate risks in the region. shortages and damage to infrastructure assets. xxii E X ECU TI VE S U MMA RY Figure 2: Multi-model mean of the percentage of austral summer months (DJF), with highly unusual temperatures (normally unlikely to occur more than once in several hundred years) in a 2°C world (left) and a 4°C world (right) in 2071–2099 and relative to the 1951–1980 base line period. • Increased droughts and higher mean temperatures are rainfall events can quickly overwhelm natural drainage chan- projected to decrease water supplies and affect most ecosys- nels in the landscape as well as urban drainage systems that tems and agroecosystems. The increasing risk of drought will are unlikely to have been designed for the projected more raise the risk of forest fires, large-scale climate-induced forest intense future rainfall events and flows. degradation and the loss of associated ecosystem services. • Glaciers will melt at an even faster rate than observed, with Climate change will place at risk small-scale a peak in river runoff expected in 20–50 years, and possibly subsistence agriculture and large-scale agricultural earlier in some watersheds. Glacial lake outbursts and con- production for export nected flooding present a hazard for Andean cities. The loss of Agriculture in the Latin America and the Caribbean region is glaciers will likely impact the páramos (Andean, high carbon heavily dependent on rain-fed systems for both subsistence and stock moorlands) which are the source of water for many export crops; it is therefore vulnerable to climatic variations Andean cities. Moreover, degraded highland ecosystems have such as droughts, changing precipitation patterns, and rising less capacity to retain water and intensified downpours will temperatures. increase erosion with a subsequent increase in siltation and • Increasing risks for agriculture as warming rises beyond damage to hydropower dams, irrigation works, and hydraulic 2°C. There is a clear negative signal for a large variety of crops and river defense infrastructure. with 2°C warming, including soybeans (up to a 70 percent yield • The projected trend of more intense rainfall can significantly decline in some areas of Brazil) and maize (up to a 60 percent increase the risk of landslides especially in sloping terrain yield decline in Brazil and Ecuador) by 2050 relative to a often occupied by the poorer rural and urban communities. 1989–2009 baseline. Simulated adaptation interventions (e.g. The major landslides in 2011 in the State of Rio de Janeiro improved crop varieties, improved soil and crop management, following intense rainfall are a harbinger of the likely severity and supplementary irrigation) alleviated but did not overcome of projected impacts from more intense rainfall events. Intense the projected yield declines from climate change. Other studies xxiii TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Box 3: Selected Climate Risks in the Latin America and the Caribbean Region In a 4°C world, heat extremes, changes in hydrological cycles, tropical cyclones and changes in the El Niño Southern Oscillation (ENSO) are expected to pose severe problems with risks cascading to the agricultural sector, human health, large urban centers and the functioning of critical ecosystem services. At lower levels of warming, glacial melt in the Andes will reduce freshwater and hydropower for communities and large Andean cities during the dry season, while increasing the risks of flooding in the short term and impacting agriculture and environmental services down- stream. Severe threats are expected from sea-level rise, damages to low-lying areas and coastal infrastructures. Degrading coral reefs will endanger tourism revenues and undermine biodiversity, fisheries, and the protection of coastal zones thereby negatively impacting livelihoods. For the global community, the potential impact of climate change on the Amazon rainforest is of high relevance. With increasing warming, degradation—if not dieback—of the Amazon rainforest is increasingly possible potentially turning the forest into a large carbon source during dry years and triggering further climate change. Central America & the Caribbean Higher ENSO and tropical cyclone frequency, precipitation extremes, drought, and heat Dry Regions waves. Risks of reduced water availability, crop yields, food security, and coastal safety. Poor exposed to landslides, coastal erosion Caribbean with risk of higher mortality rates and migration, Central America negative impacts on GDP where share of coastal tourism is high. Amazon Rainforest Increase in extreme heat and aridity, risk of forest fires, degradation, and biodiversity loss. Amazon Rainforest Risk of rainforest turning into carbon source. Shifting agricultural zones may lead to conflict over land. Risks of species extinction threatening Dry Regions traditional livelihoods and cultural losses. Andes Andes Glacial melt, snow pack changes, risks of flooding, and freshwater shortages. 3RSXODWLRQ'HQVLW\ >3HRSOHSHUVTNP@ In high altitudes women, children, and Southern Cone indigenous people particularly vulnerable; and  agriculture at risk. In urban areas the poor living on steeper slopes more exposed to flooding. ² ² Dry Regions ² Increasing drought and extreme heat events leading to cattle death, crop yield declines, and ² challenges for freshwater resources. Falkland Islands (Islas Malvinas)  A DISPUTE CONCERNING SOVEREIGNTY OVER THE Risks of localized famines among remote ISLANDS EXISTS BETWEEN ARGENTINA WHICH CLAIMS THIS SOVEREIGNTY AND THE U.K. WHICH ADMINISTERS THE ISLANDS. indigenous communities, water-related health problems. Stress on resources may lead to Data sources: Center for International Earth Science Information Network, Columbia University; United conflict and urban migration. Nations Food and Agriculture Programme; and Centro Internacional de Agricultura Tropical—(2005). Gridded Population of the World, Version 3 (GPWv3): Population Count Grid. Palisades, NY: NASA Socioeconomic Southern Cone Data and Applications Center (SEDAC). This map was reproduced by the Map Design Unit of The World Bank. The boundaries, colors, denominations and any other information shown on this map do not imply, Decreasing agricultural yields and pasture on the part of The World Bank Group, any judgment on the legal status of any territory, or any endorse- productivity, northward migration of agro- ment or acceptance of such boundaries. ecozones. Risks for nutritious status of the local poor. Risks for food price increases and cascading impacts beyond the region due to high export share of agriculture. xxiv E X ECU TI VE S U MMA RY suggest that in a 3°C world, the projected negative impacts on individual crops become stronger. For example up to almost Box 4: El Niño Southern Oscillation 70 percent decline in wheat in Central America and the Carib- (ENSO) bean. This implies that climate change threatens not only smallholder farmers, and rural and indigenous communities The Latin America and the Caribbean region is particularly but also large-scale commodity (soybeans, maize) producers, exposed to the effects of strong* El Niño and La Niña events, ranchers, and agribusinesses—with potential negative repercus- which are related to the El Niño Southern Oscillation (ENSO). In sions on food security and prices in the region and beyond. Central America, El Niño usually results in excessive rainfall along the Caribbean coasts, while the Pacific coasts remain dry. Rainfall • Local food security is seriously threatened by the projected increases and floods tend to occur on the coast of Ecuador, the decrease in fishery catch potential. The Caribbean coasts, northern part of Peru, and in the southern part of Brazil, Argentina, the Amazon estuaries, and the Rio de la Plata are expected to Paraguay, and Uruguay while drought appears in the Andean zones be particularly affected by declines in catch potential of more of Ecuador, Peru and Bolivia and in north eastern Brazil. All these than 50 percent as fish stocks migrate in response to warming changes can substantially impair livelihoods through impacts on waters. The Caribbean waters could see declines in the range agricultural productivity, critical ecosystems, energy production, of 5–50 percent. These estimates are for warming of 2°C by water supply, infrastructure, and public health in affected countries. For example, the extreme 1997–98 El Niño event resulted in many 2050, by which time many of the coral reefs—an important fish billions of dollars in economic damages, and tens of thousands nursery and habitat—would be subject to annual bleaching of fatalities worldwide, including severe losses in Latin America. events, further undermining the marine resource base. Ocean Substantial uncertainties remain regarding climate change impact acidification could affect fish populations directly, including projections on the intensity and frequency of extreme El Niño events. through physiological damages at early life stages. The effects However, evidence of changes to ENSO-driven precipitation vari- on the food chain, however, are not yet clear. ability in response to global warming has emerged recently and • The Southern Cone (Chile, Argentina, Uruguay, Paraguay, represents an update to the assessment of ENSO projections in the and southern Brazil) as a major grain and livestock produc- IPCC AR5 report. Recent climate model inter-comparison studies suggest the likelihood of global warming leading to the occurrence of ing region is susceptible to climate shocks, mainly related more frequent extreme El Niño events over the 21st century. to changing rainfall patterns and rising heat extremes. This is expected to severely impact maize and soy yields, important * “The Oceanic Niño Index (ONI) is the standard that NOAA uses for identify- export commodities. For example, maize productivity is pro- ing El Niño (warm) and La Niña (cool) events in the tropical Pacific. It is the jected to decline by 15–30 percent in comparison to 1971–2000 running 3-month mean sea-surface temperature (SST) anomaly for the Niño 3.4 region (i.e., 5oN–5oS, 120o–170oW). Events are defined as 5 consecutive levels under warming of 2°C by 2050, and by 30–45 percent overlapping 3-month periods at or above the +0.5o anomaly for warm (El under 3°C warming. Strong and/or extreme El Niño events Niño) events and at or below the –0.5 anomaly for cold (La Niña) events. The resulting in floods or droughts in the cropping season pose threshold is further broken down into Weak (with a 0.5 to 0.9 SST anomaly), Moderate (1.0 to 1.4) and Strong (Ն 1.5) events” [Source: http://ggweather further substantial risks to agriculture in the region. .com/enso/oni.htm] A stronger prevalence of extreme events is projected that would affect both rural and urban communities, particularly on sloping lands and in • Risks associated with El Niño events and tropical cyclones coastal regions. would occur contemporaneously with a sea-level rise of The region is heavily exposed to the effects of more frequent and 38–114 cm thus greatly increasing the risks storm surges. intense extreme events, such as those that occur during strong El Sea-level rise is projected to be higher at the Atlantic coast than Niño events and tropical cyclones. at the Pacific coast. Sea-level rise off Valparaiso, for example, is • An increase of approximately 40 percent in the frequency of projected at 0.35 m for a 2°C world and 0.55 m for a 4°C world the strongest north Atlantic tropical cyclones is projected for (medium estimate). Recife sees projections of approximately a 2°C world, and of 80 percent for a 4°C world, compared to 0.39 m and 0.63 m respectively, with the upper estimates as present. In LAC, close to 8.5 million people live in the path of high as 1.14 m in a 4°C world—the highest in the region. hurricanes, and roughly 29 million live in low-elevation coastal • Extreme events will strongly affect the rural and urban poor zones. The Caribbean is particularly vulnerable as more than who often reside in informal settlements in high-risk areas 50 percent of its population lives along the coast, and around (e.g., flood plains and steep slopes). In 2005, the percentage 70 percent live in coastal cities. More intense tropical cyclones of people living in informal settlements in Latin America was would interact adversely with rising sea levels, exacerbating highest in Bolivia (50 percent) and in the Caribbean highest coastal flooding and storm surge risks, putting entire econo- in Haiti (70 percent). The negative effects of extreme events mies and livelihoods at risk (particularly for island states). xxv TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL also affect rural communities as they strongly depend on their energy markets among the most heavily subsidized in the world. environment and its natural resource base. The region is very diverse in terms of socio-economic and political conditions. Thus, adaptive capacity and vulnerability to climate risks • In the Caribbean, substantial adverse impacts on local criti- varies enormously within the region, especially between the Arab cal ecosystems, agriculture, infrastructure, and the tourism Gulf States and the other Middle East and North Africa countries. industry can be expected in a 2°C world. This is due to loss The Middle East and North Africa region emerges as one of the and/or degradation of important assets from the combined effects hotspots for worsening extreme heat, drought, and aridity condi- of increasing sea levels and associated impacts of saline intru- tions. Agriculture, where 70 percent is rain-fed, is highly exposed sion and storm surges, ocean acidification, bleaching of coral to changing climatic conditions. Warming of 0.2°C per decade has reefs, and loss of the physical protection afforded to coastlines been observed in the region from the 1961–1990, and since then from dead and degrading reefs. Impacts from these and other the region is warming at an even faster rate. Projections indicate climatic changes can be expected to grow substantially with that more than 90 percent of summers will have highly unusual increasing warming, especially given the increasing likelihood heat extremes in a 4°C world compared to between 20–40 percent of more frequent intense tropical cyclones. of summers in a 2°C world (Figure 3). Given its high import dependency, the region is vulnerable The Middle East and North Africa to effects beyond its borders. While societal responses to such changes remain hard to predict, it is clear that extreme impacts, The Middle East and North Africa (MENA) is one of the most such as a more than 45 percent decrease in annual water discharge diverse regions in the world in economic terms, with per-capita projected for a 4°C world in parts of the region, would present annual GDP ranging from $1,000 in Yemen to more than $20,000 unprecedented challenges to the social systems affected. Climate in the Arab Gulf States. Qatar, Kuwait, the United Arab Emirates, change might act as a threat multiplier to the security situation Morocco, the Arab Republic of Egypt, and Yemen rank 4, 12, 27, in the region by imposing additional pressures on already scarce 130, 132, and 151 in GDP per capita on a list of 189 countries. In resources and by reinforcing pre-existing threats connected to consequence, adaptive capacity and vulnerability to climate risks migration following forced displacement. Box 5 gives an overview varies enormously within the region. of the key climate risks in the region. The region’s population is projected to double by 2050, which together with projected climate impacts, puts the region under Changing precipitation patterns and an increase enormous pressure for water and other resources. The region in extreme heat pose high risks to agricultural is already highly dependent on food imports. Approximately production and regional food security. 50 percent of regional wheat and barley consumption, 40 percent Most agriculture in the region takes place in the semi-arid climate of rice consumption, and nearly 70 percent of maize consump- zone, either close to the coast or in the highlands, and is highly tion is met through imports. The region has coped with its water vulnerable to the effects of climate change. scarcity through a variety of means: abstraction of groundwater, • Rainfall is predicted to decline by 20–40 percent in a 2°C desalinization, and local community coping strategies. Despite world and by up to 60 percent in a 4°C world in parts of the its extreme water scarcity, the Gulf countries use more water per region. Agricultural productivity is expected to drop in parts capita than the global average, with Arab residential water and of the Middle East and North Africa region with increasing Figure 3: Multi-model mean of the percentage of boreal summer months (JJA), with highly unusual temperatures (normally unlikely to occur more than once in several hundred years) in a 2°C world (left) and a 4°C world (right) in 2071–2099 and relative to the 1951–1980 base line period. xxvi E X ECU TI VE S U MMA RY Box 5: Selected Climate Risks in the Middle East and North Africa Region The region will be severely affected at 2°C and 4°C warming, particularly because of the large increase in projected heat extremes, the substantial reduction in water availability, and expected consequences for regional food security. High exposure to sea-level rise in the coming decades is linked to large populations and assets in coastal areas. In a 2°C world already low annual river discharge levels are projected to drop by more than 15 percent and highly unusual heat extremes are projected to affect about a third of the land. Crop yield declines coupled with impacts in other grain-producing regions could contribute to increasing food prices in the region. The growing food import dependency further exacerbates such risks. Deteriorating rural livelihoods may contribute to internal and international migration, adding further stress on particularly urban infrastructure, with associated risks for poor migrants. Migration and climate related pressure on resources (e.g. water) might increase the risk of conflict. e k a s h r M Maghreb &HQWUDO$UDE 3HQLQVXOD 3RSXODWLRQ'HQVLW\ >3HRSOHSHUVTNP@  Southern ² Arab Peninsula ² ² ²  Maghreb Mashrek and Eastern Parts Arabian Peninsula Strong warming reduction in annual Highly unusual heat and decrease in annual Highly unusual heat extremes in central precipitation, increased water stress and precipitation will increase aridity, decrease Arabian Peninsula. In southern parts reduced agricultural productivity. Large in snow water storage and river runoff for relative increase in annual precipitation, but coastal cities exposed to sea level rise. example in Jordon, Euphrates and Tigris. uncertain trend of annual precipitation in Adverse consequences for mostly rain-fed central part. Sea level rise in the Arabian Climate change risks will have severe agricultural and food production. Sea likely higher than in Mediterranean implications on farmers’ livelihoods, country and Atlantic coasts with risk of storm economy, and food security. Exposure of Climate change risks will have severe surges and adverse consequences for critical coastal assets would have impact implications on farmers’ livelihoods, country infrastructure. on the economy, including tourism. There is economy, and food security. There is a risk risk for accelerated migration flows to urban for accelerated migration flows to urban More heat extremes expected to increase areas and social conflict. areas and social conflict. thermal discomfort, posing risk to labor productivity and health. Data sources: Center for International Earth Science Information Network, Columbia University; United Nations Food and Agriculture Programme; and Centro Internacional de Agricultura Tropical—(2005). Gridded Population of the World, Version 3 (GPWv3): Population Count Grid. Palisades, NY: NASA Socioeconomic Data and Applications Center (SEDAC). This map was reproduced by the Map Design Unit of The World Bank. The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of The World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. xxvii TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL water scarcity and higher temperatures which are expected to extremes would occur on average in one of the summer deviate more and more from the temperature optima of several months in each year from the 2040s onward. In a 4°C world, crops (and possibly even exceed their heat tolerance levels). this frequency would be experienced as early as the 2030s, and would increase to two summer months by the 2060s and • Crop yields in the region may decrease by up to 30 percent virtually all months by the end of the century. Unprecedented at 1.5–2°C warming in Jordan, Egypt, and Libya and by almost heat extremes are absent in a 2°C world and affect about half 60 percent (for wheat) at 3–4°C warming in the Syrian Arab the summer months by the end of the century in a 4°C world. Republic. The strongest crop reductions are expected for legumes and maize as they are grown during the summer period. • The period of consecutive hot days is expected to increase, particularly in cities due to the urban heat island effect. For • With 70 percent of agricultural production being rain-fed, the example, in a 2°C warmer world the number of consecutive hot sector is highly vulnerable to temperature and precipitation days is projected to increase annually from four days to about changes and the associated potential consequences for food, two months in Amman, from eight days to about three months social security, and rural livelihoods. Forty-three percent of in Baghdad, and from one day to two months in Damascus. the population lives in rural areas and poor rural farmers are The number of hot days in Riyadh is expected to increase particularly vulnerable to hunger and malnutrition as a direct even more—from about three days to over four months. The consequence of yield loss and high food prices. In combination number of hot days in a 4°C warmer world is projected to with non-climatic pressures the decline in rural livelihood options exceed the equivalent of four months in most capital cities. could trigger further urban migration, potentially exacerbating urban vulnerability and intensifying the potential for conflict. • Heat stress levels can approach the physiological limits of people working outdoors and severely undermine regional • The increase in demand for irrigation water will be dif- labor productivity, putting a burden on health infrastruc- ficult to meet due to the simultaneous decrease in water ture. High temperatures can cause heat-related illnesses (e.g., availability in the Middle East and North Africa countries heat stress, heat exhaustion, and heat stroke) especially for which have traditionally invested in agriculture to improve the elderly, people with chronic diseases or obesity, and preg- the performance in the agriculture sector—about 30 percent of nant women, children, and people working outside. Climate the agricultural land is irrigated whereas agriculture consumes change is expected to undermine human health in other ways 60 to 90 percent of all water used in the region. as well. For instance, the relative risk of diarrheal disease as • Rising food prices that often follow production shocks and a consequence of climatic changes and deteriorating water long-term declines make the growing number of urban poor quality is expected to increase 6–14 percent for the period increasingly vulnerable to malnutrition, particularly against the 2010–39 and 16–38 percent for the period 2070–99 in North background of increasing local food insecurity. Evidence suggests Africa; and 6–15 percent and 17–41 percent, respectively, in that child malnutrition could rise in the event of significant food the Middle East. price increases or sharp declines in yields. Child malnutrition is already high in parts of the Middle East and North Africa, with an Sea-level rise will pose serious challenges to the average of 18 percent of children under age five developmentally region’s coastal population, infrastructure, and stunted. Childhood stunting has been linked to lifelong adverse economic assets. consequences, including lower economic productivity in adulthood. The dense concentration of people and assets in coastal cities • With its high and growing import dependency the region is translate into high exposure to the effects of sea-level rise. particularly vulnerable to worldwide and domestic agricul- • Projections show that all coastlines are at risk from sea-level tural impacts and related spikes in food prices. For example, rise. Depending on the city, sea levels are projected to rise climatic and hydrological events (droughts and floods), together by 0.34–0.39 m in a 1.5°C world and 0.56–0.64 m in a 4°C with global market forces, were contributing factors to high world (medium estimate), with the highest estimate reaching wheat prices in Egypt and affected the price of bread in 2008. 1.04 m in Muscat. • The Maghreb countries of Egypt, Tunisia, Morocco, and Libya Heat extremes will pose a significant challenge have been identified as among the most exposed African for human health countries in terms of total population affected by sea-level People in the region face a variety of health risks, many of which rise. In Morocco, for example, more than 60 percent of the are exacerbated by the hot and arid conditions and relative water population and over 90 percent of industry is located in key scarcity that characterize the region. coastal cities. For example, Alexandria, Benghazi, and Algiers • A substantial rise in highly unusual heat extremes is expected have been identified as particularly vulnerable to a sea-level in the coming decades. In a 2°C world, highly unusual heat rise of only 0.2 m by 2050. The United Arab Emirates also xxviii E X ECU TI VE S U MMA RY ranks among the ten most vulnerable countries to sea-level to the agriculture-water-energy nexus in Central Asia; climate rise worldwide. extreme in the Western Balkans, and the forests in Russia. While the economic and political profiles of the countries differ greatly, a • Key impacts of climate change in coastal zones include common denominator is their transition from various types of closed, inundation resulting from slow onset sea-level rise, flooding, planned economies to open, market-based systems. The region is and damages caused by extreme events (including storms, characterized by relatively low levels of per-capita annual GDP, storm surges, and increased coastal erosion). The exposure ranging from $800 in Tajikistan to $14,000 in Russia. Agricultural of critical assets may cause other impacts to have repercus- production plays an important role in the national economies of sions for the economy (e.g., where tourism infrastructure the region, particularly those of Tajikistan, the Kyrgyz Republic, is exposed). In Egypt, for example, the ocean acidification Uzbekistan, and Albania. Large portions of the population in Cen- and ocean warming threatens coral reefs and is expected to tral Asia (60 percent) and the Western Balkans (45 percent) live in place the tourism industry—an important source of income rural areas, making them dependent on natural resources for their revenue—under severe pressure. livelihoods and thus particularly vulnerable to climate change. • Impacts on groundwater levels are significant, with poten- The parts of the Europe and Central Asia region covered by tial negative repercussions on human health for local and this report are projected to experience greater warming than the migrant populations. The Nile Delta, home to more than 35 global average. The region displays a clear pattern where areas million people and providing 63 percent of Egypt’s agricultural in the southwest are becoming drier and areas further northeast, production, is especially vulnerable to salinization under chang- including most of Central Asia, are becoming wetter as the world ing climate conditions. These impacts will be exacerbated by warms toward 4°C. The projected temperature and precipitation land subsidence, especially in the eastern part of the delta, changes translate into increased risks for freshwater supplies and by extensive landscape modification resulting from both that not only jeopardizes the sustainability of hydropower and coastal modification and changes in the Nile’s hydrogeology. agricultural productivity but also negatively impacts ecosystem services such as carbon sequestration for most of the region. A Europe and Central Asia selection of sub-regional impacts is provided in Box 6. Europe and Central Asia (ECA) in this report covers 12 countries9 Water resources in Central Asia increase during within Central Asia, the Western Balkans, and the Russian Federa- the first half of the century and decline thereafter, tion. The analysis focuses on specific climate challenges related amplifying the challenge of accommodating competing water demands for agricultural production and hydropower generation. 9 The World Bank Europe and Central Asia region in this report includes only the Water resource systems in Central Asia (notably glaciers and following countries: Albania, Bosnia and Herzegovina, Kazakhstan, Kosovo, the Kyrgyz Republic, the former Yugoslav Republic of Macedonia, Montenegro, the snow pack) are sensitive to projected warming; with consequent Russian Federation, Serbia, Tajikistan, Turkmenistan, and Uzbekistan. impacts on water availability in the agriculture and energy sectors. Figure 4: Multi-model mean of the percentage of boreal summer months (JJA) with highly unusual temperatures (normally unlikely to occur more than once in several hundred years) in a 2°C world (left) and a 4°C world (right) in 2071–2099 and relative to the 1951–1980 base line period. xxix TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Box 6: Selected Climate Risks in the Europe and Central Asia Region Increasing precipitation and glacial melt lead to increased water availability and flood risk in Central Asia in the coming decades. After mid-century and especially with warming leading to a 4°C world, unstable water availability poses a risk for agriculture and competing demands for hydropower generation. In the Western Balkans, extreme heat with a strong decrease in precipitation and water availability are projected to lead to large reduc- tions in crop yields, adverse effects on human health, and increasing risks to energy generation for a 4°C world; but would already be present in a 2°C world. The Russian forests store enormous amounts of carbon in biomass and soils. While their productivity may generally increase with warmer temperatures, large-scale forest dieback and the release of carbon resulting from interacting heat stress, insect spread and fire, have the potential to further affect boreal forests in the second half of the century. s i a n F e d e r a t i R u s o n 3RSXODWLRQ'HQVLW\ >3HRSOHSHUVTNP@  l Asia ² Centra Western ² Balkans ² ²  Western Balkans Central Asia Boreal Forests of the Russian Federation Increase in droughts, unusual heat Increasing glacial melt alters river runoff. extremes and flooding. High risks for Risks of glacial lake outbursts, flooding Unusual heat extremes and annual agriculture, human health and stable and seasonal water shortages. Increasing precipitation increase, rising risks of forest hydropower generation. competition for water resources due to fires and spread of pests leading to tree rising agricultural water demand and mortality and decreasing forest productivity. Risks for human health, food and energy demand for energy production. Possible northward shift of treeline and security. changes in species composition. Risks of Risks for poor through rising food prices permafrost melt and methane release. particularly affecting women, children and the urban poor. Risks for human health Risk for timber production and ecosystem due to spreading disease, heat waves and services, including carbon capture. Risks of flooding. substantial carbon and methane emissions. Data sources: Center for International Earth Science Information Network, Columbia University; United Nations Food and Agriculture Programme; and Centro Internacional de Agricultura Tropical—(2005). Gridded Population of the World, Version 3 (GPWv3): Population Count Grid. Palisades, NY: NASA Socioeconomic Data and Applications Center (SEDAC). This map was reproduced by the Map Design Unit of The World Bank. The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of The World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. xxx E X ECU TI VE S U MMA RY Central Asia is increasingly likely to be a hotspot for heat stress for Kazakhstan. Overall, energy demand is projected to rise for agriculture and human settlements as warming proceeds to together with population and economic growth. 2°C and 4°C especially as temperatures are not moderated by • Tajikistan and the Kyrgyz Republic, which are located upstream oceanic winds. Since the beginning of the 20th century, Central of the Syr Daria and Amu Darya, produce nearly 99 percent Asian glaciers have already seen a one-third reduction in glacier and 93 percent, respectively, of the total electricity consumed volume. Glacier volume is projected to decline by about 50 percent from hydropower. These upstream countries would have to in a 2oC world, concurrent with a 25 percent decrease in snow manage the impact of climate change on their hydropower cover for the Northern Hemisphere, and by up to 80 percent in a generation capacity, which is the backbone of their power 4°C world. Reductions in water availability are predicted to occur systems; downstream countries (Kazakhstan, Uzbekistan, and contemporaneously with an increase in demand for irrigation water. Turkmenistan), meanwhile, would be hit particularly hard by • River runoff will increase in the coming decades due to competing demands for agricultural and energy production. enhanced glacial melt rates but flows are expected to decrease for the second half of the century. By the end of Climate extremes in the Western Balkans pose the 21st century a distinct decrease in the water volume of major risks to agricultural systems, energy and the Syr Darya, and an even more distinct decrease in the Amu human health. Darya River, is expected because of declining glaciers that sup- The Western Balkans are particularly exposed to the effects of ply most of its flows. Critically, also the timing of high flow extreme events, including heat, droughts, and flooding. Heat volumes changes. For example, available data for a headwater extremes will be the new norm for the Western Balkans in a 4°C catchment (Panj) of the Amu Darya River reveal that the timing world. In a 2°C world, highly unusual heat extremes are projected of peak flows is projected to shift toward spring, leading to for nearly a third of all summer months compared to virtually all a 25 percent reduction in discharge during the mid-summer summer months in a 4°C world. Unprecedented heat extremes are (July-August) period in a 3°C world. As a result, less water projected to occur for 5–10 percent of summer months in a 2°C world will be available for agriculture during the crop-growing season compared to about two-thirds of summer months in a 4°C world. while at the same time higher summer temperatures lead to • The risk of drought is high. Projections indicate a 20 per- higher water demand for plants. cent increase in the number of drought days and a decrease • Crop productivity is expected to be negatively impacted by in precipitation of about 20–30 percent in a 4°C world. increased heat extremes and variability of supply/demand Projections for a 2°C world are uncertain. At the same time, for water that poses substantial risks to irrigated agricultural projections suggest an increase in riverine flood risk, mainly systems. Rain-fed agriculture is likely to be affected by uncer- in spring and winter, caused by more intense snow melt in tain rainfall patterns and amounts, including where irrigation spring and increased rainfall in the winter months (precipita- is important, and coupled with rising maximum temperatures tion projections are, however, particularly uncertain). can lead to the risk of heat stress and crop failure. • Most crops are rain-fed and very vulnerable to projected • Rural populations that are especially dependent on agri- climate change. While there are no projections that encompass culture for food are likely to be increasingly vulnerable to the entire region, and projections for individual countries remain any reductions in agricultural yields and nutritional quality uncertain, clear risks emerge. For example, projections for FYR of their staple food grains. Macedonia indicate potential yield losses of up to 50 percent • Unstable water availability is likely to increase the chal- for maize, wheat, vegetables, and grapes for around 2°C global lenge of competing requirements for hydropower genera- warming by 2050. Pasture yields and grassland ecosystems for tion and agricultural production at times of rising overall livestock grazing may be affected by sustained drought and demand due to projected population and economic growth in heat, and decline over large parts of the Western Balkans. Central Asia. The projected increase in highly unusual and The effects of extreme events on agricultural production are unprecedented heat extremes during the summer months (see mostly not included in assessments, but observations indicate Figure 4) can be expected to simultaneously increase energy high vulnerability. demand. As the efficiency of hydropower plants depends on • Energy systems are very vulnerable to extreme events and inter- and intra-annually stable river runoff, the potential, for changes in river water temperatures; changing seasonality example, of installed hydropower plants for small catchments of river flows can further impact hydropower production. is projected to decrease by 13 percent in Turkmenistan and by Most countries in the Western Balkans depend on hydroelectric 19 percent in the Kyrgyz Republic at around 2°C warming by sources for at least one-fifth of their electricity production. the 2050s, while an increase of nearly seven percent is projected Reductions in electricity production would be concurrent with xxxi TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL an increase in cooling demand which is projected to increase • At lower latitudes the forest is likely to give way to steppe by 49 percent in a 4°C world. ecosystems. If (partly uncertain) CO2 fertilization effects do • Extreme climate events and the appearance of new disease not enhance water-use efficiency sufficiently, the risk of fire, vectors pose serious risks to human health. The increased particularly in southern Siberia and Central Yakutia, will increase incidence and intensity of extreme heat events could cause and could lead to increased carbon emissions. Projections for the seasonality of temperature-related mortality to shift from this area indicate an increase in the annual number of high winter to summer across continental Europe. Albania and fire danger days of an average of 10 days in a 3°C world, and the Former Yugoslav Republic of Macedonia are considered 20–30 days in a 4°C world. The effects of heat waves promoting particularly vulnerable to heat waves. The net total number forest fires, and the increasing spread of pests and diseases, as of temperature-related deaths is projected to increase for the well as the interaction of these factors, may lead to decreased period 2050–2100 above 2°C warming levels. Further health productivity and even increased tree mortality. risks are likely due to climate change resulting in favorable • In a 2°C world, the thawing of the permafrost is projected conditions for the insect vectors transmitting diseases, such to increase methane emissions by 20–30 percent. The as dengue fever and Chikungunya fever. projected perturbations to Russian forest ecosystems are of global importance. If pushed beyond critical thresholds and Impacts of projected warming on Russian boreal into positive feedback with regional and global warming, forests and the permafrost can have severe large carbon stocks in the boreal forests and methane in the consequences for forest productivity and global permafrost zones may be released into the atmosphere—with carbon stocks. major implications for the global carbon budget. The boreal ecosystems of the Russian Federation that account for about 20 percent of the world´s forest cover large permafrost regions (carbon and methane-rich frozen soil layers) are likely to be quite Consequences for Development sensitive to projected warming and heat extremes. Perturbations Climate change risks undermining development and to the forest or permafrost could result in severe consequences for poverty reduction for present and future generations local ecosystem services and the global carbon budget. Although Climate change poses a substantial and escalating risk to develop- slightly warmer average temperatures may increase forest produc- ment progress that could undermine global efforts to eliminate tivity, there is a risk of increasing disturbances, such as fires and extreme poverty and promote shared prosperity. Without strong, pests, leading to widespread tree mortality. early action, warming could exceed 1.5–2°C and the resulting Above-average temperature rises and an overall increase in impacts could significantly worsen intra- and inter-generational annual precipitation is projected. In a 2°C world, highly unusual poverty in multiple regions across the globe. heat extremes are projected to occur in 5–10 percent of summer Severe threats to development outlined in this report are months, increasing to 50 percent of all summer months in a 4oC beginning to occur across many sectors in all three regions. The world. Precipitation is expected to increase by 10–30 percent in a analysis presented in this report reveals that amplified risks are 2°C world and by 20–60 percent in a 4°C world. Permafrost in the emerging from multi-sectoral impacts in particular connected to region is highly vulnerable to warming, with projected permafrost food security due to projected large and severe crop yield losses thawing rate of 10–15 percent over Russia by 2050 in a 2°C world. for warming levels above 2°C. • A northward shift of the tree line is projected in response As warming approaches 4oC very severe impacts can be to warming, causing boreal forests to spread into the northern expected to trigger impact cascades crossing critical thresholds of tundra zone, temperate forests into the present boreal zone, environmental and human support systems. Climatic conditions, and steppes (grassland plains) into temperate forests. In a 4°C heat and other weather extremes considered highly unusual or world the Eurasian boreal forest area would reduce around unprecedented today would become the new climate normal—a 19 percent and the temperate forest area increase by over world of increased risks and instability. 250 percent. With warming limited to around 1.5°C, boreal Every effort must be made to cut greenhouse gas emissions forests would decrease by around two percent and the tem- from our cities, land use, and energy systems now and transition to perate forest area would increase by 140 percent. This would a clean, low carbon pathway. Action is urgently needed on climate lead to a net gain in total temperate and boreal forest area in change, but it does not have to come at the expense of economic Eurasia of seven percent in a 4°C world and 12 percent in a growth. Immediate steps are also needed to help countries build 1.5°C world. The potential carbon gains from the expansion resilience and adapt to the climate impacts being felt today and of boreal forests in the north are likely to be offset, however, the unavoidable consequences of a rapidly warming world over by losses in the south. the coming decades. xxxii E X ECU TI VE S U MMA RY The task of promoting human development, of ending pov- be avoided by holding warming below 2°C. This will require erty, increasing global prosperity and reducing global inequality substantial technological, economic, institutional and behavioral will be very challenging in a 2°C world, but in a 4°C world there change. And, most of all, it will require leadership at every level is serious doubt whether it can be achieved at all. Many of the of society. The time to act is now. worst projected climate impacts outlined in this report could still Box 7: Projected Impacts of Climate Change in Key Sectors in the Latin America and Caribbean Region Warming levels are relative to pre-industrial temperatures. The impacts shown here are a subset of those summarized in Table 3.15 of the Main report. The arrows indicate solely the range of warming levels assessed in the underlying studies, but do not imply any graduation of risk unless noted explicitly. In addition, observed impacts or impacts occurring at lower or higher levels of warming that are not covered by the key studies high- lighted here are not presented (e.g., coral bleaching already occurs earlier than 1.5°C warming but the studies presented here only start at 1.5°C). Adaptation measures are not assessed here although they can be crucial to alleviate impacts of climate change. The layout of the figure is adapted from Parry (2010). The lower-case superscript letters indicate the relevant references for each impact.10 If there is no letter, the results are based on additional analyses for this report. 1°C 1.5°C 2°C 3°C 4°C 5°C Land area affected by highly unusual heat Heat & 10% 30% 30-40% 65% 90% Drought Drought longer by(a) 1-4 days 2-8 days 8-17 days Glaciers Tropical glacier volume loss(b) 78-94% 66-97% 91-100% Southern Andes glacier volume loss(b) 21-52% 27-59% 44-72% 20- 40% 60- 80% Sea Probability of annual coral reef bleaching in Caribbean (high risk of extinction) (c) Sea level rise 0.27-0.39m, max 0.65m 0.46-0.66m, max 1.4m Fish catch potential(d) Up to +100% in South; up to -50% in Caribbean Water 10-30% decrease of mean runoff in Central America(e) Mean river discharge decreases in Northeast Brazil(f) Increasing biomass and carbon losses in Amazon(g) Forests & Increasing species range shifts/contractions and/or extinctions for mammals, marsupials, Biodiversity 10 birds, plants, amphibians (h) Rice and sugarcane yields possibly increase but high yield declines for wheat and maize (i) Food Beef cattle numbers in Paraguay(j) -16% -27% +5-13% Risk of diarrheal diseases(k) +14-36% Health +12-22% Increase in dengue (Mexico)(l) +40% Malaria increases in extra-tropics and highlands and decreases in the tropics (m) 10 a) Sillmann et al. (2013b); (b) Marzeion et al. (2012); Giesen and Oerlemans (2013); Radic et al. (2013); (c) Meissner et al. (2012); (d) Cheung et al. (2010); (e) Hidalgo et al. (2013); (f) Döll and Schmied (2012); (g) several studies without considering C02-fertilization, see Table 3.1; (h) several studies, see Table 3.1; (i) several studies, see Table 3.1; (j) ECLAC (2010); (k) Kolstad and Johansson (2011); (l) Colon-Gonzalez et al. (2013); (m) Beguin et al. (2011); Caminade et al. (2014); Van Lieshout et al. (2004). xxxiii TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Box 8: Projected Impacts of Climate Change in Key Sectors in the Middle East and North Africa Region Warming levels are relative to pre-industrial temperatures. The impacts shown here are a subset of those summarized in Table 4.10 of the Main Report. The arrows solely indicate the range of warming levels assessed in the underlying studies; but do not imply any graduation of risk unless noted explicitly. In addition, observed impacts or impacts occurring at lower or higher levels of warming that are not covered by the key studies highlighted here are not presented (e.g., increase in drought and aridity is already observed, but the respective study does not assess impacts below 1.5°C). Adaptation measures are not assessed here although they can be crucial to alleviating the impacts of climate change. The layout of the figure is adapted from Parry (2010). The lower-case superscript letters indicate the relevant references for each impact.11 If there is no letter, the results are based on additional analyses for this report. 1°C 1.5°C 2°C 3°C 4°C 5°C Heat Land area affected by highly unusual heat 5% 25% 30% 75% allmost all Drought & Moderate drought in Maghreb and Mashrek < 0.5 months per year ~ 1.5 months per year > 6 months per year Aridity Area classified as hyper-arid or arid 84% 87% Sea Level Sea level rise above present 0.20-0.64m 0.38-1.04m Food Loss of rainfed agriculture land(a) over 8,500 km over 170,000 km Reduction in crop yields(b) up to 30% up to 57% Decrease in snow melt water affecting the Water Euphrates and Tigris basin(c) by 55% by 77-85% by 87% 11 17% reduction in daily runoff for the tributaries of the Jordan River(d) Health Risk for diarrheal disease Increase in thermal discomfort by 35-70 days(e) increase by 6-15%(g) Risk for diarrheal disease increase by 16-41%(g) More people exposed to risk of malaria(f) 20-34 million 39-62 million Coast People affected from Egypt: 1.9 million Egypt: 3.6 million Morocco: 1.8 million Morocco: 2.1 million flooding(h) Loss of 25% of Nile Delta´s land area(i) 11 (a) Evans (2008); (b) several studies, see Table 4.1; (c) Bokurt and Sen (2013); (d) Samuels et al. (2010); (e) Giannakopoulos et al. (2013); (f) van Lieshout et al. (2004); (g) Kolstad and Johansson (2011); (h) Brown et al. (2011); (i) Dasgupta et al. (2009). xxxiv E X ECU TI VE S U MMA RY Box 9: Projected Impacts of Climate Change in Key Sectors in the Europe and Central Asia Region Warming levels are relative to pre–industrial temperatures. The impacts shown here are a subset of those summarized in Table 5.7 of the Main report. The arrows solely indicate the range of warming levels assessed in the underlying studies but do not imply any graduation of risk unless noted explic- itly. In addition, observed impacts or impacts occurring at lower or higher levels of warming that are not covered by the key studies highlighted here are not presented (e.g., an increase in Tien Shan glacier melt is already observed, but the respective study does not assess the observed impacts). Adaptation measures are not assessed here, although they can be crucial to alleviating the impacts of climate change. The layout of the figure is adapted from Parry (2010). The lower-case superscript letters indicate the relevant references for each impact.12 If there is no letter, the results are based on additional analyses conducted for this report. 1°C 1.5°C 2°C 3°C 4°C 5°C Heat & 5% 10% 15% 50% 85% Land area affected by highly unusual heat Drought 60% aridity increase in Western Balkans Aridity in Russian Federation: 10-40% decrease up to 60% decrease Glaciers 31% of Tien Shan 50% (31-66%)(b) 57% (37-71%)(c) 67% (50-78%)(d) glaciers melting(a) Central Asian glacier mass loss Significant runoff formation Water Strong river runoff 30-60 days peak shifts in decline in Central Asia(f) reduction in the Syr Dara basin(a) 45-75% increased water discharge in Balkans(e) North-Eastern Russia(s) Soil desertification and salinization (g) Food Yield declines due to 20% grape and droughts and floods in Increasing length of growing (i) 20-50% yield olive yield losses 30% yield drop Western Balkans(h) season in Albania(j) losses in in Tajikistan(k) Uzbekistan (j) Health 12 Balkans become suitable Increased vulnerability of Western Balkans to dengue and Heat mortality increased for dengue-transmitting (l) Tenfold increase in to 1000 per million(m) chikungunya mosquito(l) mudflow risk in Kazakhstan(m) 6-19% less capacity of nuclear and Energy 2.58% increase in hydropower fossil-fueled power plants in Europe(o) 35% decreased hydropower potential in Central Asia(n) potential in Croatia(p) Boreal Increase in timber Large decreases in timber harvest(q) harvest for larch and Dramatic changes in vegetation(r) Forests pine(q) 10 days increase in fire risk(r) 20-30 days increase in fire risk(r) 12 (a) Siegfried et al. (2012); (b) Marzeion et al. (2012); (c) Marzeion et al. (2012); Giesen and Oerlemans (2013); Radic et al. (2013); (d) Marzeion et al. (2012); Giesen and Oerlemans (2013); Radic et al. (2013); (e) Dimkic and Despotovic (2012); (f) Hagg et al. (2013); (g) Thurmann (2011); World Bank (2013f); World Bank (2013d); World Bank (2013a); (h) Maslac (2012); UNDP (2014); (i) Sutton et al. (2013a); Sommer et al. (2013); (j) Sutton et al. (2013a); (k) World Bank (2013m); (l) Caminade et al. (2012); (m) BMU and WHO-Europe (2009); (n) Hamududu and Killingtveit (2012); (o) van Vilet et al. (2012); (p) Pasicko et al. (2012); (q) Lutz et al. (2013b); (r) Tchebakova et al. (2009); (s) Schewe et al. (2013). xxxv Abbreviations °C degrees Celsius JJA June, July, and August (the summer season of the $ United States Dollars northern hemisphere; also known as the boreal AI Aridity Index summer) AOGCM Atmosphere-Ocean General Circulation Model LAC Latin America and the Caribbean AR4 Fourth Assessment Report of the LDC Least Developed Countries Intergovernmental Panel on Climate Change MAGICC Model for the Assessment of Greenhouse Gas AR5 Fifth Assessment Report of the Intergovernmental Induced Climate Change Panel on Climate Change MCMA The Mexico City Metropolitan Area BAU Business as Usual MENA Middle East and North Africa CaCO3 Calcium Carbonate MGIC Mountain Glaciers and Ice Caps CAT Climate Action Tracker NAO North Atlantic Oscillation CMIP5 Coupled Model Intercomparison Project Phase 5 NDVI Normalized Differenced Vegetation Index (used as CO2 Carbon Dioxide a proxy for terrestrial gross primary production) DGVM Dynamic Global Vegetation Model NH Northern Hemisphere DIVA Dynamic Interactive Vulnerability Assessment NPP Net Primary Production DJF December, January, and February (the winter OECD Organization for Economic Cooperation and season of the northern hemisphere) Development ECA Europe and Central Asia PDSI Palmer Drought Severity Index ECS Equilibrium Climate Sensitivity PgC Petagrams of Carbon (1 PgC = 1 billion tons of ENSO El-Niño/Southern Oscillation carbon) FAO Food and Agricultural Organization ppm Parts Per Million FPU Food Productivity Units PPP Purchasing Power Parity (a weighted currency GCM General Circulation Model based on the price of a basket of basic goods, GDP Gross Domestic Product typically given in US dollars) GFDRR Global Facility for Disaster Reduction and Recovery RCM Regional Climate Model GLOF Glacial Lake Outburst Flood RCP Representative Concentration Pathway HCS Humboldt Current System SCM Simple Climate Model IAM Integrated Assessment Model SLR Sea-level Rise IEA International Energy Agency SRES IPCC Special Report on Emissions Scenarios IPCC Intergovernmental Panel on Climate Change SREX IPCC Special Report on Managing the Risks of Extreme ISI-MIP Inter-Sectoral Impact Model Intercomparison Events and Disasters to Advance Climate Change Project Adaptation ITCZ Intertropical Convergence Zone xxxvii TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL TgC Teragrams of Carbon (1 TgC = 1 million tons of UNHCR United Nations High Commissioner for Refugees carbon) USAID United States Agency for International UNCCD United Nations Convention to Combat Development Desertification WBG World Bank Group UNDP United Nations Development Programme WGI Working Group I (also WGII, WGIII) UNEP United Nations Environment Programme WHO World Health Organization UNFCCC United Nations Framework Convention on Climate Change xxxviii Glossary Aridity Index: The Aridity Index (AI) is an indicator designed for CO2 fertilization: The CO2 fertilization effect refers to the effect identifying structurally arid regions; that is, regions with a long- of increased levels of atmospheric CO2 on plant growth. It may term average precipitation deficit. AI is defined as total annual increase the rate of photosynthesis mainly in C3 plants and increase precipitation divided by potential evapotranspiration, with the water use efficiency, thereby causing increases in agricultural pro- latter a measure of the amount of water a representative crop type ductivity in grain mass and/or number. This effect may to some would need as a function of local conditions such as temperature, extent offset the negative impacts of climate change on crop yields, incoming radiation, and wind speed, over a year to grow, which although grain protein content may decline. Long-term effects is a standardized measure of water demand. are uncertain as they heavily depend on a potential physiological long-term acclimation to elevated CO2 and other limiting factors, Biome: A biome is a large geographical area of distinct plant and including soil nutrients, water, and light. (See also Box 2.4 on the animal groups, one of a limited set of major habitats classified CO2 fertilization effect on crop productivity.) by climatic and predominant vegetative types. Biomes include, for example, grasslands, deserts, evergreen or deciduous forests, CMIP5: The Coupled Model Intercomparison Project Phase 5 and tundra. Many different ecosystems exist within each broadly (CMIP5) brought together 20 state-of-the-art GCM groups, which defined biome, all of which share the limited range of climatic generated a large set of comparable climate-projection data. The and environmental conditions within that biome. project provided a framework for coordinated climate change experi- ments and includes simulations for assessment in the IPCC AR5. C3/C4 plants: C3 and C4 refer to two types of photosynthetic biochemical pathways. C3 plants include more than 85 percent Development narratives: Development narratives highlight of plants (e.g., most trees, wheat, rice, yams, and potatoes) and the implications of climate change impacts on regional devel- respond well to moist conditions and to additional CO2 in the opment. The Turn Down the Heat series, and in particular this atmosphere. C4 plants (e.g., savanna grasses, maize, sorghum, report, discuss the potential climate change impacts on particu- millet, and sugarcane) are more efficient in water and energy use larly vulnerable groups along distinct storylines—the so called and outperform C3 plants in hot and dry conditions. development narratives. These development narratives were developed for each region in close cooperation with regional CAT: The Climate Action Tracker is an independent, science- World Bank specialists. They provide an integrated, often cross- based assessment that tracks the emissions commitments of and sectoral analysis of climate change impacts and development actions by individual countries. The estimates of future emissions implications at the sub-regional or regional level. Furthermore, deducted from this assessment serve to analyze warming scenarios the development narratives add to the report by allowing the that could result from current policy: (i) CAT Reference BAU: a science-based evidence of physical and biophysical impacts to lower reference business-as-usual scenario that includes existing be drawn out into robust development storylines to characterize climate policies but not pledged emissions reductions; and (ii) CAT the plausible scenarios of risks and opportunities—showcasing Current Pledges: a scenario additionally incorporating reductions how science and policy interface. currently pledged internationally by countries. xxxix TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL GCM: A General Circulation Model is the most advanced type The Fourth Assessment Report (AR4) was published in 2007. The of climate model used for projecting changes in climate due to Fifth Assessment Report (AR5) was published in 2013/2014. increasing greenhouse gas concentrations, aerosols, and external forcing (like changes in solar activity and volcanic eruptions). ISI-MIP: The first Inter-Sectoral Impact Model Intercomparison These models contain numerical representations of physical pro- Project (ISI-MIP) is a community-driven modeling effort which cesses in the atmosphere, ocean, cryosphere, and land surface provides cross-sectoral global impact assessments based on the on a global three-dimensional grid, with the current generation newly developed climate Representative Concentration Pathways of GCMs having a typical horizontal resolution of 100–300 km. and socioeconomic scenarios. More than 30 models across five sectors (agriculture, water resources, biomes, health, and infra- GDP: Gross Domestic Product is the sum of the gross value added structure) were incorporated in this modeling exercise. by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the product. Pre-industrial Level (what it means to have present 0.8°C warm- It is calculated without deductions for depreciation of fabricated ing): Pre-industrial level refers to the level of warming before/at assets or for depletion and degradation of natural resources. the onset of industrialization. The instrumental temperature records show that the 20-year average of global-mean, near-surface air GDP PPP: This is GDP on a purchasing power parity basis divided temperature in 1986–2005 was about 0.6°C higher than the aver- by population. Whereas PPP estimates for OECD countries are age over 1851–1879. There are, however, considerable year-to-year quite reliable, PPP estimates for developing countries are often variations and uncertainties in the data. In addition, the 20-year rough approximations. average warming over 1986–2005 is not necessarily representa- tive of present-day warming. Fitting a linear trend over the period Highly unusual and Unprecedented: In this report, highly 1901–2010 gives a warming of 0.8°C since “early industrialization.” unusual and unprecedented heat extremes are defined using Global mean, near-surface air temperatures in the instrumental thresholds based on the historical variability of the current local records of surface-air temperature have been assembled dating climate. The absolute level of the threshold depends on the natural back to about 1850. The number of measurement stations in the year-to-year variability in the base period (1951–1980), which is early years is small and increases rapidly with time. Industrializa- captured by the standard deviation (sigma). Highly unusual heat tion was well on its way by 1850 and 1900, which implies using extremes are defined as 3-sigma events. For a normal distribution, 1851–1879 as a base period, or 1901 as a start for linear trend 3-sigma events have a return time of 740 years. The 2012 U.S. analysis might lead to an underestimate of current and future heat wave and the 2010 Russian heat wave classify as 3-sigma warming. However, global greenhouse-gas emissions at the end of and thus highly unusual events. Unprecedented heat extremes the 19th century were still small and uncertainties in temperature are defined as 5-sigma events. They have a return time of several reconstructions before this time are considerably larger. million years. Monthly temperature data do not necessarily fol- low a normal distribution (for example, the distribution can have RCP: Representative Concentration Pathways are based on long tails, making warm events more likely) and the return times carefully selected scenarios for work on integrated assess- can be different from the ones expected in a normal distribution. ment modeling, climate modeling, and modeling and analysis Nevertheless, 3-sigma events are extremely unlikely and 5-sigma of impacts. Nearly a decade of new economic data, informa- events have almost certainly never occurred over the lifetime of tion about emerging technologies, and observations of such key ecosystems and human infrastructure. environmental factors as land use and land cover change are reflected in this work. Rather than starting with detailed socio- Hyper-aridity: This refers to land areas with very low Aridity Index economic storylines to generate emissions scenarios, the RCPs (AI) scores, generally coinciding with the great deserts. There is are consistent sets of projections of only the components of no universally standardized value for hyper-aridity, and values radiative forcing (the change in the balance between incoming between 0 and 0.05 are classified in this report as hyper-arid. and outgoing radiation to the atmosphere caused primarily by changes in atmospheric composition) that are meant to serve as IPCC AR4, AR5: The Intergovernmental Panel on Climate Change inputs for climate modeling. These radiative forcing trajectories (IPCC) is the leading body of global climate change assessments. are not associated with unique socioeconomic or emissions It comprises hundreds of leading scientists worldwide and on a scenarios; instead, they can result from different combina- regular basis publishes assessment reports which provide a com- tions of economic, technological, demographic, policy, and prehensive overview of the most recent scientific, technical, and institutional futures. RCP2.6, RCP4.5, RCP6 and RCP8.5 refer, socioeconomic information on climate change and its implications. xl GLOS S A RY respectively, to a radiative forcing of +2.6 W/m², +4.5 W/m², making different assumptions about the driving forces determining +6 W/m² and +8.5 W/m² in the year 2100 relative to pre- future greenhouse gas emissions. Scenarios were grouped into four industrial conditions. families (A1FI, A2, B1 and B2), corresponding to a wide range of high- and low-emissions scenarios. RCP2.6: RCP2.6 refers to a scenario which is representative of the literature on mitigation scenarios aiming to limit the increase of SREX: The IPCC published a special report on Managing the global mean temperature to 2°C above pre-industrial levels. This Risks of Extreme Events and Disasters to Advance Climate Change emissions path is used by many studies that have been assessed Adaptation (SREX) in 2012. The report provides an assessment of for the IPCC 5th Assessment Report and is the underlying low the physical and social factors shaping vulnerability to climate- emissions scenario for impacts assessed in other parts of this related disasters and gives an overview of the potential for effective report. In this report, the RCP2.6 is referred to as a 2°C world disaster risk management. (with the exception of sea-level rise, where the subset of model used actually leads to 1.5°C world—see Box 2.1, Definition of Tipping element: Following Lenton et al. (2008), the term tipping Warming Levels and Base Period in this Report). element describes large scale components of the Earth system pos- sibly passing a tipping point. A tipping point “commonly refers to RCP8.5: RCP8.5 refers to a scenario with a no-climate-policy base- a critical threshold at which a tiny perturbation can qualitatively line with comparatively high greenhouse gas emissions which is alter the state or development of a system” (Lenton et al. 2008). used by many studies that have been assessed for the IPCC Fifth The consequences of such shifts for societies and ecosystems are Assessment Report (AR5). This scenario is also the underlying likely to be severe. high-emissions scenario for impacts assessed in other parts of this report. In this report, the RCP8.5 is referred to as a 4°C world Virtual water: A measure of the water resources used in the pro- above the pre-industrial baseline period. duction of agricultural commodities. International trade in such commodities thereby implies a transfer of virtual water resources Severe and extreme: These terms indicate uncommon (negative) from one country to another embedded in the products. consequences. These terms are often associated with an additional qualifier like “highly unusual” or “unprecedented” that has a WGI, WGII, WG III: IPCC Working Group I assesses the physical specific quantified meaning. scientific aspects of the climate system and climate change. IPCC Working Group II assesses the vulnerability of socio-economic SRES: The Special Report on Emissions Scenarios, published by the and natural systems to climate change, negative and positive IPCC in 2000, has provided the climate projections for the Fourth consequences of climate change, and options for adapting to it. Assessment Report (AR4) of the Intergovernmental Panel on Climate IPCC Working Group III assesses the options for mitigating climate Change. The scenarios do not include mitigation assumptions. The change through limiting or preventing greenhouse gas emissions SRES study included consideration of 40 different scenarios, each and enhancing activities that remove them from the atmosphere. xli Chapter 1 Introduction This report describes the challenges for development and poverty reduction under anthropogenic climate change in Latin America and the Caribbean (LAC), the Middle East and North Africa (MENA), and Europe and Central Asia (ECA). Climate change projections are presented for each region alongside an assessment of the most recent literature on climate change impacts that are expected in key sectors for different warming levels, from the current baseline of 0.8°C through to 2°C and 4°C above pre-industrial levels in 2100. These impacts are then discussed in relation to existing social vulnerabilities to stress the implications of biophysical climate change impacts for development in the different regions. In order to provide a better understanding of how climate change and its expected impacts may affect development, storylines—referred to here as development narratives—are presented. These follow the chain of impacts from the physical to the biophysical and social level and specify how climate change could affect human well- being. Attention is paid to the specific ways in which different population groups are affected. The development narratives stress plausible scenarios and plausible impacts rather than giving probability ranges for the impacts estimated. Following the chain of impacts, the assessment shows that climate change poses severe challenges for human wellbeing and development, with the poor and underprivileged often hardest hit. The recent publications of the Fifth Assessment Report (AR5) by rise over the next 2,000 years. However sea levels would rise to the Intergovernmental Panel on Climate Change (IPCC) show that around 3.6 m under a 2°C warming scenario, and to around 8 m greenhouse gas emissions have continued to increase and in fact over the same period under a 4°C scenario (Levermann et al. have accelerated toward the end of the 2000s (IPCC 2014a). Global 2013). It is also important to note that greenhouse gas emissions mean temperature has already increased by 0.8°C from 1880–2012 and concentrations leading to a warming of 4°C by 2100 would (IPCC 2013a), and climate change impacts are increasingly being commit the world to much higher warming levels exceeding 6°C experienced on all continents and across a range of human and or more in the long term (IPCC 2013b). natural systems (IPCC 2014b). More severe impacts are projected with This report takes as its starting point the IPCC´s 5th Assessment further warming, and the resulting challenges for eradicating poverty Reports. In addition it provides regional and sub-regional narratives and promoting human wellbeing could be immense. If efforts and on the implications of climate impacts on development. The report achievements in reducing greenhouse gas emissions continue at the emphasizes topics that are of particular relevance for the focus current pace, warming levels of higher than 4°C cannot be ruled out. regions and considers scientific studies published after the literature Recent efforts to project the effects of current national poli- cutoff dates of the AR5. It thus takes up the guiding questions of the cies indicate that there is about a 40 percent chance of exceeding first two Turn Down the Heat reports (World Bank 2012a; 2013) by 4°C warming above pre-industrial levels by about 2100. Critically, focusing on three previously not assessed regions and digging deeper timing is of the essence. With rising temperatures, the risks for into the social and development consequences of climate change human lives and development trajectories increase, and a number for those affected. It does so along the following lines of inquiry: of impacts will soon be locked in for decades, if not for centuries 1. What are the key biophysical climate change impacts in the to come. For example, if present temperatures were to be main- case study regions under different levels of warming (par- tained, the world would be committed to around 2.3 m of sea-level ticularly 2°C and 4°C)? 1 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL 2. What are the crucial development impacts triggered by the bio- changes in precipitation combine with glacial melt to affect physical impacts of climate change within and across sectors? the seasonality of river discharge). 3. What are the implications of climate change impacts (physical, With a deliberate focus on the impacts of climate change, the biophysical, and social) for (differential) social vulnerability report frames the critical need for adaptation, but an assessment within the case study regions? of climate change adaptation options is outside the scope of this report. It is clear, however, from the evidence presented here that Through extensive data analysis and literature review, this there are a range of low-hanging and no-regret adaptation options report shows that the mitigation of greenhouse gas emissions that could increase the resilience of natural and social systems that cause climate change, adaptation to the consequences, and across sectors. These include, among others, closing the yield gap coping with the unavoidable impacts, must be part and parcel of between potential and actual agricultural yields through techno- the fight against poverty if risks are to be minimized and benefits logical changes, climate-smart agriculture, and/or climate-smart promoted. The biogeographic and development context of the urban development, as well as boosting health care systems in three regions defines the nature and extent of climate impacts developing regions. The IPCC WGII provides a comprehensive and the discussion of the development implications. For example: review of these and other options (cf. IPCC 2014b). • Latin America and the Caribbean is a highly heterogeneous Although adaptation is crucial to alleviate the impacts of climate region in terms of economic development and social and change, the potential for additional adaptation, beyond current adapta- indigenous histories. The rural poor depend on their natural tion, to reduce risks of climate change impacts to low levels is limited resource base, including subsistence agriculture and ecosystem even in a 2°C world (IPCC 2014c). In a 4°C world, the effectiveness services. In the Andean region, housing built on steep terrain of adaptation measures is thought to be even more limited for many is critically exposed to heavy precipitation events, landslides, systems and sectors, which will not be able to minimize risks by and glacial lake outbursts associated with glacial melt. Coastal simply doubling the effort to adapt to a 2°C world. For example, livelihoods, particularly in the Caribbean region, face the risks of an increasing severity and frequency in extreme events is likely to degrading marine ecosystems and coastal flooding, concurrent severely undermine a population’s adaptive capacity for consecutive with damage to critical infrastructure and freshwater supplies. impacts (World Bank and GFDRR 2013). A 4°C world would also likely mean irreversible changes in the Earth system, with impacts • The Middle East and Northern Africa relies heavily on agri- materializing well after warming has stopped. It is thus clear that culture as a source of food and income, not only in the his- investments that boost the adaptive capacity of people are a must torically important fertile crescent of the Euphrates and Tigris and it is also evident that climate change mitigation is required. The region but also at the Mediterranean coast and along the Nile. challenges that climate change poses are the focus of this report. At the same time, much of the region is covered by drylands and deserts. Seventy percent of the agricultural production is currently rain-fed, which leaves the region highly vulnerable 1.1 Development Narratives to the consequences of changes in precipitation patterns and temperature changes—with associated consequences for food Recent work has fostered understanding of what climate change security, social security, and rural livelihoods. This, in com- means for development (World Bank 2010a) and decision mak- bination with social changes and strong urbanization rates, ing in the face of uncertainty (Hallegatte et al. 2012). The Turn marks a very vulnerable future for the region, particularly for Down the Heat series, and this report in particular, reports the both the urban and rural poor. potential impacts on particularly vulnerable groups along distinct • Europe and Central Asia encompasses a wide terrain of storylines—the so-called development narratives. These develop- geographic features ranging from the mountainous and partly ment narratives provide an integrated, often cross-sectoral analysis coastal Western Balkans to the vast plains of Central Asia to of climate change impacts and development implications at the the boreal forests of the Russian Federation. In climatic terms sub-regional and regional scale. The development narratives the region displays a clear dipole, whereby regions in the also add to the report by allowing the science-based evidence southwest are becoming drier and regions in the northeast of physical and biophysical impacts to be effectively drawn out are becoming wetter as the world warms toward 4°C. The into robust development storylines to characterize the plausible most pronounced warming is expected to occur in two distinct scenarios of risks and opportunities—showcasing how science and regions: Northern Russia bordering the Barents-Kara Sea, and policy interact. It is widely accepted that climate change affects the Black Sea coastal region, including the Western Balkans. particular socioeconomic groups, such as the poor, the elderly, These changing conditions lead to a number of pronounced children, and women, the hardest (Leichenko and Silva 2014). This vulnerabilities, with a high risk of drought in the west and report includes a framework for assessing climate change impacts challenges for stable freshwater supplies in the east (where in light of existing social vulnerability (see Box 1.1). 2 I NTROD U CTION Box 1.1: Social Vulnerability Box 1.2: Climate Change Projections, Impacts, and Uncertainty Social vulnerability refers to the lack of capability of individuals, groups, or communities to cope with and adapt to external stresses In this report the projections of future climate change and its sectoral placed on their livelihoods and wellbeing. This is determined by the impacts are based, necessarily, on modeling exercises. The quantita- availability of resources and by the entitlement of individuals and tive results discussed take into account the inherent uncertainties groups to call on these resources (Füssel 2012). of model projections. The analysis of temperature and precipitation Social vulnerability, and how it differs according to socioeco- changes, as well as heat extremes and aridity, is based on a selection nomic and demographic conditions, is a common denominator in of state-of-the-art Coupled Model Intercomparison Project Phase 5 all three regions in this report. Examples of past extreme events (CMIP5) climate models. Following Hempel et al. (2013), precipitation expose the uneven distribution of impacts among different popula- data was bias-corrected, such that it reproduces the historical mean tions (IPCC 2014c), such as the impact of Hurricane Katrina in the and variation in precipitation. Results are reported as the mean of the United States or glacial lake outburst floods in the Peruvian Andes. models and their variability. Where relevant, a measure of agreement/ They show that factors such as socioeconomic class, race, gender, disagreement of models on the sign of changes is indicated. The and ethnicity affect the magnitude with which impacts are experi- projections might therefore provide more robust and consistent trends enced. Such examples also show that it is the underprivileged in than a random selection of model results, even at regional scales. rich nations who bear high burden of climate impacts. In addition, Results reported from the literature are, in most cases, based such extremes as hurricanes, floods, and heat waves leave in their on climate impact models and are likewise faced with issues about wake a trail of damage and human suffering which can extend well uncertainty. As with the case for climate projections, there are limita- beyond the point of impact in terms of both time and space. Dam- tions on the precision with which conclusions can be drawn. For ages to supply chains can transmit impacts across an ever more this reason, where possible conclusions are drawn from multiple globalized world and have long-lasting economic effects (Levermann lines of evidence across a range of methods, models, and data 2014). Already under present levels of warming, which have reached sources, including the Intergovernmental Panel on Climate Change about 0.8°C above pre-industrial levels, the number of local record- Fifth Assessment Report and the Special Report on Managing the breaking monthly temperature extremes is around five times higher Risks of Extreme Events and Disasters to Advance Climate Change than would be expected if no warming had occurred (Coumou et al. Adaptation (IPCC 2012). 2013). As the IPCC WGI AR5 (IPCC 2013b) has shown, and as it is outlined in this report, the likelihood of extreme events is projected to increase under rising temperatures. While large uncertainties exist development narratives present possible scenarios that are based on about the magnitude of the poverty effects of climate change and assumptions drawn from the scientific literature and shaped by the the scientific debate surrounding the issue is far from settled, the close cooperation with regional development specialists. As a result, evidence demonstrates a major reason for concern. it is not feasible to provide uncertainty ranges of the likelihood of the scenarios presented in the development narratives. Rather, a risk-based approach indicating plausible high-impact consequences 1.2 Methodological Approach under 2°C and 4°C global mean warming was chosen to form a basis of further policy and research. The projections on changes in temperature, heat extremes, pre- cipitation, aridity, and sea-level rise are based on original analysis 1.3 Structure of the Report of output from state-of-the-art General Circulation Models (GCM) (see Box 1.2 and Appendix). The development narratives combine The report is structured as follows. Chapter 2 explores the probability knowledge that has been gained through large and small scale of warming reaching 4°C above pre-industrial levels and discusses quantitative and qualitative research. The sectoral analysis for the the feasibility of significantly limiting global mean warming to three regions is based on existing literature. The literature review below 2°C. It further provides an update on global climate impact was almost exclusively conducted in the English language and it projections for different levels of global warming. The updated followed a prescribed hierarchy of sources: peer-reviewed scientific analysis of the risks at the global level further complements the publications were given most weight, followed by peer-reviewed first two Turn Down the Heat reports (World Bank 2012a; 2013) reports. Where these sources were lacking and additional infor- and provides a framework for the regional case studies. Chapter 2 mation was obtainable from other sources, those were used (but also presents a framework for how social vulnerability and climate least weight was given to them). As the studies assessed were not change interact. Chapters 3, 4 and 5 present analyses of climate conducted within an integrated framework, their integration does impacts and the development narratives for the Latin America and not allow for a coherent analysis of overall impacts and human the Caribbean, the Middle East and North Africa, and Europe and consequences. Rather, the causal pathways mapped out in the Central Asia regions respectively. 3 Chapter 2 The Global Picture The Fifth Assessment Report (AR5) of the IPCC provides a very comprehensive analysis of the physical science basis and the observed and projected impacts of climate change as well as of the economics of climate change mitigation. The following chapter should be read as an addition to the AR5, with emphasis on topics that are of particular relevance for the focus regions of this report and where scientific studies published after the literature cutoff dates of the AR5 led to an update of the findings from IPCC for specific issues. The IPCC projections always draw on the largest model ensemble Copenhagen pledges in this assessment.12 As a consequence, available (which differs significantly in size ranging from less than it is projected on the basis of these assessments that under 10 to more than 30 climate models). This report restricts most of recent trends and current policies there is about a 40 percent the projections presented in this chapter and throughout to five chance of exceeding 4°C by 2100 and a 10 percent risk of state-of-the-art CMIP5 models that are bias-corrected and used in exceeding 5°C. the ISI-MIP framework (see Appendix). Although the robustness • Based on a sample of 114 energy-economic model scenarios of projections based on smaller ensembles is generally lower, this estimating emissions in the absence of further substantial policy approach allows for direct comparison between different impacts action (baseline scenarios), climate-model projections reach in an impact cascade (e.g., of changes in precipitation patterns, a warming of 4.0–5.2°C above pre-industrial levels by 2100 river discharge, and agricultural impacts). (95 percent range of scenarios—grey shaded area Figure 2.1) (Blanford et al. 2014; Kriegler et al. 2013; Kriegler, Tavoni et al. 2.1 How Likely is a 4°C World? 2014; Kriegler, Weyant et al. 2014; Luderer et al. 2013; Riahi et al. 2013; Tavoni et al. 2014). Only under extreme assump- The previous Turn Down the Heat reports estimated that current tions regarding autonomous improvements leading to large emissions reductions pledges by countries worldwide, if fully decreases in energy intensity and an energy demand by 2100 implemented, would lead to warming exceeding 3°C before 2100. (e.g., 35–40 percent lower than under default assumptions) New assessments of business-as-usual emissions in the absence (Blanford et al. 2014; Kriegler, Weyant et al. 2014) does the of strong climate mitigation policies, as well as recent reevalua- lower end of the baselines’ 95 percent range decrease to below tions of the likely emissions consequences of pledges and targets 4°C (3.5°C) in 2100. adopted by countries, point to a considerable likelihood of warm- ing reaching 4°C above pre-industrial levels within this century. • The IPCC Working Group III Fifth Assessment Report assess- ment of baseline scenarios leads to a warming of 3.7–4.8°C • Assessments of recent trends and “current policies” in the by 2100 (80 percent range of scenarios; 2.5–7.8°C including world’s energy system analyzed by the International Energy climate-system uncertainty), which includes the scenarios with Agency in its World Energy Outlook 2012 indicate global-mean assumptions on autonomous improvements in energy intensity. warming above pre-industrial levels would approach 3.8°C by 2100. Another assessment, by Climate Action Tracker, of these trends and policies leads to a warming of 3.7°C, about 12 http://climateactiontracker.org/news/151/In-talks-for-a-new-climate-treaty-a-race- 0.6°C higher than the median estimate of the effect of the to-the-bottom.html. 5 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 2.1: Projections for surface-air temperature increase, showing the scenarios assessed in this report in the context of baseline projections (no further substantial climate policy from the recent energy-economic model literature). The medium-dark-grey shaded area (bar on right-hand side indicates year-2100 range) indicates the 90 percent uncertainty range over 114 baseline sce- narios from the literature, excluding a class of variants that assume very high autonomous improvements in energy intensity. The dark-grey range on the right-hand side indicates the broadening of the 90 percent scenario range, if one includes these variants. For comparison, the light-grey area indicates the 80 percent range over all scenarios, as assessed in IPCC WGIII AR5 (2014d). The climate model underlying these calculations is the same as applied for all emissions scenarios in the IPCC WGIII AR5 (2014d). The climate-system uncertainty derived from such modeling is depicted for RCP2.6 as the blue- shaded area (66 percent uncertainty). Note that the estimates here are comparable to the full CMIP5 model range, which are slightly cooler at the end of the 21st century than the ISI-MIP subset of CMIP5 models used for most projections in this report (see Appendix). In this subset, the RCP 2.6 scenario represents a 1.8°C warming and the RCP 8.5 represents a 4.6°C warming above pre-industrial levels for the period 2081–2099. • In relation to the effects of pledges, the updated UNEP Emis- (IPCC 2013b). These AR5 model runs were designed to take into sions Gap Assessment 2013 (UNEP 2013) assessed present account a fixed pathway of concentrations and do not, therefore, emissions trends and pledges. Global emissions estimated for include the effects of carbon-cycle feedbacks on the response of 2020 are consistent with emissions pathways that reach warm- the climate system to CO2 emissions. Including these feedbacks ing in the range of 3–5°C by 2100,13,14 and are closest to levels raises the median estimate of warming in RCP8.5 from 4.3 to 4.5°C consistent with pathways leading to 3.5–4°C warming by 2100. by the 2080s and widens the total uncertainty, in particular at the high-temperature end, from 3.2–5.3°C to 3.1–6.2°C by that time, On average, the RCP8.5 is illustrative of a range of business-as- relative to pre-industrial levels (Collins et al. 2013). usual scenarios,15 reaching a global-mean warming level of about 4°C above pre-industrial levels by the 2080s, and gives a median warming of about 5°C by 2100 (Figure 2.1). The IPCC AR5 WGI 2.1.1 Can Warming be Held Below 2°C? noted that 62 percent of high-complexity climate models exceed Climate policy has not to date succeeded in curbing global green- 4°C for RCP8.5 by the 2080s,16 and all model runs exceed 3°C house gas emissions, and emissions are steadily rising (Peters et al. 2013). However, recent high-emissions trends do not imply a 13 lock-in to a high-emitting pathway (van Vuuren and Riahi 2008) The Climate Action Tracker projections of the effects of pledges if fully implemented is about 3.1°C warming by 2100 (median estimate)—i.e., at the lower end of this range. if there is a move toward rapid, technically and economically 14 This applies to the “unconditional pledges, strict rules” case. feasible mitigation. As was confirmed in the 2013 UNEP Emissions 15 Not including those with extreme assumptions on autonomous improvements in Gap Report (UNEP 2013) and in successive International Energy energy intensity. 16 A probability of >66 percent is labeled “likely” in IPCC’s uncertainty guidelines Agency Assessments (International Energy Agency 2013), there are adopted here. many measures that could close the gap between estimated global 6 TH E GLOBA L P ICTU RE Box 2.1: Definition of Warming Levels and Base Period in This Report This report and the previous Turn Down the Heat reports reference future global warming levels against the pre-industrial period 1850–1900 consis- tent with the IPCC WGI AR5. To study the impacts of climate change at different levels of global mean warming in this report, the literature and present climate impacts for different warming levels above the pre-industrial period were reviewed using the following classification: WARMING CATEGORY OBSERVED 1°C 1.5°C 2°C 3°C 4°C Range [°C] <0.8 0.8–1.25 1.25–1.75 1.75–2.25 2.25–3.5 >3.5 Given the diversity of different base periods for projections, emissions scenarios, and models or model ensembles used, this report adopts a standardized approach to convert any given warming level with any given base period to its corresponding warming level relative to the pre-industrial period (see Appendix). This allows, within limits, for a classification of climate impacts independent of the underlying emissions scenario and model or model ensemble used. This stringent approach of classifying warming levels is new in this Turn Down the Heat report and allows for a consistent comparison of climate change impacts across studies and sectors. Median warming for the full CMIP5 model ensemble under the RCP2.6 is about 1.6°C (and therefore on the border between the 1.5° and 2°C warming categories presented above); 22 percent of the models nonetheless projecting a warming above 2°C. For the estimation of heat extremes, precipitation, and aridity, this report uses a subset of the CMIP5 models (as in the ISI-MIP project) showing a median warming of 1.8°C above pre- industrial levels by 2081–2100 for the RCP 2.6 scenario. As in the earlier reports, impacts in a “2°C world” refer to the impacts assigned to the 2°C warming category. Where the results of the ISI-MIP ensemble for RCP 2.6 are used to describe a 2°C world, readers need to be aware that this is at the low end of the 2°C impact category. Sea-level rise projections presented in this report are based on a larger model ensemble with an ensemble mean warming of less than 1.75°C; as a result, end-of-century sea-level rise in RCP2.6 is classified as 1.5° warming. The “4°C world” refers to impacts assigned to the 4°C category as described above. The median warming of the RCP8.5 CMIP5 ensemble for the period 2081–2100 is 4.3°C, whereas the projected warming for the ISI-MIP ensemble (used to project heat extremes, precipitation, and aridity) is 4.6°C above pre-industrial levels (and thus at the high end of the 4°C warming category). Impacts shown in tables refer to the exact warming category above pre-industrial levels, as they allow for more detail than the stylistic futures drawn at 2°C and 4°C global mean warming. The terms “2°C world” and “4°C world” always refer to end-of-21st-century impacts’ for impacts referring to earlier time periods the convention of “2°C warming by 20xx” is chosen. greenhouse gas emissions levels by 2020 and levels consistent with 2070 [2060 to 2080]. The lowest published emissions scenarios in pathways that keep warming below 2°C. The required emission recent literature (Luderer et al. 2013; Rogelj et al. 2013a; Rogelj et reductions over the 21st century, were estimated by IPCC WGIII al. 2013b) lead to warming projected to peak at around 1.5°C and AR5 to lead to an annualized reduction in consumption growth decline to a median level of 1.3°C above pre-industrial by 2100. limited to 0.04–0.14 percentage points, relative to baseline growth of 1.6–3 percent per year (IPCC 2014d). This does not include the co-benefits, including for example health and environmental 2.2 Climate Sensitivity and Projected benefits from reduced co-emited air pollutants, fuel poverty reduc- Warming tions and net employment gains (IPCC 2014d). Delaying additional mitigation increases mitigation costs in the medium- to long-term. Although the past decade has been the warmest on record glob- The recent IPCC AR5 WGI (summary for policymakers) showed ally, observations suggest that the rate of warming during the that “warming is unlikely to exceed 2°C for RCP2.6” and “likely last decade has been slower than earlier decades. This has led to to exceed 1.5°C . . . for all RCP scenarios except RCP2.6” (IPCC discussions of a so-called “warming hiatus”, for example in the 2013a). The energy-economic modeling behind the RCP2.6 emis- IPCC AR5 WGI: sions scenario (van Vuuren et al. 2011) shows that large-scale “In summary, the observed recent warming hiatus, defined transformations in the energy system are feasible to achieve the low as the reduction in GMST trend during 1998–2012 as emissions levels of RCP2.6 and net-negative global energy-related compared to the trend during 1951–2012, is attributable CO2 emissions by the 2070s. The IPCC WGIII AR5 showed that there in roughly equal measure to a cooling contribution from is a broad category of low-emissions mitigation scenarios that reach internal variability and a reduced trend in external forcing emissions levels comparable to or lower than RCP2.6 (IPCC 2014d). (expert judgment, medium confidence).” (Stocker et al. On average, these scenarios also reach net-zero CO2 emissions by 2013, Box TS.3) 7 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Slower and faster decades of warming occur regularly, related Figure 2.2: Climate-model projections of global-mean surface- to variations in forcing (e.g., volcanic eruptions, solar activity) air temperature for RCP2.6 (lower curves) and RCP8.5 (top and to internal redistribution of heat in the oceans driven by curves). large-scale patterns of climate variability (including the El Niño/La Niña-Southern-Oscillation—see also Section 2.3.2) causing natural variations of surface warming (Balmaseda et al. 2013; England et al. 2014; Foster and Rahmstorf 2011).17 An additional factor is data uncertainty. Without taking data uncertainties or the physical explanation of a remaining “hiatus” into account, the recent slower warming has led to media attention that suggests the sensitivity of the climate system to anthropogenic emissions might be smaller than estimated previously. IPCC AR5 WGI estimated equilibrium climate sensitivity (ECS) at 1.5–4.5°C, which has a lower low-end estimate than the IPCC AR4 at 2–4.5°C. On the contrary, values substantially higher than 4.5°C still cannot be ruled out. Climate projections for the 21st century are, however, not highly sensitive to equilibrium warming estimates. Rogelj et al. (2014) evaluated the implications of both AR4 and AR5 assessments of ECS in a climate-model framework. While the uncertainty ranges of global-mean temperature increases by 2100 for RCP2.6 and RCP8.5 are slightly wider for the AR5 than for the AR4 model version, the difference is small and the median estimates are virtually unaffected (Figure 2.2). Hence, a change in estimated Lines indicate median estimates, grey shaded areas 66 percent uncertainty equilibrium climate sensitivity, among others informed by a recent ranges, consistent with the AR4 assessment of equilibrium climate sensi- warming hiatus, has no significant effect on the climate projections tivity (2–4.5°C). On the right-hand side, the increases averaged over the presented in this report. years 2081–2100 are shown for both the model version constrained by a representation of the AR4 assessment and the AR5 (blue) assessment (1.5–4.5°C). Bold right-hand side columns show a 66 percent uncertainty 2.3 Patterns of Climate Change range, thin columns a 90 percent range, diamonds median estimate, circles mean estimates (higher than the median, because of skewed probability distributions due to significant likelihood of very high climate sensitivity This report gives an update of the projected patterns of climate values). Source: Rogelj et al. (2014). change presented in the earlier Turn Down the Heat reports with a particular focus on temperature and precipitation extremes as well as changes in droughts and river runoff. 2.3.1 Observed Trends in Extreme Events Since the 1960s, a robust increase in the number and magnitude of hot temperature extremes is observed globally that is consistent with the increase in global mean temperature over the same time period (Donat, Alexander et al. 2013; Seneviratne et al. 2012). 17 Consequently, the recent IPCC AR5 assesses a human contribution This can be explained by natural external forcings, like those of solar and volcanic origin, and physical mechanisms within the climate system itself. This includes a to this trend as very likely (IPCC 2013a). Coumou et al. (2013) large role played by the El Niño/La Niña-Southern-Oscillation, a pattern of natural find that new record-breaking monthly mean temperatures can be fluctuations in heat transfer between the ocean’s surface and deeper layers. If such attributed to climate change with an 80 percent probability. Despite fluctuations are filtered out of the observations, a robust continued warming signal emerges over the past three decades. It is this signal that should be compared to the a decade of slowed-down global mean temperature increase, the average warming of climate models, because the latter exhibit the same upswings number of observed hot temperature extremes is continuously on and downswings of warming as the observational signal, but at different times, due the rise—with a trend that is strongest for the most extreme events to the natural chaotic nature of the climate system. Taking an average from many (Seneviratne et al. 2014; Sillmann et al. 2014). At the same time, models filters out these random variations; hence, this must also be done with observational datasets before comparing with model results. an increase in frequency and duration of heat waves has been 8 TH E GLOBA L P ICTU RE Figure 2.3: ENSO and extreme events. al. 2013). Figure 2.3 illustrates the increase in precipitation-related disasters recorded in the EM-DAT database since the 1960s. While the observed changes in heavy precipitation are sta- tistically robust and the level of agreement between different studies and datasets is high for most world regions (Donat et al. 2014), this is not the case for dry spells and droughts (Dai 2012; Donat, Alexander et al. 2013; Sheffield et al. 2012; Trenberth et al. 2014). Although global trends remain uncertain, robust dry- ing signals emerge from the observational record (e.g., for the Mediterranean) (Donat, Peterson et al. 2013; Hoerling et al. 2012; Sousa et al. 2011). Additionally, a strengthening in the seasonal cycle and regional contrast has been observed, meaning that the strongest increase in heavy precipitation events has been found during wet seasons of already wet regions, while the strongest drying signals emerge during the dry season of already dry areas (Chou et al. 2013); this further amplifies flood and drought risks in the respective regions. Despite a substantial increase in meteorological disasters Upper panel: Number of climate-related disasters from 1960–2013 (based recorded in the EM-DAT database (compare Figure 2.3) that are on the EM-DAT database18). A robust increase in all types of climate-related disasters is observed. Lower panel: El Niño and La Niña events identified related to tropical and extra-tropical storms, confidence in large- on the basis of the Niño 3.4 sea-surface temperature index.19 scale trends in meteorological indices remains low for extra-tropical storms and tropical cyclones as well as for such small-scale meteo- rological events as hail and thunderstorms (Stocker et al. 2013, TS.2.7.1). For tropical cyclones, however, a robust global trend in poleward migration is observed (Kossin et al. 2014). At the same observed globally, although trends differ strongly among regions time, the frequency and intensity of the strongest tropical cyclones (Donat, Alexander et al. 2013; IPCC 2013a, Table SPM.1).1819 have increased significantly in the North Atlantic since the 1970s Unlike the patterns of change for extreme temperature indices, (Grinsted et al. 2012; IPCC 2013a, Table SPM.1; Kossin et al. 2013) changes in extreme precipitation appear to be more heterogeneous. with profound consequences for the Caribbean, Central America, Significant increases both in frequency and intensity of heavy pre- and southeastern-North America. cipitation events are observed over eastern North America, large A steep rise in climate-related disasters by one order of mag- parts of Eastern Europe, Asia, and South America; a decrease is nitude from about 30 in the early 1960s to more than 300 in the reported for the Mediterranean, South East Asia, and the northwest- early 21st century (trend: approximately 70 events per decade from ern part of North America (Donat, Alexander et al. 2013). While 1960–2014) is apparent from the EM-DAT database (see Figure 2.3). being spatially heterogeneous, most areas of the globe experience The absolute values of this increase should be interpreted with an increase in precipitation extremes—and a human contribution caution, since this signal is distorted by an increase in climate- to this increase can be clearly identified (IPCC 2013a, Table SPM.1; related disaster reporting over the same time frame that is very Min et al. 2011). The median intensity of extreme precipitation is difficult to quantify. Still, this increase in climate-related disaster found to increase by about 6–8 percent per degree rise in global reporting is assumed to have happened predominantly before the mean temperature (Kharin et al. 2013; Trenberth 2011; Westra et mid-1990s (and the advent of modern information technology); while the number of disasters counted nearly doubled between 18 EM-DAT: The OFDA/CRED International Disaster Database. Available at www the mid-1990s and 2010–2014. .emdat.be. 19 Although not necessarily as strong as for climate-related disas- Following NOAA guidelines, the index is derived based on running-mean 3-month SST anomalies in the Niño 3.4 region (5°N–5°S, 120°–170°W)]. At least five con- ters, such an increase in reporting is also assumed to be responsible secutive overlapping 3-month periods above 0.5°C (below –0.5°C) are identified for the observed increase in geophysical disasters (volcanic eruptions as El Niño (La Niña) events. El Niño events are classified as weak (moderate) if and earthquakes). This trend, however, is an order of magnitude at least three consecutive overlapping 3-month periods exceed 0.5°C (1°C) and as strong if they exceed 1.5°C. La Niña events are similarly classified. Source: NOAA, smaller than what has been observed for climate-related extremes Oceanic Niño Index (ONI). (about 6.5 events per decade over 1960–2014). A robust trend also 9 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL emerges from the analysis for severe climatological disasters, for even though such an effect may already be occurring but obscured which a reporting bias can be assumed to be absent. Thus, the by natural inter-decadal variability (Christensen et al. 2013; Latif increase in climate-related disasters in the EM-DAT database can and Keenlyside 2009; Stevenson et al. 2012; Wittenberg 2009). be attributed in part to climate change, but the exact influence A tree-ring-based reconstruction of ENSO strength suggests that of increased reporting cannot be quantified. The lower panel of ENSO over the past 700 years has never been as variable as dur- Figure 2.3 depicts the time series of the El Niño Southern Oscilla- ing the last few decades, and that the ENSO cycle thus may have tion index highlighting low, moderate, and strong El Niño and La intensified due to global warming (Jinbao et al. 2013). Niña events that have a profound imprint on the tropical climate regime and extreme events statistics globally. 2.3.2.2 ENSO Projections Despite model uncertainties, there is high confidence that the mean 2.3.2 El-Niño/Southern Oscillation climate in the tropical Pacific will change under global warming. It is likely that these changes will affect ENSO through one or One of the largest sources of climate variability in terms of scale several of the associated atmospheric or oceanic feedbacks; but and impact is the El Niño/Southern Oscillation (ENSO). ENSO is models disagree on the exact nature of the ENSO changes (Col- a coupled atmosphere-ocean phenomenon in the tropical Pacific lins et al. 2010; Guilyardi et al. 2012; Latif and Keenlyside 2009; region and the dominant global mode of variability on an inter- Power et al. 2013). annual timescale. Although substantial uncertainties remain Some robust evidence of changes to ENSO-driven precipitation with respect to how ENSO will respond to rising atmospheric variability in response to global warming has emerged recently, and temperature, recent model intercomparison studies suggest a trend it represents an update to the assessment of ENSO projections in toward more extreme El Niño events over the 21st century (Cai et the IPCC AR5. A majority of GCMs shows that the rainfall anoma- al. 2014; Power et al. 2013). lies associated with El Niño events in the tropical Pacific (i.e., dry During El Niño events, the heat that is stored in the ocean conditions in the west and more abundant rainfall in the central is released into the atmosphere, leading to changes in the tropi- and eastern Pacific) will become stronger (Power et al. 2013) and cal atmospheric circulation (see Box 2.2) and, consequently, to impacts of El Niño thus more intense. This result is consistent variations in weather patterns around the world. Anomalous El with another study showing a major increase in the frequency of Niño-type conditions are related to disastrous flooding events extreme El Niño events (Cai et al. 2014). These particularly strong in Latin America and droughts in Australia and large areas events have caused devastating impacts (recent examples include of South East Asia, and can have far-reaching effects on the the 1982–83 and the 1997–98 El Niño events) and GCMs that are Atlantic hurricane activity and the global monsoon system (e.g. able to simulate such events project approximately a doubling in Donnelly and Woodruff (2007); Kumar et al. (2006)). All these frequency in a 4°C world (Cai et al. 2014). changes can substantially impair livelihoods (e.g., through impacts on agricultural productivity, infrastructure, and public 2.3.3 Projected Changes in Extreme health) (Kovats et al. 2003; Wilhite et al. 1987). For example, Temperatures the extreme El Niño of 1997–98 resulted in billions of dollars of economic damages and tens of thousands of fatalities worldwide, Figure 2.5 depicts projected regional boreal summer (JJA) warm- with severe losses in Latin America in particular (McPhaden et al. ing by 2071–2099 for a 4°C world (RCP8.5) and a 2°C world 2006; Vos et al. 1999). Recent research has also suggested an (RCP2.6). The upper panels show the northern hemisphere sum- ENSO influence on the risk of civil conflict around the world mer temperature anomalies relative to the 1951–1980 base period (Hsiang et al. 2011). for the 4°C world (RCP8.5) and the 2°C world (RCP2.6). In a 2°C world, summer warming anomalies exceed 2°C for large areas 2.3.2.1 Observed Changes in ENSO of the Northern Hemisphere; this regional warming is enhanced It seems unlikely that a complex system consisting of numerous disproportionally for some regions in a 4°C world, where up to feedback mechanisms would not be affected by anthropogenic 8°C boreal summer warming is reached in northern central Russia global warming (Collins et al. 2010). The amplitude, frequency, and in the western United States. In addition, the Mediterranean seasonal timing, and spatial pattern of ENSO events, as well as appears as a warming hotspot; so do Northern Africa and Central their links to weather patterns around the world, might all be Asia, which is likely due to regional amplification as a result of a altered in what could potentially be one of the most prominent drying trend in these regions. manifestations of climate change (Guilyardi et al. 2012; Vecchi The lower panels of Figure 2.5 illustrate the projected warming and Wittenberg 2010). But the instrumental record is short and so in terms of regional climate variability relative to the 1951–1980 far yields no clear indication of a climate change effect on ENSO, base period (Hansen et al. 2012). The local absolute warming is 10 TH E GLOBA L P ICTU RE Box 2.2: Mechanisms Behind the El-Niño/Southern Oscillation In simple terms, the oceanic part of ENSO is characterized by the movement of a pool of warm surface water in an east-west direction along the equator. In the mean state, this warm pool is confined to the western tropical Pacific, whereas in the Eastern Pacific, sea surface temperatures (SST) are much cooler and the surface water layer, separated from the cold, deep ocean by the so-called thermocline, is shallower (see Figure 2.4 a). In the west, the warm surface water facilitates strong atmospheric convection and resulting precipitation; together with sinking motion over the colder East Pacific and easterly surface winds that follow the SST gradient an overturning movement of air is formed—the so-called Walker circulation. This wind pattern, in turn, drives upwelling of cold deep water in the East Pacific and reinforces the concentration of warm surface water in the west. This positive feedback between oceanic and atmospheric processes renders the mean state unstable. Small fluctuations in a part of the system are amplified into oscillations of the entire system. Every couple of years, when the easterly surface winds weaken and the warm water from the West Pacific extends so far to the east that it approaches the American coast, it creates an El Niño event, with a characteristic pattern of positive SST anomalies in the central and eastern tropical Pacific (see Figure 2.4 b). In some years the pendulum swings the other way, and easterly winds strengthen and the warm water pool gets concentrated further in west than on average. This is called a La Niña event. Both El Niño and La Niña events occur every 2–7 years and with varying magnitudes. Figure 2.4: Idealized schematic showing atmospheric and oceanic conditions of the tropical Pacific region and their interactions during normal conditions, El Niño conditions, and in a warmer world. (a), Mean climate conditions in the tropical Pacific, indicating sea-surface temperatures (SSTs), surface winds and the associated Walker circulation; the mean position of atmospheric convection over the Western tropical Pacific; the mean oceanic upwelling in the Eastern tropical Pacific; and the position of the oceanic thermocline separating surface and cooler nutrient rich deep ocean waters. (b), Typical conditions during an El Niño event. SSTs are anomalously warm in the east; convection moves into the central Pacific; the trade winds weaken in the east and the Walker circulation is disrupted; the thermocline flattens; and the upwelling is reduced. Source: Collins et al. (2010). divided (normalized) by the standard deviation (sigma) of the A 3-sigma deviation (see Box 2.3) would be considered a very local monthly temperature dataset over the reference period, rare extreme month under present conditions and a deviation by which represents the normal year-to-year changes in monthly five sigma (or more) unprecedented. Since natural variability is temperature because of natural variability (see Box 2.3). Since lower in the tropics, the same change in absolute temperature is ecosystems and humans are adapted to local climatic conditions much stronger relative to natural variability in this region, posing and infrastructure is designed with local climatic conditions and a potential threat to ecosystems even under low levels or warm- its historic variations in mind, this approach helps to highlight ing. Under a 4°C warming scenario (RCP8.5), about 50 percent of regions most vulnerable to warming. the global land surface is projected to be covered on average by 11 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 2.5: Multi-model mean global temperature anomaly for RCP2.6 for (2°C world, left) and RCP8.5 (4°C world, right) for the boreal summer months (JJA). Temperature anomalies in degree Celsius (top row) are averaged over the time period 2071–1999 relative to 1951–1980. The bottom row offers a different perspective on these changes by depicting them relative to a measure of currently normal variations in monthly temperature (standard deviation)–a factor of 2 indicates that projected changes in boreal summer temperature are twice as large as currently occurring variations in summer temperatures across different years. Box 2.3: Heat Extremes This report defines two types of monthly temperature extremes using thresholds based on the historical variability of the current local climate (similar to Hansen et al. 2012). The absolute level of the threshold thus depends on the natural year-to-year variability in the base period (1951–1980), which is captured by the standard deviation (sigma). While there is a range of impact relevant temperature extreme measures on a daily or multi-daily basis, this report focuses on monthly data. 3-sigma Events—Three Standard Deviations Outside the Normal • Highly Unusual at present • Extreme monthly heat • Projected to become the norm over most continental areas by the end of the 21st century 5-sigma Events—Five Standard Deviations Outside the Normal • Essentially absent at present • Unprecedented monthly heat: new class of monthly heat extremes • Projected to become common, especially in the tropics and in the Northern Hemisphere mid-latitudes during summertime Assuming a normal distribution, 3-sigma events would have a return time of 740 years. The 2012 U.S. heat wave and the 2010 Russian heat wave classify as 3-sigma events (Coumou and Robinson 2013). 5-sigma events have a return time of several million years. Monthly temperature data do not necessarily follow a normal distribution (for example, the distribution can have long tails, making warm events more likely) and the return times can be different from the ones expected in a normal distribution. Nevertheless, 3-sigma events are extremely unlikely and 5-sigma events have almost certainly never occurred over the lifetime of key ecosystems and human infrastructure. 12 TH E GLOBA L P ICTU RE Figure 2.6: Estimates of world population experiencing highly unusual monthly boreal summer temperatures (JJA, averaged over centered 20-year time intervals) for the RCP2.6 (2°C world, left panel) and the RCP8.5 (4°C world, right panel). Population estimates are based on the Shared Socioeconomic Pathway 2 (SSP2) and shown in terms of population density.20 highly unusual heat extremes by 2050; this number increases to constrained to the tropical areas of South America, Africa, and 90 percent by the end of the century. The departure from the his- South East Asia—with unprecedented heat extremes being rare. torical climate regime (which means that the coldest month will be Using population estimates from the intermediate Shared Socio- warmer than the warmest month during the 1860–2005 reference economic Pathway 2 (SSP2), the population affected by extreme period) for large parts of the tropics is projected to happen by the heat can be approximated over the 21st century. It is important to 2050s–2060s under such a scenario (Mora et al. 2013). Also in a highlight that these population estimates are based on a range of 4°C world, about 60 percent of the global land surface is projected assumptions about societal development that cannot be constrained. to be covered on average by unprecedented heat extremes by the Unlike physically based projections (e.g., temperature projections), end of the century. This implies a completely new climatic regime these population estimates should be interpreted merely as a pos- posing immense pressure globally on natural and human systems. 20 sible future, not claiming any kind of predictive skill. In a 2°C world (RCP2.6), the projected land area experiencing The policy choices made with respect to a pathway toward highly unusual heat extremes is limited to about 20 percent and a 4°C or 2°C world will make a fundamental difference in terms of the population affected by heat extremes in the 2020s and thereafter. In 2025, about 17 percent of the world's population 20 The basis for the gridded population estimates is the National Aeronautics and is estimated to experience highly unusual monthly summer Space Administration GPWv3 y-2010 gridded population dataset, which is linearly temperatures for a 4°C world (and 11 percent of the population scaled up on a country basis to match the SSP projections, thus neglecting population redistribution within countries. The SSP2 population estimates on country basis with for a 2°C world), mostly located in tropical regions (compare 5-year time steps were obtained from the SSP database as in Schewe et al. (2013). Figure 2.6, upper panel). Heading toward a 4°C world (RCP8.5), 13 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL the affected population increases to about 50 percent in 2040, 10 percent in a 4°C world by the end of the century relative to including some of the most densely populated areas in South East the 1961–1990 baseline. The patterns emerging from the analysis Asia, the Americas, and the Mediterranean. By 2080, 96 percent in this report are broadly consistent with the overall “wet gets of the global population is projected to experience highly unusual wetter” and “dry gets drier” trend under climate change noted summer temperatures and 71 percent to experience unprecedented in the previous Turn Down the Heat reports. The subtropical extreme temperatures (meaning that this would be the new norm regions, particularly those in the northern hemisphere, stick out for more than two-thirds of global population). Estimates are con- as aridity hotspots. This broad global trend does not apply to all siderably lower for a 2°C world (RCP2.6), where highly unusual regions equally, however. One notable exception is the Amazon, summer temperatures are projected to affect about 25 percent which is projected to experience drying despite its very humid of the global population in 2080 and unprecedented summer present-day climate; another exception is East Africa, where wet- temperatures would be very rare. ting is projected. Southern Europe, especially the Balkans and the eastern Mediterranean, are expected to experience aridity index (AI) changes of up to –60 percent in a 2°C world; an AI increase 2.3.4 Projected Changes in Extreme of about 30 percent is projected for large parts of Siberia. These Precipitation trends are strongly amplified in a 4°C world. While regional pre- cipitation projections are very sensitive to the underlying model A warming of the lower atmosphere is expected to strengthen the ensemble, the subtropical drying trend and the Siberian wetting hydrological cycle. The increase in extreme precipitation event inten- trend are consistently found in a wide range of climate models sity over the 21st century is found to be about six percent per °C and model generations (Knutti and Sedlᡠcek 2012). for the CMIP5 model ensemble and thus about three times stronger These trends in aridity also lead to changes in annual water than the increase in mean precipitation of about 1.5–2.5 percent discharge that can be taken as a first-order approximation of the per °C (Kharin et al. 2013). A widely used indicator for persistent water resources available to humans. Figure 2.7 shows the relative heavy rain conditions (and, as such, for potential flood risk) is changes in annual water discharge for a 2°C versus a 4°C world the annual maximum 5-day precipitation accumulation (RX5day). based on Schewe et al. (2013). Profound changes in river runoff The RX5day is found to intensify by about six percent for RCP2.6, are already evident for a 2°C world. A robust reduction in dis- but the change exceeds 20 percent for RCP8.5 for the 2080–2100 charge in the subtropical regions (including Meso-America and the period. While heavy precipitation events become more intense, Caribbean, parts of central and southern South America, Southern they are also projected to increase in frequency. Africa, Western Australia, and the Mediterranean) between 15–45 Kharin et al. (2013) find that annual extremes of daily precipi- percent relative to the 1986–2005 reference period is projected. On tation that are estimated to have a return time of 20 years during the contrary, an increase in water availability is projected for the 1986–2005 will be about 3–4 times as common by the end of high northern latitudes, parts of South and South East Asia (due the 21st century in a 4°C world. In a 2°C world, such events are to an intensification of the Indian Monsoon system), and parts of projected to occur 25 percent more often. central and eastern Africa. However, some of the strongest relative increases are plotted relative to a very low baseline precipitation 2.3.5 Aridity and Water Scarcity (e.g., the Sahara desert or the Arabian Peninsula), which means that these regions will remain very dry. As mean precipitation over large regions of land is increasing, so The changes in projected precipitation patterns for a 4°C world is evaporation, since higher temperatures provide more energy for are not simply a linear extrapolation of the patterns observed for evaporation. Under a changing climate, the land surface warming a 2°C world. While, for a 2°C scenario, discharge changes for will be particularly pronounced and the saturated water vapor central and Eastern Europe and North America are inconclusive, concentration will increase in the lower atmosphere over land a robust drying of 15–45 percent is projected for a 4°C world for surface. If this saturated water vapor concentration exceeds the the southeast, west, and north central North America and for most growth in actual water vapor concentration, this may result in a parts of central and Eastern Europe and the Balkans. Interestingly, drying trend even though absolute precipitation increases. only a slight increase or even decrease in discharge is projected A drying trend over land has been observed globally; this is for central and eastern Africa and the Arabian Peninsula, with strongly heterogeneous, however, with some regions experienc- some countries (e.g., Pakistan) exhibiting a robust increase in ing profound wetting. This is in agreement with Feng and Fu discharge under a 2°C warming scenario that is reduced to zero (2013), who report an increase in global dry land area of about or even negative in a 4°C world. 14 TH E GLOBA L P ICTU RE Figure 2.7: Relative change in annual water discharge for a 2°C world and a 4°C world. Relative to the 1986–2006 period based on an ISI-MIP model intercomparison using climate projections by CMIP5 GCMs as an input for global hydrology models (GHMs)21 (adjusted from Schewe et al. 2013). Colors indicate the multi-model mean change, whereas the saturation indicates the agreement in sign of change over the ensemble of GCM—GHM combinations. Other features present already for a 2°C world (e.g., the sub- Figure 2.8 shows the relative increase in the number of days under tropical drying trend) get more pronounced, reaching reductions in drought conditions for a 4°C world relative to the 1976–2005 baseline annual mean discharge of up to 75 percent in the Mediterranean. (Prudhomme et al. 2013). The subtropical and tropical regions in the In addition, the high-latitude discharge is robustly projected to Mediterranean, Meso-America, Caribbean, southern South America, increase further for a 4°C world, dominated by an increase in and Australia are projected to experience the strongest increase in winter precipitation. A warming of about 3°C is found to lead to an increase in the number of people living under absolute water scarcity (which means less <500 m3 per capita per year); 40 percent higher than projected due to population growth alone Figure 2.8: Percentile change in the occurrence of days under (Schewe et al. 2013).21 drought conditions by the end of the 21st century (2070–2099) in a 4°C world relative to the 1976–2005 baseline. 2.3.6 Droughts While an upward trend for extreme temperature and heavy pre- cipitation events can be clearly deduced from the observational record, there is still considerable debate as to whether or not there is a trend in global drought in recent decades (Dai 2012; Sheffield et al. 2012). Drought projections depend strongly on the underlying methodology, the applied indicators, and the reference periods used (Trenberth et al. 2014). In accordance with the overall drying trend in several world regions, an increase in intensity and duration of droughts is estimated to be likely by the end of the 21st century (IPCC 2013a, Table SPM.1). 21 Please note that not all models of the ISI-MIP ensemble reach 4°C in the 21st cen- tury, which is why these projections are just based on three CMIP5 GCMS, namely White regions: Hyper-arid regions for which runoff is equal to zero more HadGEM2-ES, IPSL-CM5A-LR, MIROC-ESM-CHEM. As in Schewe et al. (2013), these than 90 percent of the time in the reference and future periods. Reprinted GCM projections are then combined with 11 global hydrology models. from Prudhomme et al. (2013). 15 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 2.9: Median yield changes (%) for major crop types in a 4°C world relative to the 1980–2010 baseline from the AgMIP (Rosenzweig et al. 2013a). Please note that this projection only considers global gridded crop models that explicitly account for CO2 effects and nitrogen stress. days under drought conditions. In fact, more than six months per provide a clearer picture of agricultural impacts and risks than year on average is projected under at least moderate drought con- was previously possible. Figure 2.9 shows end of 21st century ditions for the Mediterranean over the 2080–2100 period in a 4°C (2070–2099) changes in yield projections relative to the 1980–2010 world compared to less than a month in a 2°C world (Orlowsky base period from the AgMIP project (Rosenzweig et al. 2013a) for and Seneviratne 2013). Substantial drought risk also emerges for models that explicitly account for CO2 and nitrogen effects, thus large parts of South America, including the Amazon (representing a including the most important processes influencing yield dynam- considerable threat to the Amazonian rainforest (see chapter 3.4.5)). ics (although the magnitude and the interplay of these effects are highly uncertain) (see Box 2.4). 2.3.7 Agricultural Yields While gains in agricultural yields are projected for the high latitudes, substantial losses are projected for the tropical and sub- The impacts of climate change on agricultural production have tropical regions and all major crop types. For wheat and maize, been observed for different crops. The recent IPCC AR5 Working losses may even exceed 50 percent on average for large parts of Group II report states with medium confidence that wheat and the tropical land area. This is consistent with a meta-analysis of maize production has been affected negatively in many regions more than 1,700 published studies on agricultural yield changes and at a globally aggregated level (IPCC 2014c). under climate change that found robust indications for a reduc- Even though projected warming and drying trends present a tion in wheat, rice, and maize production for a local warming of major threat to agriculture, in particular in tropical and sub-tropical 2°C in both temperate and tropical regions without adaptation regions, there remains substantial uncertainty in the projections (Challinor et al. 2014). All studies show a downward trend at of agricultural yields over the 21st century. This is due to uncer- local warming levels of 1°–3°C for both temperate and tropical tainties in climatological forcing as well as in the response of the regions if no adaptation is taken into account. The strongest agricultural models used and their representation of carbon diox- results (40 percent yield decreases under 5°C local warming) ide, nitrogen, and high temperature-related effects on agricultural were found for wheat in tropical areas. Notably, this decrease is yields (Asseng et al. 2013; Rosenzweig et al. 2013b; see Box 2.4 still projected to be about 30 percent for wheat in tropical regions for a discussion of the CO2 fertilization effect). when adaptation measures are considered. The results indicate Recent model intercomparison projects and the meta-analysis significant negative aggregate impacts of 4.9 percent yield losses studies which provide the basis for the analysis in this report per degree of warming. 16 TH E GLOBA L P ICTU RE Box 2.4: The CO2 Fertilization Effect Increasing carbon dioxide concentrations in the atmosphere lead not only to higher air temperatures due to the greenhouse effect but also affect plant productivity and vegetative matter (Ackerman and Stanton 2013). Higher CO2 concentrations also lead to improved water use efficiency in plants through reducing transpiration by decreasing the plant’s stomatal conductance (Hatfield et al. 2011). Plants can be categorized as C3 and C4 plants according to their photosynthetic pathways. Maize, sorghum, sugar cane and other C4 plants use atmospheric CO2 more efficiently, which means that C4 plants do not benefit as much as C3 plants from increasing CO2 concentrations except during drought stress (Leakey 2009). C3 plants, including wheat, rice, and soybeans, use atmospheric CO2 less efficiently and therefore profit from increasing CO2 concentrations (Ackerman and Stanton 2013; Leakey 2009); this possibly attenuates the negative effects of climate change for C3 plants. The degree to which the CO2 fertilization effect will increase crop yields and compensate for negative effects is uncertain, as differing experi- mental designs have shown. For example, Free-Air CO2 Enrichment (FACE) experiments suggested a 50-percent lower CO2 fertilization effect than enclosure studies (Ackerman and Stanton 2013; Long et al. 2006).* Moreover, the yield response to elevated CO2 concentrations seems to be plant and genotype specific and depends on the availability of water and nutrients (Porter et al. 2014). It seems certain, however, that water-stressed crops benefit stronger from elevated CO2 concentrations, and thus rain-fed cropping systems could benefit more than irrigated systems (Porter et al. 2014). A debate is ongoing over whether results from FACE experiments, laboratory experiments, or modeling approaches are closest to reality and results among the different approaches differ greatly (Ainsworth et al. 2008; Long et al. 2006; Tubiello, Amthor et al. 2007). For this reason, many climate impact studies include a scenario with the CO2 fertilization effect and one scenario without considering the fertilization effect. Elevated CO2 concentrations are thought to positively influence future food security through faster plant growth, but they could also have a negative impact through a change in the grain’s protein concentration (Pleijel and Uddling 2012; Porter et al. 2014). This change comes about when the increased biomass accumulation happens faster than the corresponding nitrogen or nitrate uptake, leading to increasing yields but reduced protein concen- tration; this is also known as growth dilution (Bloom et al. 2010; Pleijel and Uddling 2012). This decrease in grain protein concentration, which in experiments with wheat ranged from 4–13 percent (and 7.9 percent in FACE experiments), and with barley ranged from 11–13 percent, is assumed to have negative effects on the nutritional quality of grains (Bloom et al. 2010; Erbs et al. 2010; Högy et al. 2013; Pleijel and Uddling 2012; Porter et al. 2014). In addition, elevated CO2 concentrations are associated with significant decreases in the concentrations of zinc and iron in C3 grasses and legumes (Myers et al. 2014). *The study by Long et al. (2006) triggered a debate and resulted in a critical response by Tubiello et al. (2007) which in turn was answered by the original authors in Ainsworth et al. (2008). Figure 2.10: Global ocean acidification as expressed by a gradual decrease of ocean surface pH (indicating a higher concentration of hydrogen ions—or acidity). Projections for scenarios in this figure are produced by a simple model (bold lines) derived from one of the complex models that is included in the range of projections from IPCC AR5 WGI (shaded ranges). Also indicated is a local historical measurement series at Aloha normalized to (global) model levels. Source: Methodology in Bernie et al. (2010) combined with climate projections as in Figure 2.1; IPCC data WGI IPCC (2013); observed data Dore et al. (2009). 17 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Although crop-level adaptation is found to increase simulated low-lying coastal regions are particularly densely populated and are yields by 7–15 percent, there are still substantial climate-related expected to experience further population increases. The density threats to regional and global crop production. These threats of people living in flood prone, coastal zones and megacities may represent a major reason for concern about global food security, increase by about 25 percent until 2050 (Aerts et al. 2014). Since particularly given population (and food demand) increases. the potential impacts of fast-rising seas will be severe for both the built and natural environments in these regions, avoiding sea-level 2.3.8 Ocean Acidification rise is a strong reason to engage in mitigation efforts. Paleo-reconstructions show that sea levels were significantly Rising atmospheric CO2 concentrations not only cause land surface higher during warmer periods of earth history. The rate of sea-level and sea-surface temperatures to rise but also leave their imprint rise increased from tenths of mm per year in the stable climate of on ocean chemistry. In order to restore the balance between the the last 5,000 years to 1.7 mm per year during the 20th century— atmosphere’s and the ocean’s CO2 concentration, the oceans absorb and 3.2 mm per year from 1993–2010 (Church et al. 2013). The present additional CO2 as atmospheric concentrations rise. The oceans have rate of sea level rise is the highest in recent history, i.e. higher than taken up approximately 25 percent of anthropogenic CO2 emis- all rates within the past 6,000 years as inferred from proxy records sions in the period 2000–06 (Canadell et al. 2007). CO2 dissolves (Lambeck et al. 2014). Sea-level rise is the result of the thermal in seawater and eventually forms a weak acid—a process known expansion of sea water and melting of land ice from glaciers and ice as ocean acidification (Caldeira and Wickett 2003). sheets, with an increasing contribution from the latter in the past 20 The increase in CO2 concentrations to the present-day value of years. In its latest assessment report, the IPCC projects rates of up to 396 ppm has caused the ocean surface pH to drop by 0.1 from pre- 16 mm per year by the end of the 21st century (Church et al. 2013). industrial days (Raven 2005), equivalent to a 30 percent increase Projections of global and regional sea-level rise in this report are in ocean acidity. In fact, current rates of ocean acidification appear draw upon the IPCC AR5 WGI report (Masson-Delmotte et al. unprecedented over the last 300 million years (WMO 2014). A 2013) and more recent estimates of the Antarctic contribution 4°C or higher warming scenario by 2100 corresponds to a CO2 (Bindschadler et al. 2013; Hinkel et al. 2014; Levermann et al. concentration above 800 ppm and will lead to a further decrease 2014) (see methods in Appendix). This report follows a process- of pH by another 0.3, equivalent to a 150 percent acidity increase based approach based on the outcome of physical models. This over pre-industrial levels (World Bank 2012a). Such changes in differs from the earlier Turn Down the Heat reports (World Bank ocean chemistry will induce dramatic, though uncertain biological 2012a, 2013), where semi-empirical methods (Rahmstorf 2007; responses. A recent meta-analysis of biological responses suggests Schaeffer et al. 2012) were applied due to the lack of sufficient that the effects are likely to interact with other changes, including progress at the time in developing and validating process-based rising temperatures (Harvey et al. 2013). models. The adoption of process-based estimates in this report An important effect of the reaction of CO2 with seawater is is the result of recent advances in modeling and the narrowing the reduction of carbonate ions available for skeleton- and shell- gap between process-based estimates and observations (Church forming organisms in the form of calcium carbonate (CaCO3). et al. 2013; Gregory et al. 2013). Compared to the IPCC AR5, the Surface waters are usually super-saturated with aragonite, which projections for Antarctica in this report are scenario-dependent is a mineral form of CaCO3. In the case of coral reefs, decreasing and yield higher upper bounds (up to about 0.2 m higher) as availability of carbonate ions is expected to increase vulnerability they explicitly account for different levels of ocean warming and to the effects of rising temperatures and to hinder recovery fol- resulting ice sheet melt (see discussion in Appendix). lowing hurricanes and other extreme events (Dove et al. 2013). It is important to note that large uncertainties remain in predict- In combination with warming waters and other anthropogenic ing future sea-level rise and, in particular, the contribution from stresses, including overfishing and pollution, ocean acidification potentially unstable regions of marine ice in Antarctica (Church poses severe threats to marine ecosystems. Section 3.4.6 provides et al. 2013). The results in this report incorporate the direct effect an overview of some of the most recent scientific publications on of Southern Ocean warming on ice-shelf basal melting and related the expected impacts. While those impact projections are focused on ice stream acceleration in Antarctica; as in the IPCC AR5, however, Latin America and the Caribbean, this does not imply that they are not they do not include amplifying feedbacks responsible for marine expected to occur in other regions. In fact, the levels of acidification ice sheet instability. In light of model shortcomings and increasing are projected to be above average in cold waters at higher latitudes. evidence of marine ice sheet instability, this report cannot provide 2.4 Sea-Level Rise a very likely range for sea-level rise from Antarctica. This report thus follows the IPCC AR5 approach and assesses the model-based, Sea-level rise is one of the main consequences of global warming 90 percent range as a likely (67 percent) range only. This applies with direct and fundamental impacts on coastal regions. Many not only to Antarctica but to all sea-level rise contributions (see 18 TH E GLOBA L P ICTU RE Figure 2.11: Global mean sea-level rise projection within the Table 2.1: Sea-level rise projections to 2081-2100 above the 21st century. 1986–2005 baseline, in meters (unless indicated otherwise). RCP2.6 RCP8.5 (1.5°C WORLD) (4°C WORLD) Steric 0.13 (0.1, 0.18) 0.27 (0.2, 0.32) Glacier 0.12 (0.07, 0.17) 0.18 (0.13, 0.27) Greenland 0.07 (0.02, 0.12) 0.11 (0.06, 0.21) Antarctica 0.04 (–0.01, 0.19) 0.04 (–0.03, 0.3) SLR in 2081–2100 0.36 (0.20, 0.60) 0.58 (0.40, 1.01) SLR in 2046–2065 0.22 (0.14, 0.35) 0.27 (0.19, 0.43) SLR in 2100 0.4 (0.21, 0.67) 0.68 (0.48, 1.23) Rate of SLR in 2046–2065 4.2 (2.2, 7.8) 7.1 (5.3, 12.9) (mm/yr) Rate of SLR in 2081–2100 3.9 (1.3, 7.2) 10.8 (7.5, 21.9) (mm/yr) The lower and upper bounds are shown in parentheses (likely range24). The sum of median contributions is not exactly equal to the total due to rounding. The upper and lower bounds of the total sea-level rise (SLR) are Time series for sea-level rise for the two scenarios RCP2.6 (1.5°C world, smaller than the sum of each contribution’s upper and lower bound because blue) and RCP8.5 (4°C world, green). Median estimates are given as lines errors are not necessarily correlated (see methods in Appendix). Note that and the lower and upper bound given as shading (likely range23). The esti- the land-water contribution is not included. mate includes contributions from thermal expansion, glaciers and ice caps, and the Greenland and the Antarctic ice sheet, but not from anthropogenic groundwater mining (estimated to 0.04 +/– 0.05 m by the IPCC AR5 over the projection period). The estimates are skewed toward high values mainly because of a possible, yet less probable, large Antarctic contribution to sea-level rise. The sea-level rise baseline is 1986–2005, which represents 2100. The slow response of sea levels also explains the small dif- a sea level of about 0.2 m higher than pre-industrial levels. ference between the RCP2.6 and RCP8.5 sea-level rise estimates in 2100 relative to the seemingly much larger divergence in global mean temperature projections (Table 2.1). The effect of a large methods in Appendix). The lower and upper bounds can thus be proportion of 21st century emissions on sea-level rise will only interpreted as likely ranges. Note also that sea-level rise contri- become visible in the decades and centuries beyond 2100, when butions not related to climate warming, such as resulting from sea level projections diverge more strongly between the RCP2.6 groundwater mining, are not included and should be added on and RCP8.5 scenarios (Church et al. 2013). top of our global projections. The future divergence is foreshadowed by the difference in the This report projects 0.58 m of globally averaged sea-level rise rates of sea-level rise toward the end of the century. For RCP2.6, in a 4°C world (RCP8.5) for the period 2081–2100 compared to this report projects a rate of 3.9 (1.3–7.2) mm per year as the the reference period 1986–2005,22 with the low and high bounds mean over the 2081–2100 period, which is comparable with the being 0.40 and 1.01 m (Figure 2.11). present-day rate of 3 mm per year. This is in strong contrast to For the scenario RCP2.6 (classified as a 1.5°C world for the the report’s projections for RCP 8.5, where the rate of sea-level model ensemble used here, see Box 2.1) this report projects 0.36 m rise, at 10.8 (7.5–21.9) mm per year, is two to three times higher. of sea-level rise (0.20 m–0.60 m) for the same period, a reduc- In addition, this report projects the risk of a rate of sea-level rise tion of almost 40 percent compared to RCP8.5. This potential for higher than 20 mm per year in a 4°C world toward the end of sea-level rise mitigation is broadly consistent with median IPCC the century. Sea-level rise is projected to continue for centuries AR5 estimates and emphasizes a larger benefit than previously to millennia to come. Based on paleo-evidence as well as model estimated from emissions reductions.23 results, Levermann et al. (2013) estimate the sea-level commitment Even when global mean temperatures stabilize, as in the over the next 2000 years to be about 2.3 m per degree of global RCP2.6 scenario, sea level is projected to continue to rise beyond mean temperature warming.24 22 The 1986–2005 baseline period is about 0.2 m higher than pre-industrial times. 23 24 The likely range (67 percent) is computed from the model-based 90-percent range. The likely range (67 percent) is computed from the model-based 90-percent range. 19 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL 2.4.1 Marine Ice Sheet Instability oceans (Lombard et al. 2009), where additional heat from global warming is not stored equally. Associated density changes (steric Post-IPCC literature provides further evidence of marine ice sheet changes) are further modulated by salinity changes, for example instability (Mercer 1978; Oppenheimer 1998) in both West and freshening in high latitudes from increased precipitation, runoff, East Antarctica (Joughin et al. 2014; Mengel and Levermann 2014). and ice melt (Bamber et al. 2012; Durack et al. 2012; Pardaens Marine ice sheets predominantly rest on solid rock below sea level. et al. 2011). Rapid disintegration of the ice sheets can be triggered by oceanic Wind patterns are also critical in redistributing the heat and melting underneath the floating ice shelves surrounding the ice shaping the sea surface (Timmermann et al. 2010). They are sheets and is associated with a self-enforcing dynamic feedback responsible for strong natural, cyclical variability such as related mechanism, where the retreat of the grounding line25 being exposed to the Pacific Decadal Oscillation, where high sea levels in the to warming waters accelerates poleward on downward-sloping Eastern Pacific occur along with low sea levels in the Western bedrock. This rapid retreat of the grounding line is currently being Pacific (e.g. Bromirski et al. 2011), and inversely. Sustained observed for several large glacier systems in West Antarctica, includ- changes in wind-induced sea level trends are also expected as the ing the Pine Island and Thwaites glacier (Rignot et al. 2014), and climate warms. The ocean dynamical response to changing density numerical models indicate that a collapse of these glaciers might and atmospheric conditions is complex and GCMs show a large already be underway (Favier et al. 2014; Joughin et al. 2014). If spread (Pardaens et al. 2011; Yin 2012). A well-known example this proves to be true, it would result in an irreversible additional of dynamic-steric sea-level rise is a projected 10–30 cm additional sea-level rise of about one meter on multi-centennial time scales rise in the northeastern coast of the United States associated with with the potential for a destabilization of the entire West Antarctic a slowing down of the Atlantic meridional overturning circulation ice sheet (containing ice that is equivalent to about 4 m of sea-level (Schleussner et al. 2011; Yin and Goddard 2013; Yin et al. 2009). rise). In East Antarctica, Mengel and Levermann (2014) identified Signs of such an accelerated sea-level rise at this coastline are a potential marine ice sheet instability of the Wilkes basin (3–4 m already visible in the tide gauge record (Sallenger et al. 2012). sea-level equivalent). In addition to ocean circulation and density changes, melting The potential of warm water intrusion has been shown for at glaciers and ice sheets induce a redistribution of mass as melt least one large ice shelf cavity (Hellmer et al. 2012). However, the water spreads into the oceans, reducing the gravitational pull of skill of climate models in predicting the oceanic dynamics sur- the formerly glaciated regions and potentially causing local land rounding the ice sheets is still low; as a result, it is very uncertain uplift (Bamber et al. 2009; Mitrovica et al. 2001). This results in how and at what levels of global warming disintegration of this ice above-average sea levels far away from melting ice masses and sheet could be triggered. At the same time, new insights into the below-average rise or even relative drops in sea levels drop in bed topography of the Greenland ice sheet reveal deeply incised their proximity. submarine glacial valleys that control about 88 percent of total ice In the projections in this report, these processes are accounted discharge from Greenland into the ocean; this indicates a much for by analyzing steric-dynamic GCM outputs and combining projec- greater sensitivity of the ice sheet to oceanic melt than previously tions for glacier and ice sheet contributions with regional fingerprints thought (Morlighem et al. 2014). (see Appendix). This is similar to previous Turn Down the Heat reports. This report focuses on long-term changes and thus does 2.4.2 Regional Distribution of Sea-Level Rise not attempt to predict natural, up to multi-decadal, variability in sea levels. Consistent with the previous Turn Down the Heat reports, Sea-level rise is not distributed equally across the globe. This is this report only includes climate-related contributions to sea-level clearly visible in satellite observations (Meyssignac and Cazenave rise and omits such other, more local contributions as ongoing 2012) and tide-gauge reconstructions (Church and White 2011), glacial isostatic adjustment since the last glacial ice age (Peltier and this inequality is projected to be amplified under future sea- and Andrews 1976), sediment transport—especially in river deltas level rise (Perrette et al. 2013; Slangen et al. 2011). Recent sea level (Syvitski et al. 2009), and local mining (Poland and Davis 1969). trend patterns have been dominated by thermal expansion of the These local factors may provoke vertical land movement and contribute to enhanced (in case of subsidence) or reduced (in case of uplift) relative sea-level rise at the coasts. They should 25 The point at where the ice sheet is grounded on solid rock, marking the transition from the floating part that already contributes to global sea levels (ice shelf) to the be accounted for by local planners, for example by using Global grounded ice sheet that does not. Positioning System measurements (Wöppelmann and Marcos 20 TH E GLOBA L P ICTU RE Figure 2.12: Patterns of regional sea-level rise. Median (left column) and upper range (right column) of projected regional sea-level rise for the RCP2.6 scenario (1.5°C world, top row) and the RCP8.5 scenario (4°C world, bottom row) for the period 2081–2100 relative to the reference period 1986–2005. Associated global mean rise are indicated in the panel titles, consistent with Table 2.1. 2012). Another limitation of the projections in this report results Near-polar regions are projected to experience sea-level from the poor skills of global models at representing local oce- rise below the mean, with sea-level fall occurring at the coasts anic processes, especially in semi-enclosed basins such as the very close to the mass losses of the big ice sheets. A map of Mediterranean Sea. regional anomalies from the global mean rise shows that, in Regional sea-level rise is projected to exceed the global mean most coastal areas away from the poles, sea-level rise tends to at low latitudes (Figure 2.12). Since the ice sheets are concentrated remain within +/– 10 cm of the global mean rise of 0.58 m near the poles and melting is projected to increase significantly (Figure 2.13, top). In these areas, uncertainty in global mean toward the end of the century, their decreasing gravitational sea-level rise dominates the total uncertainty (excluding local, pull piles up water in the low latitudes (Figure 2.13, bottom). non-climatic processes). It is also clear from Figure 2.12 that This effect is stronger than freshening-induced steric rise in the global sea level uncertainty is comparable in magnitude to dif- Arctic (Figure 2.13, middle). Due to the large contribution of the ferences between emissions scenarios, but both the “medium” Antarctic ice sheet in this report’s high estimate, sea-level rise at and “high” scenarios in this report indicate a mitigation poten- low latitudes will exceed the global average (Figure 2.12, right tial of about 40 percent in sea-level rise between RCP8.5 and column). Therefore, large ice sheet melt should have strongest RCP2.6. A more detailed analysis is conducted for each region impacts at the tropical coastlines. covered in this report. 21 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 2.13: Regional anomaly pattern and its contributions in 2014). Figure 2.14 presents a framework for understanding key the median RCP8.5 scenario (4°C world). factors leading to social vulnerability to climate change, based on an extensive literature review in the three focus regions and beyond (see Appendix, Summary of Evidence Concerning Social Vulnerability, and Table 2.2). Chapters 3–5 draw on elements of this framework as they highlight the potential social impacts of climate change in the three focus regions. 2.5.1 Interaction of Key Current and Future Development Trends with Climate Change Looking toward the late 21st century and the possible consequences of up to 4°C global warming, changes in demographic, economic, technological, sociocultural, and political conditions are likely to profoundly shape the vulnerability and adaptive capacity of societies and particular social groups (Hallegatte et al. 2011). Key development trends include population dynamics, migration, and urbanization; overall economic development, which will influ- ence both patterns of emissions and the resources at different social groups’ disposal to respond to threats to their livelihoods and wellbeing; and the availability of and demand for natural resources. Adaptation and mitigation policies will also significantly influence different social groups’ vulnerability to the effects of climate change. These trends will both affect and be affected by the impacts of climate change. 2.5.2 Understanding Vulnerability, Adaptive Capacity, and Resilience How different groups experience climate change is strongly influ- enced by their capacity to adapt, which itself depends largely on access to resources (Moser et al. 2010; Wisner et al. 2012)—economic Total sea-level rise (top), steric-dynamic (middle), and land-ice (bottom) assets and incomes, natural and physical resources, social networks, contributions to sea-level rise, shown as anomalies with respect to the global mean sea-level rise. Global mean contributions to be added on top cultural knowledge, and access to political power (Chambers and of the spatial anomalies are indicated in the panel titles (see also Table 2.1). Conway 1991). Processes underpinning inequality and social exclu- sion (related for example, to ethnicity or class) frequently lead to increased exposure to hazards and prevent certain groups from acquiring the resources needed to protect themselves and recover from shocks (i.e., to become resilient to climate change). The typol- ogy developed by Moser et al. (2010) defines three key elements 2.5 Social Vulnerability to Climate of vulnerability: spatial and physical vulnerability, socioeconomic Change vulnerability, and political/legal vulnerability. People’s vulnerability to climate change, their capacity to adapt, 2.5.3 Spatial and Physical Vulnerability and their resilience in the face of its impacts reflect a combina- tion of geographical exposure to hazards and varied demographic, Coasts and deltas, tropical forests, mountainous regions, arid and socioeconomic, and political factors (Beck 2010; Hewitt 1997; semi-arid areas and the Arctic are all identified as geographical Ribot 2010; Wisner et al. 2004). For the first time, the most regions particularly sensitive to climate change (IPCC 2014a; World recent report of the IPCC WGII includes a chapter on the liveli- Bank 2012a). Within these regions, vulnerabilities vary consider- hoods and poverty dimensions of climate change (Olsson et al. ably. Typically people living in areas that are prone to hazards, 22 TH E GLOBA L P ICTU RE Figure 2.14: Framework for understanding social vulnerability to climate change Expected climate change: Current/future development trends: • Rising air and sea • Population dynamics (growth, migration, urbanisation) surface temperatures • Globalisation and economic processes • Sea level rise • Demand for and availability of natural resources • Changing precipitation • Mitigation policies • Extreme weather events Vulnerability and access to resources: Environmental impacts Social implications: on ecosystems: Spatial and physical vulnerability • Food insecurity • Geographical location • Loss of biodiversity • Increased health • Lack of physical resources • Changes in land risks productivity Socio-economic vulnerability • Increased poverty • Ocean acidification • Poverty/lack of income • Aggravated • Water scarcity • Social exclusion inequalities • Depletion of fish stocks • Limited access to formal education • Increased migration • Glacier retreat • Exacerbated Political-legal vulnerability • Salinization of farm land conflict • Lack of adequate policies • Desertification • Unequal power structures • Lack of institutional support Adaptive capacity: Social inequalities on the basis of: • Coping strategies • Class • Adaptation strategies (Autonomous/Planned) • Gender • Social protection • Race/Ethnicity • (Dis)ability • Age Source: Adapted from Verner (2010). such as flooding, landslides or cyclones, and in areas affected by 2.5.4 Socioeconomic Vulnerability drought or heat extremes are most likely to be negatively affected by climate change. Loss of ecosystem services, related for example 2.5.4.1 Poverty to loss of forest cover, reduction in overall water supply or increased Economically and socially marginalized people’s access to resources salinity, is likely to put increasing stress on livelihoods and wellbe- to adapt to anticipated environmental stresses such as slow-onset ing in affected areas, such as, for example, growing pressures on climatic change, and to cope with extreme events, is often impeded water availability in many Latin American cities. There is a strong because their needs are overlooked or due to discrimination (Wisner (but not complete) overlap between geographical and economic et al. 2012; Tanner and Mitchell 2009). Economic and social mar- vulnerability—with a growing risk of people being ‘trapped’ in ginalization, in turn, reinforces geographical vulnerability. Thus, for climatically stressed areas because they lack the financial capital or example, over one million people live in Rio de Janeiro’s favelas that social connections to move, and are unable to diversify into more sprawl over the slopes of the Tijuca mountain range, making them resilient livelihood activities (Black et al. 2011). Likewise, poorer particularly at risk from mudslides (Hardoy and Pandiella 2009). people often have no choice but to live in hazard-prone locations, Poorer people often take longer to bounce back after such where housing is cheaper (Winchester 2000). For example, many shocks as flooding and cyclones, reflecting their more limited of the victims of Saudi Arabia’s 2009 floods were migrant workers physical, financial, human, and social assets. After Hurricane who lived in poorly constructed, informal shanty houses in the Mitch in Honduras, for example, many poor people took 2–3 years wadi (natural drainage) area (Verner 2012). longer than better-off people to rebuild their livelihoods (Carter et 23 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL al. 2007). In Mexico, meanwhile, Hurricanes Stan and Wilma had Where gender relations are more egalitarian, differentials in much longer-lasting effects in Chiapas (a poorer state) than the disaster mortality are much lower. In some contexts, however, similarly affected but better off Yucatan (Rossing and Rubin 2010). norms of masculinity that mandate heroic behavior can lead to Poorer people’s livelihoods often depend on sectors such as greater death rates among men. This was seen, for example, in agriculture, forestry, fishing, and pastoralism that are particularly Central America after Hurricane Mitch (Neumayer and Plumper sensitive to the effects of slow-onset climatic changes (Leichenko 2007; Bradshaw and Fordham 2013). Both disasters and daily pres- and Silva 2014). Poor people’s adaptive capacity is often under- sures can increase the work needed to ensure basic survival, which mined by lower education levels, limited alternative livelihood can disproportionately affect women if household tasks become options, discriminatory social norms that affect their access to more time consuming. Additionally, they can also exacerbate stress labor markets and decent work, and a lack of long-term institu- factors and may increase violence against women and girls (Azad tional planning, policy, and programmatic support for resilience- et al. 2014; Brody et al. 2008; Enarson 2003). strengthening activities (UNISDR 2009). Although significant There is increasing evidence that promoting gender equality reductions in extreme poverty are projected by 2030–50,26 up to is an important component of an effective strategy for developing 325 million extremely poor people will be living in the 45 coun- resilience to climate change, since gender differentials in educa- tries most exposed to drought, extreme temperatures, and flood tion, access to and control of assets, access to information and hazards in 2030. Most of these people will reside in isolated rural social networks can all limit disadvantaged women’s adaptive areas in South Asia, Sub-Saharan Africa, and Central America and capacity (Ahmad 2012; Verner 2012; World Bank 2011), and thus the Caribbean (Shepherd et al. 2013). As a result, poverty and undermine overall resilience. geographical exposure to climate change are likely to continue to reinforce one another in the medium term. 2.5.4.3 Vulnerability Related to Age and Disability Age greatly influences people’s ability to face climate-related 2.5.4.2 Gender Dimensions of Vulnerability threats. During extreme climatic events, older people’s reduced Different gender roles and norms mean that men, women, girls, and mobility, strength, and health; impaired eyesight and hearing; boys are likely to be affected in different ways by climate change and greater vulnerability to heat and cold can restrict their ability (Demetriades and Esplen 2009; Denton 2002). For example, water both to cope and to escape danger (HelpAge International 2012). scarcity can lead to particular challenges for women and girls, who Thus, for example, the European heat wave of 2003 (Vandentor- in many cultures have primary responsibility for obtaining water ren et al. 2006) and the 2013 heat wave in Pudong, China (Sun et al. for domestic use, and for cooking, laundry, and bathing of young 2014) led to disproportionate death rates among older people. children. In contexts where much subsistence farming is carried In both cases, people over 80 were among the most affected. In out by men, and social norms concerning masculinity frame men France, both older people from poorer socioeconomic groups and as breadwinners, boys and men are more likely to migrate for work people with more limited social networks were disproportionately if climate-related stresses on agriculture undermine livelihoods. affected (Vandentorren et al. 2006). Similarly death rates during Rural women’s access to land, property rights, financial Chicago’s 1995 heat wave were lower in more socially connected resources, and representation in decision-making processes is also neighborhoods (Klinenberg 2002). often restricted compared to men’s access (Agarwal 1994; Brody Poorer older people face particular challenges in adapting et al. 2008). This constrains women’s capacity to diversify into their housing to cope with changing climates; in the absence of alternative livelihoods in the event of climate or other stresses effective social protection systems they may also be unable to (Demetriades and Esplen 2009; Verner 2012) and to anticipate and afford to purchase food, fuel, and water as prices rise. This can prepare for disasters and environmental stresses (Enarson 2003; make them reliant on farming and ecosystem services, and con- Fordham 2012). Furthermore, in some contexts (e.g., in parts of sequently even more vulnerable to climate change (Wang et al. the Middle East and North Africa, and in areas of Latin America) 2013). Globally, the number of older people is expected to triple social norms restrict girls from learning important survival skills by 2050, accounting for one-third of the population in developed (e.g., swimming). This is one reason why women are typically regions and one-fifth of the population in developing countries more likely than men to die as a result of climate related disasters. (UN 2010). The combination of aging and urbanization could lead to a significant increase in the number of people vulnerable to 26 This is expected to be a result of growth in emerging economies (Dadush and climate-related stresses. Stancil 2010; Edward and Sumner 2013), but these projections do not typically take into account the effects of climate change. As a result, they may underestimate Disabled people, are often more dependent on household and climate-change-related impoverishment. community members to fulfill their daily basic needs; they are 24 TH E GLOBA L P ICTU RE thus at higher risk if supportive infrastructures and social relation- agriculture and construction, where they are vulnerable to heat ships are strained or limited. Disabled people are over-represented stress and to unfavorable weather conditions (Lowry et al. 2010). among the very poor in low-income countries and face greater risk This is seen, for example, in the Gulf States (Verner 2012). In of death, injury, discrimination, and loss of autonomy (Priestley addition, migrant workers are often neglected when it comes to and Hemingway 2006). Moreover, they are often ignored in pre- disaster recovery. This is often compounded where migrants lack paredness, recovery and adaptation planning. the legal status to apply for support (Abramovitz 2011). Children are also disproportionately affected by climate change (O’Brien et al. 2008; UNICEF 2008). Food insecurity can have 2.5.4.5 Political/Legal Vulnerability particularly negative effects on their development (Bartlett 2008; Social status, discrimination patterns, and access to resources are Shepherd et al. 2013). Children are also at greater risk than adults almost always determined by political processes and the distribution of mortality and morbidity from malnutrition, disasters and their of power (Mascarenhas and Wisner 2012). The most vulnerable consequences, and from diseases (e.g., malaria and waterborne groups are often the least vocal, mobilized, and empowered in diseases) that may become more widespread as a result of climate decision-making processes, and they often suffer disproportion- change. An additional challenge is that some family coping strate- ately from weak local institutions. As a result, they are frequently gies, such as withdrawing children from school or marrying off deprioritized for infrastructure investments (such as storm drains) daughters to reduce the number of mouths to feed (or to being that would reduce their vulnerability to extreme events (Hardoy new assets into the household) end up jeopardizing the wellbeing and Pandiella, 2009) and for services like training and education of children (Brown et al. 2012). that could help them build adaptive capacity (Rossing and Rubin Lost human development opportunities in childhood can have 2011). Vulnerable groups are also less likely to be able to influence lifetime consequences. Evidence from Zimbabwe, for example, disaster risk reduction planning (Douglas et al. 2008; Tacoli 2009) indicates that children affected by drought and food insecurity and to get adequate relief after disasters (Ruth and Ibarran 2009). in infancy never catch up on lost growth (Hoddinott and Kinsey After the cyclone that hit Orissa, India, in 1990, for example, more 2001). If these negative consequences (e.g., increased rates of than 80 percent of disabled persons faced food shortages due to malnutrition, lost educational opportunities) become more com- a lack of clear information on the location of relief supplies and mon, climate change could lead to an increase in intergenerational how to access them (Handicap International 2008). poverty cycles (Harper et al. 2003)—thus compounding vulner- Weaker institutions in areas primarily populated by marginal- ability to climate change. ized groups increases their vulnerability to the effects of disasters (Kahn 2005)—possibly because building codes and zoning controls 2.5.4.4 Ethnicity and Belonging to Minority Groups are less well enforced where institutions are weak. Marginalized Indigenous and minority groups disproportionately live in areas groups are also often powerless to prevent processes of adaptation already affected by climate change where livelihoods are increas- or development that increase their vulnerability (e.g., capture of ingly undermined—including the Amazon (Kronik and Verner water resources by better-off groups and/or large-scale businesses 2010), the Andes (Hoffman and Grigera 2013), and dryland areas in some parts of the Caribbean and the Andes (Buytaert and De (Macchi et al. 2008). They are typically more likely than majority Bièvre 2012; Cashman et al, 2010)); they also lack the political clout groups to be poor and to have limited access to public services, to ensure that mitigation strategies (such as REDD+) take their employment, education, and health care (UNPFII 2009; Care 2013; interests into account. Where marginalized groups are involved World Bank 2014). In addition, indigenous and minority groups in adaptation planning, however, measures to reduce the risk of often have less access to early warning systems; likewise, they disasters can be much more effective. In St Lucia, for example, are often excluded from or underrepresented in decision-making participatory planning in a low-income urban community involv- processes (Abramovitz 2011; Salick and Ross 2009; Vásquez- ing both local people and engineers led to the identification of Léon 2009). Research with Aymara villagers in Bolivia found, for effective ways to stabilize slopes and a reduction in vulnerability example, that stresses related to the impact of the retreat of the to rain-induced landslides (Arnold et al. 2014). Likewise, where Mururata glacier on water supply were accompanied by historical disadvantaged groups organize for development purposes, these marginalization due to a lack of official identity cards, land titles, can enable individuals and communities to implement natural and access to bilingual basic education (McDowell and Hess 2012). resource management measures that enhance resilience, as India’s Another challenge for indigenous and minority groups in self-help group experience shows (Arnold et al. 2014). many countries is that migrant laborers (often members of ethnic Political/legal vulnerability also involves a lack of recogni- minorities) are more likely to work outdoors, in sectors such as tion of rights. For example, many poor people’s vulnerability is 25 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL compounded by an insecurity regarding tenure rights to housing climate change impacts is provided in Appendix, Summary of and land. This can lead to people being forced to live on marginal Evidence Concerning Social Vulnerability. land that is highly vulnerable to flooding; it can also limit incen- As Table 2.2 shows, the strongest evidence of the likely social tives to invest in hazard-proofing housing where land rights are impacts of climate change relates to impacts on human health: unclear and people can be evicted at any time (Moser et al. 2010). there is clear evidence of the negative effects of increasing tem- Effective public policies can play an important role in reduc- peratures and of the spread of some vector- and water-borne ing vulnerability and boosting adaptive capacity. Social protec- diseases. There is also clear evidence of impacts on some sectors tion measures range from disaster risk preparedness and rescue (and livelihoods), such as fisheries and crop production, as well to cash transfers and insurance to protect assets and livelihoods as moderate evidence of negative impacts on many other areas of (Kuriakose et al. 2012). Finally, the extent to which effective food security, leading to projections of increased poverty among adaptation or mitigation policies, climate-sensitive social protec- small-scale producers and low-income urban consumers. tions, and disaster risk reduction measures are institutionalized There is also moderate evidence that climate change is likely reflects, at least in part, a government’s political orientation and to lead to increased migration, as people move to try to develop institutional capacity as both factors contribute to increasing or more climate-resilient livelihoods or as a result of disaster-related reducing the vulnerability of exposed groups to climate change displacement. While many studies have probed whether climate (Cannon 2000; Hewitt 1997). change is likely to be associated with reduced social cohesion and increased future conflict, there is no overall consensus; 2.5.5 Evidence of the Social Implications the studies generally agree, however, that the risk of increased of Climate Change tensions and violence cannot be ruled out. Two areas that are increasingly flagged in the literature on social vulnerability, but The framework outlined in Figure 2.14 focuses on five main areas for which there is as yet limited evidence, are the impacts of of the social dimensions of climate change: food security and climate change on mental health and its relation to incidences nutrition, income/consumption poverty, human health, migra- of domestic violence. tion, and social cohesion (i.e. the issues in the central orange Overall, the literature reviewed suggests that helping vulnerable box). Table 2.2 below provides a summary assessment of the people build stronger, more climate-resilient lives and livelihoods state of evidence on the social implications of climate change is critical for reducing social vulnerability to climate change. Devel- in these areas, drawing both on literature concerning the three opment investments that reduce poverty by enabling people to regions of focus in this report and the wider literature from other build assets of all types (including stronger physical and financial parts of the globe covered in the previous Turn Down the Heat assets) is a vital component of this; so is much greater investment reports. Table 2.2 follows IPCC Working Group II judgments on in disaster risk reduction and response capacity. Focused efforts the strength of evidence made in the 5th Assessment Report, with to combat the drivers of social exclusion and systemic inequali- author assessments on issues not covered by the IPCC.27 A more ties (such as gender inequalities) that underlie vulnerability will detailed summary of available evidence of social vulnerability to also be necessary. Effective poverty reduction and disaster risk management require investments in governance: to build strong 27 On issues where there are no assessments in the IPCC Working Group II 5th Assess- institutions that are capable of planning and implementing policies ment Report, this report has used the following criteria for assessing the strength of and programs, and to ensure that the voices of affected or likely-to- evidence: a strong evidence base denotes a consensus among studies, and/or eight be-affected people (including those of women, young people, and or more studies with similar findings; a moderate evidence base denotes mixed findings, or 4–7 studies with similar findings; a limited evidence base indicates other socially excluded groups) are much more strongly taken into inconclusive findings or fewer than three studies. These relatively small numbers account in both disaster preparedness and longer-term planning reflect the limited evidence base on many aspects of social vulnerability. and management of climate resilience activities. 26 TH E GLOBA L P ICTU RE Table 2.2: Evidence Summary—Social Vulnerability to Climate Change. POTENTIAL IMPACTS OF INTERACTING CLIMATE CHANGE EVIDENCE AND BROADER DEVELOPMENT BASE AND EXAMPLES OF AFFECTED REGIONS, EXAMPLES OF AFFECTED SOCIAL TRENDS CONFIDENCE SECTORS, AND AREAS GROUPS Food Security and Nutrition Reduction of land available suitable for crops MENA : Israel, Occupied Palestinian Small scale farmers and marginalized groups and ecosystems Territory, Lebanon, Syria, Iraq and the Islamic likely to be displaced by competition for land Republic of Iran Indigenous communities and small-scale farmers who lack land entitlements Reduction in crop productivity especially Tropical and subtropical regions Rural food producers for wheat and maize and negative yields Rain-fed agriculture in LAC Low income urban consumers impacts for nuts and fruit trees Western Balkans Groups reliant on glacial melt water Central Asia Reduction in affordability of food and/or Low-income and food-importing countries Low-income people in rural and urban areas variability of food prices Africa Children at risk of malnutrition LAC (northeast Brazil, parts of the Andean region) Central Asia MENA Increased livestock vulnerability and Arid and semi-arid regions Agro-pastoralists and pastoralists mortality Europe and North America Disruption to fishery and shellfishery Tropical developing countries Artisanal fishermen production, including fish migrations Decrease in fishery catch potential at the People engaged in fish processing and trading Caribbean coasts, the Amazon estuaries, Small coastal communities and the Rio de la Plata Increase in fishery production at higher latitudes Declines in coral reefs resulting in declines in Caribbean Small coastal communities relying on coral fish stocks Western Indian Ocean ecosystems People engaged in fish processing and trading Poverty Impacts Increase in poverty headcount rate and Sub-Saharan Africa (Malawi, Mozambique, Urban poor groups and urban wage laborers risk of chronic poverty in different warming Tanzania, Zambia), Bangladesh, and Mexico Residents of informal settlements scenarios Dwellers in rural hotspots where hunger is expected to become prevalent Increase in disaster related impoverishment Exposed areas globally (e.g., low-lying Low-income groups and destruction of assets; risk of chronic coastal areas, flood-prone land, mountain Children and adolescents (stunting and missing poverty compounded by limited access to slopes) education) disaster relief Self-employed urban groups Coping strategies with negative social Exposed areas globally Low-income groups impacts Children (e.g. child labor, removal from school) Girls/young women (e.g., forced marriage) Strained social cohesion and decline in Exposed areas globally Low-income groups, groups experiencing reciprocity sudden impoverishment and competition for resources Strong Evidence Moderate Evidence Limited Evidence 27 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Table 2.2: Continued. POTENTIAL IMPACTS OF INTERACTING CLIMATE CHANGE EVIDENCE AND BROADER DEVELOPMENT BASE AND EXAMPLES OF AFFECTED REGIONS, EXAMPLES OF AFFECTED SOCIAL TRENDS CONFIDENCE SECTORS, AND AREAS GROUPS Worsening poverty as a result of mitigation Tropical forests globally and farmland in Sub- Groups with limited land rights and socially strategies Saharan Africa used for biofuels excluded (indigenous groups, women, smallholders without formal tenure) Migration Migration as a means for securing livelihoods Coastal cities and fertile deltas likely to People with few or no land holdings are more in the face of slow-onset climatic stress attract more migrants likely to migrate Small islands and coastal plains likely to see Men are more likely to migrate but this out-migration at higher levels of sea rise depends on local social norms and labor (e.g., Caribbean and Mediterranean Coast market opportunities; women left behind will countries) face additional work burdens Maghreb countries serving as receiving and transit countries for Sahelian and other Sub- Saharan African migrants Russian Arctic likely to experience flooding, subsidence and emigration, as well as some in-migration to exploit emerging farming and extractive opportunities Displacement (as a result of extreme events) Areas prone and vulnerable to hazards Elderly and poorest are less likely to leave, and when they leave, they are at greater risk of permanent displacement Displaced women sometimes find it more difficult to generate a livelihood (discriminatory labor markets) Health Increase in malaria Highland areas People who lack immunity MENA and Colombia (LAC) Children in poverty affected areas Migrants Low-income groups Increase in dengue fever Tropical cities Low-income groups Increase in water-borne diseases (diarrheal Tropical cities Children in poverty affected areas disease and cholera) Coastal populations Elderly populations Low-income groups Increase in respiratory diseases Globally Older people and children Women at risk from indoor air pollution Low-income groups Increase in food borne infectious diseases Globally Low-income groups Older people Children Strong Evidence Moderate Evidence Limited Evidence 28 TH E GLOBA L P ICTU RE POTENTIAL IMPACTS OF INTERACTING CLIMATE CHANGE EVIDENCE AND BROADER DEVELOPMENT BASE AND EXAMPLES OF AFFECTED REGIONS, EXAMPLES OF AFFECTED SOCIAL TRENDS CONFIDENCE SECTORS, AND AREAS GROUPS Reduction in availability of clean water and Coastal zones with low lying populations Low-income groups sanitation (e.g., Caribbean) Women and children (increased workloads and Cities reliant on highland or on declining higher violence risks) ground water sources (e.g., the Andes, parts Poor children are especially vulnerable to of Central Asia) disease Increase in heat-related illnesses and Globally Elderly reduced labor productivity Densely populated large cities Manual laborers and those working outdoors MENA, the Arabian peninsula (more exposed to heat stress) Overweight people Displaced people living in shelters Low-income groups Residents of urban heat islands Increased mortality rates from extreme Low elevated coastal zones and land prone Women and girls at increased risk (if social weather events and disasters to flooding and landslides in all three regions norms prevent them from acquiring survival skills) Men/ older boys (if expected to risk their lives to rescue others) Children and older people Low income households Increase in mental illnesses Globally (areas exposed to extreme events or Low-income groups affected by slow-onset change) Displaced people Increasing malnutrition Sub-Saharan Africa, South Asia, Central Children (especially infants) America, and MENA Subsistence farmers (in low rainfall areas) Urban poor Women (particularly in South Asia) Potential for increased risk of domestic and Globally (areas exposed to extreme events or Women and children sexual violence affected by slow-onset change) Conflict and Security Risk of land and water scarcity (or excess of Countries already affected by conflict (e.g., Land holders water) contributing to conflict/tensions North Africa and Sub-Saharan Africa) Farmers/ Subsistence Farmers Countries where there are tensions between Farmers vs Herders the mining industry and farmers/indigenous Indigenous groups groups (e.g., Peruvian Andes) Low-lying areas Extreme weather events or sudden disasters More common where governance is weak or Low-income people and children leading to conflict/social unrest visibly inequitable Protests related to increased food or fuel More common where governance is weak or Low-income urban groups prices visibly inequitable Increased risk of conflict through climate/ Countries where resources are scarce/or Low-income groups extreme event-induced displacement physically vulnerable to climate change with People who lack political recognition inequalities along ethnic/regional lines Strong Evidence Moderate Evidence Limited Evidence 29 Chapter 3 Latin America and the Caribbean The Latin America and the Caribbean region encompasses a huge diversity of landscapes and ecosystems. The region is highly heterogeneous in terms of economic development and social and indigenous history. It is also one of the most urbanized regions in the world. In Latin America and the Caribbean, temperature and precipitation changes, heat extremes, and the melting of glaciers will have adverse effects on agricultural productivity, hydrological regimes, and biodiversity. In Brazil, without additional adaptation, crop yields could decrease by 30–70 percent for soybean and up to 50 percent for wheat at 2°C warming. Ocean acidification, sea level rise, and more intense tropical cyclones will affect coastal livelihoods and food and water security, particularly in the Caribbean. Local food security is also seriously threatened by the projected decrease in fishery catch potential. Reductions and shifts in water availability would be particularly severe for Andean cities. The Amazon rainforest may be at risk of large‐scale forest degradation that contributes to increasing atmospheric carbon dioxide concentration and local and regional hydrological changes. 3.1 Regional Summary of the region are mainly rain-fed and, as a result, susceptible to variable rainfall and temperatures. In the Andean regions, houses The Latin America and Caribbean region is highly heterogeneous built on the often steep terrain are critically exposed to storm in terms of economic development and social and indigenous surface flows, glacial lake outbursts, and landslides. Coastal history with a population of 588 million (2013), of which almost residents, particularly in the Caribbean region, face the risks of 80 percent is urban. The current GDP is estimated at $5.655 loss of ecosystem services and livelihoods from degrading marine trillion (2013) with a per capita GNI of $9,314 in 2013. In 2012, ecosystems, loss of physical protection from degrading reefs, and approximately 25 percent of the population was living in poverty coastal flooding, as well as from damages to critical infrastructure and 12 percent in extreme poverty, representing a clear decrease (especially in the beach front tourism sector) and threats to fresh- compared to earlier years. Undernourishment in the region, for water from sea water intrusion due to sea level rise. example, declined from 14.6 percent in 1990 to 8.3 percent in 2012. Despite considerable economic and social development progress 3.1.1 Regional Patterns of Climate Change in past decades, income inequality in the region remains high. The region is highly susceptible to tropical cyclones and strong 3.1.1.1 Temperatures and Heat Extremes El Niño events, as well as to rising sea levels, melting Andean By 2100, summer temperatures over the region will increase by glaciers, rising temperatures and changing rainfall patterns. The approximately 1.5°C under the low-emissions scenario (a 2°C rural poor who depend on a natural resource base are particularly world) and by about 5.5o C under the high-emissions scenario vulnerable to climate impacts on subsistence agriculture and (a 4°C world) compared to the 1951–1980 baseline (Figure 3.1). ecosystem services; the urban poor living along coasts, in flood Along the Atlantic coast of Brazil, Uruguay, and Argentina, the plains, and on steep slopes are particularly vulnerable to extreme warming is projected to be less than the global average, ranging precipitation events and the health impacts of heat extremes. The between 0.5–1.5°C in a 2°C world and 2–4°C in a 4°C world. intensive grain-producing cropping systems in the southern part In the central South American region of Paraguay, in northern 31 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 3.1: Multi-model mean temperature anomaly for Latin America and the Caribbean for RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) for the austral summer months (DJF). Temperature anomalies in degrees Celsius are averaged over the time period 2071–2099 relative to 1951–1980. Argentina, and in southern Bolivia, warming is likely to be more under climate change, most dry regions will get drier and most pronounced, up to 2.5°C in a 2°C world and up to 6°C in a 4°C wet regions will get wetter. The exception is central Brazil. The world by 2071–2099. Similar levels of warming are projected for annual mean precipitation here is projected to drop by 20 percent the equatorial region, including eastern Colombia and southern in a 4°C world by the end of the century. In general, more intense Venezuela. Projections indicate that in a 4°C world almost all and frequent extreme precipitation events also become more likely. land area (approximately 90 percent) will be affected by highly In a 4°C world, the Amazon basin, the full land area of Brazil unusual,28 and more than half of the land area (approximately except the southern coast, southern Chile, the Caribbean, Central 70 percent) by unprecedented, summer heat extremes. America, and northern Mexico, are expected to be under severe to extreme drought conditions relative to the present climate by the end 3.1.1.2 Precipitation, Drought, and Aridity of the 21st century. The total area of land classified as hyper-arid, arid, In general, in a 2°C world, precipitation changes are relatively small or semi-arid is projected to grow from about 33 percent in 1951–1980 (+/–10 percent) and models exhibit substantial disagreement on to 36 percent in a 2°C world, and to 41 percent in a 4°C world. the direction of change over most land regions. In a 4°C world, the models converge in their projections over most regions, but 3.1.1.3 Tropical Cyclones inter-model uncertainty remains over some areas (such as north- Observations over the last 20–30 years show positive trends in ern Argentina and Paraguay) (Figure 3.2). Tropical countries on tropical cyclone frequency and strength over the North Atlantic the Pacific coast (Peru, Ecuador, and Colombia) are projected to but not over the eastern North Pacific. While Atlantic tropical see an increase in annual mean precipitation of about 30 percent. cyclones are suppressed by the El Niño phase of ENSO, they are Similarly, Uruguay on the Atlantic coast (and bordering regions enhanced in the eastern North Pacific. Under further anthropogenic in Brazil and Argentina) will get wetter. Regions which are pro- climate change, the frequency of high-intensity tropical cyclones jected to become drier include Patagonia (southern Argentina and is generally projected to increase over the western North Atlantic Chile), Mexico, and central Brazil. These patterns indicate that, by 40 percent for 1.5–2.5°C global warming and by 80 percent in a 4°C world. Global warming of around 3°C is associated with 28 In this report, highly unusual heat extremes refer to 3-sigma events and unprec- an average 10 percent increase in rainfall intensity averaged over edented heat extremes to 5-sigma events (see Appendix). a 200 km radius from a tropical cyclone’s center. Although there 32 LATI N AME R I CA A ND THE CA RIBBEA N Figure 3.2: Multi-model mean of the percentage change in the aridity index under RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) for Latin America and the Caribbean by 2071–2099 relative to 1951–1980. Hatched areas indicate uncertain results, with two or more out of five models disagreeing on the direction of change. Note that a negative change cor- responds to a shift to more arid conditions.29 is some evidence from multiple-model studies for a projected to experience above-average sea-level rise (Recife: median estimate: increase in frequency of tropical cyclones along the Pacific coast 0.63 m, low estimate: 0.41 m, high estimate: 1.14 m; Rio de Janeiro: of Central America, overall projections in this region are currently median estimate: 0.62 m, low estimate: 0.46 m, high estimate: inconclusive. Despite these inconclusive projections, however, any 1.11 m). Sea-level rise is exacerbated at low latitudes due to both increase in Pacific and Atlantic storms (not necessarily cyclones) increased ocean heat uptake and the gravity-induced pattern of ice making landfall simultaneously would potentially entail more sheets and glaciers. As an example, Guayaquil on the Pacific Coast damaging impacts than increasing frequency of any individual of Ecuador is projected to experience 0.62 m (low estimate: 0.46 m, Pacific or Atlantic cyclone.29 high estimate: 1.04 m) of sea-level rise in a 4°C world. In contrast, Puerto Williams (Chile) at the southern tip of the South American 3.1.2 Regional Sea-Level Rise continent is projected to experience only 0.46 m (low estimate: 0.38 m; high estimate: 0.65 m). Port-Au-Prince (Haiti) is projected Sea-level rise is projected to be higher at the Atlantic coast than at the to experience 0.61 m (low estimate: 0.41 m, high estimate: 1.04 m) Pacific coast. Valparaiso (median estimate: 0.55 m for a 4°C world) of sea-level rise in a 4°C world (Figure 3.11); it serves as a typical is projected to benefit from southeasterly trade wind intensification example for sea-level rise in other Caribbean islands. over the Southern Pacific and associated upwelling of cold water leading to below-average thermosteric (due to ocean temperature 3.1.3 Sector-based and Thematic Impacts rise) sea-level rise. In contrast, the Atlantic coast of Brazil is projected 3.1.3.1 Glaciers and Snowpack Changes 29 Some individual grid cells have noticeably different values than their direct neigh- Glacial recession in South America has been significant. The bors (e.g., on the border between Peru and Bolivia). This is due to the fact that the tropical glaciers in the Central Andes in particular have lost major Aridity index is defined as a fraction of total annual precipitation divided by potential portions of their volume in the course of the 20th century. A clear evapotranspiration (see Appendix). It therefore behaves in a strongly non-linear way, and thus year-to-year fluctuations can be large. Since averages are calculated over a trend of glacial retreat is also visible for glaciers in the southern relatively small number of model simulations, this can result in these local jumps. Andes, which have lost about 20 percent of their volume. 33 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL The recession of the tropical glaciers in the Central Andes will 3.1.3.3 Climate Change Impacts on Agriculture, continue as rapidly as it has in recent decades. Even for low or Livestock, and Food Security intermediate emissions scenarios inducing a global warming of The results of the climate change impact projections on crop yields 2–3°C above pre-industrial levels, two comprehensive studies con- differ among studies, but most authors agree that climate change sistently project a glacial volume loss of 78–97 percent. Both studies will very likely decrease agricultural yields of important food predict an almost complete deglaciation (93–100 percent) for a 4°C crops in the Latin America and Caribbean region. An exception world. Other studies are slightly less dramatic; irrespective of the is the projected increase in yield of irrigated/flooded rice in some temperature evolution in the next decades, however, large parts of regions. The few available studies on climate change impacts on the glaciers of the tropical Andes will be gone long before the end of livestock indicate that beef and dairy cattle production will decline the century. In the Southern Andes, the model spread for the 2–3°C under increasing temperatures, as heat stress is a major influenc- global warming ranges from 22–59 percent glacier volume loss; a ing factor of cattle productivity. Sheep seem to cope better with comparison for individual scenarios is difficult. In a 4°C world, warmer and drier conditions than cattle and pigs. models project a glacier volume retreat of 44–74 percent by 2100. Monitoring of snow cover in the high altitudes of Chile and 3.1.3.4 Climate Change Impacts on Biodiversity Argentina since 1950 shows no significant trend (possible trends Climate change-induced negative effects on biodiversity, from are hard to identify in the records, since the inter-annual variabil- range contractions to extinctions, are very likely in a warmer than ity is large and clearly modulated by ENSO). The lack of reliable 2°C world. As the adaptive capacity of affected species and eco- projections for snowpack and snow cover changes in the Andes systems is hard to project or quantify, models need to use simpli- is an important research gap. fied approaches as implemented in bioclimatic envelope models, species-distribution models, and dynamic global vegetation models. 3.1.3.2 Water Resources, Water Security, and Floods One clear trend regarding future warming levels is that the Although the magnitude of the change varies, there is a high more temperature is projected to increase, the more species diver- agreement on decreasing mean annual runoff and discharge in sity is affected. Mountainous regions in the tropics (e.g., cloud Central America. Water stress may increase, especially in arid forests) are projected to become very vulnerable due to the high areas with high population densities and during the dry season. number of endemic and highly specialized species which might In the Caribbean, runoff projections are of low confidence due to face mountaintop extinction. Most models do not take biotic lack of data. However, freshwater availability may decrease for interactions (e.g., food-web interactions, species competition) or several reasons, such as sea-level rise leading to an intrusion of resource limitations into account. Therefore, the realized ecological sea water into coastal aquifers. Regionally, the risk of flooding niche of species within an ecosystem might become much smaller and mudslides with high mortality rates is high. Although floods than what is potentially possible according to climatic and other often seem to be associated with land-use change, more severe environmental conditions, leading to shifts in ecological zones. flooding events may also occur in the context of climate change. Higher variability of seasonal discharge is projected for the 3.1.3.5 Amazon Rainforest Degradation, Dieback, Tropical Andes. Decreased streamflow during the dry season has and Tipping Point already been observed, and may decrease further as a result of Overall, the most recent studies suggest that the Amazon dieback is an ongoing glacier retreat. However, streamflow during the wet sea- unlikely, but possible, future for the Amazon region. Projected future son may increase. The Andean region could experience a higher precipitation and the effects of CO2 fertilization on tropical tree growth flood risk in a 4°C world (e.g. due to accelerated glacier melting). remain the processes with the highest uncertainty. Climate-driven In the Amazon Basin, runoff and discharge projections for most changes in dry season length and recurrence of extreme drought parts of the Amazon basin are diverging. For the western part of years, as well as the impact of fires on forest degradation, add to the basin a likely increase in streamflow, runoff, flood zone, and the list of unknowns for which combined effects still remain to be inundation time are projected. In southern most South America, investigated in an integrative study across the Amazon. A critical a decrease in mean runoff is projected. tipping point has been identified at around 40 percent deforesta- Although the Latin America and Caribbean region has an tion, when altered water and energy feedbacks between remaining abundance of freshwater resources, many cities depend on local tropical forest and climate may lead to a decrease in precipitation. rivers, aquifers, lakes, and glaciers that may be affected by climate A basin-wide Amazon forest dieback caused by feedbacks change—and freshwater supplies might not be enough to meet between climate and the global carbon cycle is a potential tipping demand. For example, Guadalajara (Mexico) and many Andean point of high impact if regional temperatures increase by more cities are expected to face increasing water stress and, if the current than 4°C and global mean temperatures increase by more than demand continues, low-income groups who already lack adequate 3°C toward the end of the 21st century. Recent analyses have, how- access to water will face more challenges. ever, downgraded this probability from 21 percent to 0.24 percent 34 LATI N AME R I CA A ND THE CA RIBBEA N for the 4°C regional warming level when coupled carbon-cycle including droughts, floods, landslides, and tropical cyclones; all climate models are adjusted to better represent the inter-annual of these extreme events can induce migration. variability of tropical temperatures and related CO2 emissions. Examples indicate that drought-induced migration is already This holds true, however, only when the CO2 fertilization effect is occurring in some regions. The largest level of climate migration realized as implemented in current vegetation models. Moreover, is likely to occur in areas where non-environmental factors (e.g., large-scale forest degradation as a result of increasing drought may poor governance, political persecution, population pressures, and impair ecosystem services and functions, including the regional poverty) are already present and putting migratory pressures on hydrological cycle, even without a forest dieback. local populations. The region is considered to be at low risk of armed conflict. 3.1.3.6 Fisheries and Coral Reefs However, in the context of high social and economic inequality Together with ocean acidification and hypoxia, which are very likely and migration flows across countries, disputes regarding access to to become more pronounced under high-emissions scenarios, the resources, land, and wealth are persistent. Climate change could possibility of more extreme El Niño events poses substantial risks increase the risk of conflict in the region through more resource to the world’s richest fishery grounds. Irrespective of single events, scarcity, more migration, increasing instability, and increasing the gradual warming of ocean waters has been observed and is frequency and intensity of natural disasters. further expected to affect fisheries (particularly at a local scale). Generally, fish populations are migrating poleward toward 3.1.3.9 Coastal Infrastructure colder waters. Projections indicate an increase in catch potential By 2050, coastal flooding with a sea-level rise of 20 cm could of up to 100 percent in the south of Latin America. Off the coast of generate approximately $940 million of mean annual losses in Uruguay, the southern tip of Baja California, and southern Brazil the 22 largest coastal cities in the Latin America and Caribbean the maximum catch potential is projected to decrease by more than region, and about $1.2 billion with a sea-level rise of 40 cm. The 50 percent. Caribbean waters and parts of the Atlantic coast of Central Caribbean region is particularly vulnerable to climate change due America may see declines in the range of 5–50 percent. Along the to its low-lying areas and the population’s dependence on coastal coasts of Peru and Chile, fish catches are projected to decrease by and marine economic activity. In a scenario leading to a 4°C world up to 30°percent, but there are increases expected toward the south. and featuring 0.89–1.4 m of sea-level rise, tropical cyclones in Irrespective of the sensitivity threshold chosen, and irrespective the Caribbean alone could generate an extra $22 billion by 2050 of the emissions scenario, by the year 2040, Caribbean coral reefs (and $46 billion by 2100) in storm and infrastructure damages and are expected to experience annual bleaching events. While some tourism losses, compared to a scenario leading to a 2°C world. species and particular locations appear to be more resilient to such The potential increase in tropical cyclone intensity may increase events, it is clear that the marine ecosystems of the Caribbean port downtime for ships and, therefore, increase shipping costs. are facing large-scale changes with far-reaching consequences for Beach tourism is particularly exposed to direct and indirect climate associated livelihood activities as well as for the coastal protection change stressors, including sea-level rise, modified tropical storm provided by healthy coral reefs. patterns, heightened storm surges, and coastal erosion. Coastal 3.1.3.7 Health tourist resorts are potentially two-to-three times more exposed The Latin America and Caribbean region faces increased risks of to climate change-related stressors than inland touristic resorts. morbidity and mortality caused by infectious diseases and extreme 3.1.3.10 Energy weather events. Observed patterns of disease transmission associ- The assessment of the current literature on climate change impacts ated with different parts of the ENSO cycle offer clues as to how on energy in Latin America and the Caribbean shows that there changes in temperature and precipitation might affect the incidence are only a few studies, most of which make strong assumptions of a particular disease in a particular location. Projections of how about such key issues as seasonality of water supply for hydro- malaria incidence in the region could be affected by climate change power. These studies are more qualitative than quantitative and over the rest of the century are somewhat inconsistent, with some important gaps remain. There is also a lack of studies with respect studies pointing to increased incidence and others to decreased to the impacts of climate change impacts on renewable energies. incidence. Such uncertainty also characterizes studies of the rela- In general, the impacts of climate change on energy demand tionship between climate change and malaria globally and reflects are less well studied than those on energy supply—and, yet, the complexity of the environmental factors influencing the disease. demand and supply interact in a dynamic way. For example, the 3.1.3.8 Migration and Security concomitant increase in energy demand during heat extremes While migration is not a new phenomenon in the region, it is and the decrease in energy supply through reduced river flow and expected to accelerate under climate change. There are many areas low efficiencies may put existing energy systems under increasing in the Latin America and Caribbean Region prone to extreme events, pressure in the future. 35 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL 3.1.4 Overview of Regional Development Narratives The development narratives build on the climate change impacts ana- MEXICO DOMINICAN HAITI REPUBLIC ST. KITTS AND NEVIS JAMAICA lyzed in this report (see Table 3.15: Synthesis table of climate change BELIZE ANTIGUA AND BARBUDA DOMINICA GUATEMALA HONDURAS ST. VINCENT AND THE GRENADINES impacts in LAC under different warming levels) and are presented ST. LUCIA EL SALVADOR NICARAGUA BARBADOS GRENADA in more detail in Section 3.5. Climate change impacts have manifold COSTA RICA R.B. DE TRINIDAD AND TOBAGO PANAMA VENEZUELA direct and indirect implications for development in the region. These GUYANA COLOMBIA SURINAME impacts occur on a continuum from rural to urban; not only are there many climate impacts directly affecting rural spaces leading for ECUADOR example to reduced agricultural productivity or altered hydrological regimes, but these impacts also affect urban areas through changing PERU BRAZIL ecosystem services, migration flows, and so forth. Development will likewise be impacted as the challenges of a changing climate mount BOLIVIA and interact with socioeconomic factors. In particular, glacial melt and PARAGUAY changing river flows, extreme events, and risks to food production systems will put human livelihoods under pressure. Climate change impacts are and will continue to affect devel- CHILE ARGENTINA URUGUAY opment across the region in several ways. First, changes to the hydrological cycle endanger the stability of freshwater supplies and ecosystem services. An altered hydrological system due to changing runoff, glacial melt, and snowpack changes will affect the ecosystem IBRD 41281 OCTOBER 2014 Falkland Islands services that the rural population depends on, freshwater supplies in This map was produced by the Map Design Unit of The World Bank. (Islas Malvinas) The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of The World Bank A DISPUTE CONCERNING SOVEREIGNTY OVER THE cities, and such major economic activities as mining and hydropower. Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. ISLANDS EXISTS BETWEEN ARGENTINA WHICH CLAIMS THIS SOVEREIGNTY AND THE U.K. WHICH ADMINISTERS THE ISLANDS. Second, climate change places at risk both large-scale agricultural production for export and small-scale agriculture for regional food Grenada, Guatemala, Guyana, Haiti, Honduras, Jamaica, Mexico, production. Third, a stronger prevalence of extreme events affects Nicaragua, Panama, Paraguay, Peru, Puerto Rico, St Kitts and Nevis, both rural and urban communities, particularly in coastal regions. St Lucia, St Vincent and the Grenadines, Suriname, Trinidad and At the sub-regional level, the following climate-development Tobago, Uruguay, and Venezuela. interactions are particularly important. In Central America and The region is very large and has a very diverse range of distrib- the Caribbean, extreme events threaten livelihoods and damage uted ecosystems from the Andean mountains that stretch for about infrastructure. In the Andes, changes in water resource availability 8,850 kilometers, to mountain glaciers, vast rainforests, savannas, challenge the rural and urban poor. In the Amazon, the risks of grasslands, wetlands, islands, deserts and a coastline that is over a tipping point, forest degradation, and biodiversity loss threaten 72,000 kilometers long. There are broad differences in development local communities. Hydrological changes may affect the wider levels both within and among countries (Table 3.1); these factors region. The Southern Cone faces risks to export commodities from influence the social vulnerability of the population. In addition, cur- loss of production from intensive agriculture. In the Mexican dry rent and projected climate change impacts vary strongly within the subtropical regions and northeastern Brazil, increasing drought region, with some key impacts relating to changing temperatures and stress threatens rural livelihoods and health. precipitation. Changes in extreme events (e.g., heatwaves, droughts, 3.2 Introduction tropical cyclones, and changing ENSO patterns) (see Section 2.3.2, El-Niño/Southern Oscillation) and sea-level rise are also projected to This report defines Latin America and the Caribbean (LAC) as vary across the region. These physical risk factors trigger biophysical the region encompassing the South American continent, Central impacts on hydrological flows, agricultural productivity, biodiversity America,30 the Caribbean islands, and Mexico. It is constituted in general, and forest dynamics in the Amazon in particular, coral by the following countries: Antigua and Barbuda, Argentina, the reefs, and fisheries, as well as social impacts on human health, Bahamas, Barbados, Belize, Bolivia, Brazil, Chile, Colombia, Costa security, energy systems, and coastal infrastructure. Rica, Cuba, Dominica, Dominican Republic, Ecuador, El Salvador, This report analyses these physical, biophysical, and social impacts in an integrated way using data analysis, model projec- 30 30 The World Bank Central America subregion includes the following countries: Costa tions, and an intensive review of the scientific literature. Wherever Rica, El Salvador, Guatemala, Honduras, Nicaragua, and Panama. possible, the results are regionally stratified. 36 Table 3.1: Basic Socioeconomic Indicators of LAC Countries URBAN LIFE URBAN POPULATION AGRICULTURE, EXPECTANCY INDICATOR POPULATION POPULATION GROWTH GDP PER CAPITA VALUE ADDED1 AT BIRTH2 % OF CURRENT 1000 UNIT MILLION POPULATION ANNUAL % US$ % OF GDP YEARS YEAR 2012 2012 2012 2012 2011 2011 SP.URB.TOTL ID SP.POP.TOTL .IN.ZS SP.URB.GROW NY.GDP.PCAP.CD NV.AGR.TOTL.ZS SP.DYN.LE00.IN Argentina 41.1 92.6 1.03 11.6 10.7 75.8 Antigua and Barbuda 0.1 29.9 1.01 12.7 2.5 75.5 Bahamas, The 0.4 84.4 1.75 21.9 2.3 74.8 Belize 0.3 44.6 2.01 – 13.1 73.5 Bolivia 10.5 67.2 2.27 2.6 12.5 66.6 Brazil 198.7 84.9 1.19 11.3 5.5 73.3 Barbados 0.3 44.9 1.65 14.9 1.5 75.0 Chile 17.5 89.3 1.13 15.5 3.7 79.3 Colombia 47.7 75.6 1.68 7.7 6.9 73.6 Costa Rica 4.8 65.1 2.12 9.4 6.5 79.5 Cuba 11.3 75.2 –0.07 0.0 5.0 78.9 Dominica 0.1 67.3 0.57 6.7 13.5 – Dominican Republic 10.3 70.2 2.07 5.7 6.0 73.0 Ecuador 15.5 68.0 2.43 5.4 10.4 75.9 Grenada 0.1 39.5 1.25 7.3 5.3 72.5 Guatemala 15.1 50.2 3.43 3.3 11.8 71.3 Guyana 0.8 28.5 0.88 3.6 21.3 65.9 Honduras 7.9 52.7 3.12 2.3 15.3 73.2 Haiti 10.2 54.6 3.85 0.8 – 62.3 Jamaica 2.7 52.2 0.36 5.4 6.6 73.1 St. Kitts and Nevis 0.1 32.1 1.41 14.3 1.8 – St. Lucia 0.2 17.0 –3.03 6.8 3.3 74.6 Mexico 120.8 78.4 1.60 9.7 3.4 76.9 Nicaragua 6.0 57.9 1.98 1.8 19.7 74.1 Panama 3.8 75.8 2.42 9.5 4.1 77.2 Peru 30.0 77.6 1.68 6.8 7.0 74.2 Puerto Rico 3.7 99.0 –0.64 27.7 0.7 78.4 Paraguay 6.7 62.4 2.58 3.8 21.4 72.1 El Salvador 6.3 65.2 1.40 3.8 12.5 71.9 Suriname 0.5 70.1 1.47 9.4 10.0 70.6 Trinidad and Tobago 1.3 14.0 2.26 17.4 0.5 69.7 Uruguay 3.4 92.6 0.45 14.7 9.4 76.8 St. Vincent and the 0.1 49.7 0.80 6.5 6.4 72.3 Grenadines Venezuela, RB 30.0 93.7 1.73 12.7 – 74.3 Note: Agriculture corresponds to ISIC divisions 1–5 and includes forestry, hunting, and fishing, as well as cultivation of crops and livestock production. Value added is the net 1 output of a sector after adding up all outputs and subtracting intermediate inputs. It is calculated without making deductions for depreciation of fabricated assets or depletion and degradation of natural resources. 2 Life expectancy at birth indicates the number of years a newborn infant would live if prevailing patterns of mortality at the time of its birth were to stay the same throughout its life. Source: World Bank (2013b). 37 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL 3.2.1 Social, Economic and Demographic in densely populated low-income settlements (Ravallion et al. 2008) Profile of the Latin America and Caribbean are more likely to be adversely affected by climate extremes. Region 3.2.2 Vulnerabilities to Climate Change in the The Latin America and Caribbean Region comprises a population Latin America and the Caribbean Region of 588 million (in 2013) of which almost 80 percent is urban. The current GDP is estimated at US$ 5.655 trillion (in 2013) with a GNI In the LAC region, climate change is expected to accentuate pre- per capita of US$ 9,314 in 2013. In 2012, approximately 28.2 per- existing socioeconomic vulnerabilities. People living in low-lying cent of the population were living in poverty and 11.3 percent coastal areas, slums (Douglas et al. 2008), and certain popula- in extreme poverty or deprivation (ECLAC 2014). These figures tion groups (such as poor (Ahmed et al. 2009; Hertel et al. 2010) represent a decrease of about 1.4 percent in the poverty rate with and women-led households (Kumar and Quisumbing 2011)), are respect to 2011 (ECLAC 2014). Although the number of people particularly exposed to shocks and future climate change risks. living in poverty in the region has been going down slowly, in Several socioeconomic and physical factors can contribute to absolute terms, this means that 164 million people were poor—of increasing the vulnerability of populations to climate change. For whom 66 million were extremely poor (ECLAC 2014). example, poverty hinders households’ adaptive capacity (Kelly and Despite progress in the past decade and a growing middle Adger 2000). According to Calvo (2013), the following character- class now surpassing the number of poor, inequality in the region istics of the population in the LAC region increase the exposure remains high, and may be stagnating. Thirty eight percent of to climate change impacts and the likelihood of being affected by the population live just above the poverty line on an income of economic shocks: (1) one third of the population can be classified $4–10 per day (Fereira et al. 2013). as poor or extremely poor, so that any shock can push them into Income inequalities in the region affect vulnerability to cli- further poverty; (2) there are more children in poor and extremely mate change, as the poor are more likely to be exposed to both poor households given the higher fertility rates amongst the poor, climate and economic shocks and limited ability to prepare for or so that shocks can have particularly adverse consequences for mitigate impacts. Haldén (2007) notes that disparities and divi- children, given that they are at a stage of life with greater needs sions could impede growth and undermine adaptation strategies, and dependency; and (3) poor households have members with with the additional risk that substantial inequality might also fewer years of formal education, which can limit their capacity destabilize societies and increase the likelihood of conflict in the to adapt to climate change impacts or macroeconomic shocks. light of climate change and variability (see Section 2.5, Social Vulnerability to Climate Change). These income inequalities are 3.2.3 Vulnerabilities Faced by Rural Populations further exacerbated by gender, spatial, and ethnic inequalities. Ethnicity correlates closely with poverty. In seven countries for Even though urban and rural areas have both experienced poverty which data are available, the poverty rate is 1.2–3.4 times higher reduction, the gap between the rural and urban experience is still for indigenous and afrodescendent groups than for the rest of the wide. In 2010, the rural poverty rate was twice as high as that population (ECLAC and UNFPA 2009). Furthermore, while indig- of urban areas; when considering extreme poverty, it was four enous peoples in LAC represent 10 percent of the population, their times as high (IFAD 2013). Close to 60 percent of the population income levels and human development indicators (e.g., education in extreme poverty lives in rural areas (RIMISP 2011). Many rural and health status) have consistently fallen behind those of the people in the region continue to live on less than $2 per day and rest of the population (Hall and Patrinos 2005). have poor access to financial services, markets, training, and The region’s population is expected to rise to 622 million in other opportunities. The rural poor are thus more likely to feel the 2015 and to 700 million by 2030. The distribution of the population impacts of climate change and variability given their dependency on is increasingly urban. In 2010, the urban population accounted for small-hold, rain-fed agriculture and other environmental resources 78.8 percent of the total; this number is projected to rise to 83.4 per- that are particularly susceptible to the effects of climate change cent by 2030 (ECLAC 2014). The concentration of poverty in urban (Hoffman and Grigera 2013). Moreover, these populations have settlements is a central determinant of vulnerability to climate limited political influence and are less able to leverage govern- change. In addition, differences in fertility levels of social groups in ment support to help curb the effects of climate change (Hardoy Latin America and the Caribbean show that the poor segments of the and Pandiella 2009). The dependence of the rural population on urban population contribute most to urban growth, exacerbating the land as a source of food and income, coupled with lack of physi- contribution of predominantly poor rural-urban migrants. Residents cal and financial adaptive capacity, means that poor farmers are 38 LATI N AME R I CA A ND THE CA RIBBEA N also at increased risk of harm from slow-onset change (Rossing and Rubin 2011). Box 3.1: Hurricane Mitch’s Impact in Urban Areas 3.2.4 Urban Settlements and Marginalized In 1998, Hurricane Mitch cut a swath across Central America, hitting Populations Honduras especially hard. Overall, 30 percent of the central district of Honduras, including the cities of Tegucigalpa, was destroyed. Most of In addition to high levels of urbanization, in many countries in the damage was concentrated around the four rivers that cross the the region a high proportion of the urban population lives in a cities; as a result, 78 percent of Tegucigalpa’s drinking water supply few very large cities. National economies, employment patterns, pipelines were destroyed. Factors that increased the vulnerability of and government capacities—many of which are highly centralized— the city included obsolete and inadequate infrastructure, especially are also very dependent on these large cities. This makes them regarding water, sanitation, and drainage; a lack of zoning codes; extremely vulnerable (Hardoy and Pandiella 2009). Based on two concentration of services and infrastructure in only a few areas; a lack global model studies, Ahmed et al. (2009), Hertel et al. (2010), and of official prevention and mitigation strategies; and inappropriate man- Skoufias et al. (2011) estimate that urban salaried workers will be agement of the river basins. Source: Hardoy and Pandiella (2009). the most affected by climate change given the increase in prices of food resulting from reduced agricultural production. Increasing pressure on rural economic activities induced by droughts, heat waves, or floods—also driven by future climate change impacts— ill-suited to settlement, such as areas prone to flooding or affected could result in a greater rural exodus and add further pressure on by seasonal storms, sea surges, and other weather-related risks. human and economic development in cities (Marengo et al. 2012, Such land is cheap or is state-owned land and relatively easy for 2013; Vörösmarty et al. 2002). low-income groups to occupy. In most cases, the poor have no Spatial vulnerability within urban centers is a major source of formal tenure of the land and face not only environmental risks risk. There are particularly hazardous areas within Latin American but also the risk of eviction. Left with few options, low-income cities where settlements have been built, including flood plains groups live in overcrowded houses in neighborhoods with high (Calvo 2013). These settlements already face infrastructural population densities (Hardoy and Pandiella 2009). All these factors problems that affect water supplies, sanitation, and solid waste contribute to a high level of vulnerability to floods and landslides. management as they were built for less populated cities. This leaves In most LAC cities there are concentrations of low-income these areas at greater risk of flooding and other disasters (Hardoy households at high risk from extreme weather (Hardoy et al. 2001). and Pandiella 2009) (Box 3.1). In 2004, for example, 14 percent of For example, an estimated 1.1 million people live in the favelas of the population in the LAC region (more than 125 million people) Rio de Janeiro that stretch over the slopes of the Tijuca mountain did not have access to improved sanitation, and an even-higher range (Hardoy and Pandiella 2009). Most low-income groups live percentage lacked good quality sanitation and drainage. Limited in housing without air-conditioning or adequate insulation; dur- access to sanitation and freshwater sources is also a key source of ing heat waves, the very young, the elderly, and people in poor vulnerability as this increases the risk of the spread of water-borne health are particularly at risk (Bartlett 2008; see also Section 3.4.7, diseases (McMichael and Lindgren 2011; McMichael et al. 2012). Human Health). In northern Mexico, for example, heat waves have Houses in informal settlements are built incrementally with been correlated with increases in mortality rates; in Buenos Aires, deficient materials and no attention to building or zoning regula- 10 percent of summer deaths are associated with heat strain; and tions. As a result, a significant share of the population is exposed records show increases in the incidence of diarrhea in Peru (Mata to flooding, contamination of groundwater by salt water, and and Nobre 2006). constraints on the availability and quality of drinking water, as Although LAC has an abundance of freshwater resources, well as to a rising sea level (Magrin et al. 2007). In addition, the many cities depend on local rivers, underground water, lakes, and impacts of extreme weather events are more severe in areas that glaciers that may be affected by climate change (see Section 3.4.1, have been previously affected and have not yet been able to recover Glacial Retreat and Snowpack Changes and Section 3.4.2, Water properly, with cumulative effects that are difficult to overcome. Resources, Water Security, and Floods). Considering city growth, Limited disaster preparation and a lack of planning compound environmental deterioration, and possible climate change impacts, the problems (Martí 2006). the supply of fresh water might not be enough to meet demand. A great deal of urban expansion in the region has taken Guadalajara in Mexico (Von Bertrab and Wester 2005) and many place over floodplains, on mountain slopes, and in other zones Andean cities may face increasing water stress and, if the current 39 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL situation continues, low-income groups who already lack adequate and Krellenberg 2011). A similar situation exists in Sao Paulo, access to water will be even less likely to obtain it. Quito is likely where the local catchment of Alto Tiete provides just 10 percent to face water shortages as a result of glacier retreat (Hardoy and of the water supply for 11 million people and where urban areas Pandiella 2009). In Santiago de Chile, an estimated 40 percent are expanding over agricultural and natural areas; this is impact- reduction in precipitation would impact water supply in a city that ing the area’s storm-water retention capacity, thereby making the is expecting a 30 percent population growth by 2030 (Heinrichs city more prone to flood events (Heinrichs and Krellenberg 2011). 3.2.4.1 Gender and Age-specific Vulnerabilities In the context of a male-dominated, patriarchal society, gender and Box 3.2: The Case of Mexico City age are important aspects of vulnerability in the LAC region. Many women and children are particularly vulnerable to the effects of Mexico City provides a good case study for examining the potential climate change as they have limited access to resources and fewer future impacts of climate change on urban areas in the region. capabilities and opportunities for participating in decision and The main climate-related risk factors in Mexico City stem from policy making (Hardoy and Pandiella 2009). The most vulnerable increased dry-spell periods and heat waves (see Section 3.3.2, groups seldom have an influential voice with regard to disaster Heat Extremes). Greater Mexico City, with about 20 million inhabitants, is among preparedness or response, and their needs receive little attention. the most highly populated cities on the planet. Despite a very high Economic dependency places women and children in a GDP per capita, the city exhibits a very large income inequality, particularly disadvantaged situation, and climate change could with about 13 percent of the population lacking enough money to exacerbate the problem. According to ECLAC and UNFPA (2009), meet minimum food needs and approximately 23 percent unable to poverty is 1.7 times higher among minors under 15 than in adults, access education or affordable health care (Ibarrarán 2011). and 1.15 times greater among women than men. For example, in According to UN-Habitat, Mexico City is already exposed Uruguay poverty is 3.1 times higher among children than adults; to several environmental challenges: The urban region is rapidly in Chile, it is 1.8 times greater; and, in Nicaragua, it is 1.3 times expanding, increasing the demand for space, infrastructure, water, greater. Sudden-onset disasters or a worsening of drought condi- and energy. This taxes already-deficient water supplies and an inad- tions have the potential to trigger severe acute malnutrition with equate sewage system. Waste management is similarly challenging, greater effects on women and children (see Section 2.5, Social as collection, transportation, and adequate final disposal are limited Vulnerability to Climate Change). compared to the daily volume of waste produced by the 20 million inhabitants of the city. There are various ways in which women can be affected by In addition to these preexisting socioeconomic and environmen- climate change differently than men. One way is through the rise tal vulnerabilities, climate change stressors are projected to increase in domestic violence in the context of environmental disasters. Mexico City’s overall vulnerability. Four principal climate-related Gender-based violence is already a significant problem for women stressors are projected to affect Mexico City: (1) the higher fre- in LAC, where most studies estimate that prevalent physical vio- quency of heat waves and hotter days; (2) the decreased instance of lence between intimate partners affects between 20–50 percent of cooler days; (3) the increased occurrence of flash floods; and (4) the women. While there are important differences in the estimates, extension of summer droughts (Ibarrarán 2011). An increase in the studies similarly find that 8–26 percent of women and girls report frequency of heat waves could have two potential consequences having been sexually abused (Morrison et al. 2004). Moser and in Mexico City. First, the steadily growing elderly population will be Rogers (2005) indicate that rapid socioeconomic changes—such particularly exposed as they are more sensitive to heat extremes as those that can occur as a result of climate shocks—might have than the rest of the population (Gasparrini and Armstrong 2011). destabilizing effects within families, leading to an increased risk Second, in response to heat waves, the population could purchase more air conditioning and cooling systems. This may put power of domestic violence. Although evidence of the effects of climate- plants under severe stress, particularly as they work less efficiently induced disasters in the region remains mixed and limited, accounts under higher temperatures. The extension of the summer droughts of gender-based violence have been found in Nicaragua after Hur- is projected to increase Mexico City’s water stress situation (Novelo ricane Mitch, in the Dominican Republic after Tropical Storm Noel, and Tapia 2011; Romero Lankao 2010). Furthermore, the increased and in Guatemala after Tropical Storm Agatha (Bizzarri 2012). occurrence and extension of summer droughts may disproportion- Some groups of indigenous women are also particularly ately impact the rural population, who may then be more inclined to vulnerable given their involvement in specific activities. Within migrate to cities to find less climate-dependent economic activities indigenous populations in the Colombian Amazon, for example, (Ibarrarán 2011). As a consequence, the population of Mexico’s impacts on horticulture would affect mainly women as they are urban areas is expected to grow, putting more pressure on the traditionally in charge of this activity (Kronik and Verner 2010). urban environments and resources. However, not all gender differences are necessarily worse for women and children. For example, in the case of indigenous 40 LATI N AME R I CA A ND THE CA RIBBEA N groups, impacts in the availability of fish and game will affect urbanization, indigenous populations in LAC are found both in mainly young men (Kronik and Verner 2010). urban and rural areas (Del Popolo and Oyarce 2005). Popolo et Changes in migration patterns as a result of climate change are al. (2009) found that on average 40 percent of the indigenous also likely to have important effects on women. While traditionally population in 11 countries in Latin America were living in urban it is young males who have migrated domestically or abroad, over areas in 2000/2001. Whereas the ratio varies from country to the past two decades rural indigenous women have also started country, recent census information highlights that 21.4 percent of migrating, generally with the support of their social and family indigenous peoples in Colombia (Paz 2012), 54 percent in Bolivia networks. Studies indicate that the experiences of migrant indig- (Molina Barrios et al. 2005), 55.8 percent in Peru (Ribotta 2011), enous women tend to be less favorable, however, as they become 64.8 percent in Chile, (Ribotta 2012a), and 82 percent in Argentina vulnerable and disadvantaged by discrimination, lack of previous (Ribotta 2012b) are currently living in towns and cities. crosscultural experiences, illiteracy, and language barriers. Because Understanding the climate change impacts on indigenous of these barriers, their only option for work tends to be low-wage populations requires an understanding of the cultural dimension employment in the informal sector (Andersen et al. 2010). of their livelihood strategies and the social institutions that sup- port them (Kronik and Verner 2010). In rural areas, indigenous 3.2.4.2 Indigenous People groups are particularly vulnerable to climate change because of There are about 40 million indigenous people within the LAC their reliance on natural resources, traditional knowledge systems, region, with the majority located in the cooler high regions of and culture (Kronik and Verner 2010) and due to poor access to the Andes and in Mesoamerica (Kronik and Verner 2010). The infrastructure and technology (Feldt 2011). Indigenous popula- indigenous population is made up of about 400 indigenous tions with greater territorial autonomy, and with their livelihoods groups (Del Popolo and Oyarce 2005), of which about 30 percent more intertwined with forest and water resources, are therefore are afrodescendant (Rangel 2006). Bolivia is the country with more affected by climate change when compared to indigenous the highest share of indigenous people (66 percent) and Mexico populations with restricted territorial autonomy (whose livelihoods has the highest absolute number (Table 3.2). When compared to are more diversified, and include wage labor, tourism, and other non-indigenous groups, the profile of indigenous peoples shows income-generating activities) (Kronik and Verner 2010). Indig- that they have higher levels of poverty and infant and maternal enous groups with territorial autonomy are normally located in mortality, lower levels of life expectancy, income, and schooling, the Amazon region whereas those without are more likely to be and less access to water and sanitation; together, this highlights found in the Andes (Kronik and Verner 2010). the exclusion of and discrimination against these groups (Del Kronik and Verner (2010) studied the impacts of climate change Popolo and Oyarce 2005; World Bank 2014). As a response to on indigenous populations in the LAC region, in particular on those Table 3.2: Total Population and Indigenous Population Census 2000 COUNTRY AND TOTAL INDIGENOUS % INDIGENOUS CENSUS YEAR POPULATION POPULATION POPULATION RECOGNIZED PEOPLES GROUPS Bolivia (2001) 8,090,732 5,358,107 66.2 36 groups (49.5% Quechua, 40% Aymara) Brazil (2000) 169,872,856 734,127 0.4 241 groups Costa Rica (2000) 3,810,179 65,548 1.7 Chile (2002) 15,116,435 692,192 4.6 9 groups (83% Mapuche) Ecuador (2001) 12,156,608 830,418 6.8 Guatemala (2002) 11,237,196 4,433,218 39.5 21 groups (all Maya) Honduras (2001) 6,076,885 440,313 7.2 Mexico (2000) 97,014,867 7,618,990 7.9 62 groups Panama (2000) 2,839,177 285,231 10.0 3 groups (Ngöbe-Buglé, Kuna, and Embera-Wounan) Paraguay (2002) 5,183,074 87,568 1.7 Please note that data are from 2000 but used here to provide a comprehensive overview of the share of indigenous population. Sources: Del Popolo and Oyarce (2005); Rangel (2006). 41 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL living in the Amazon, the Andes, the Caribbean, and in Central vulnerability of indigenous groups living in urban areas is more America. In the Colombian Amazon, they found the biggest direct related to the social conditions they face (e.g., discrimination and impacts related to changes in the seasonal cycle (i.e., floods, social exclusion) than linked to their livelihoods. and dry and rainy periods): river flooding affects fish and turtle reproduction, thereby impacting the food security of indigenous 3.2.4.3 Risk for Populations in Coastal Areas populations; changes in periods during which important local Coastal communities in the region are particularly exposed to fruits ripen and the succession of dry and rainy seasons affect the climate change extremes and sea-level rise (Trab Nielsen 2010). harvests of wild fruits; and changes in the length of the dry season The region’s 64,000 km coastline is one of the most densely affects agriculture productivity, particularly in alluvial plateaus gar- populated in the world (Sale et al. 2008). Coastal states have more dens. Increases in temperature and changes in precipitation affect than 521 million residents, of whom two-thirds (348 million) live mainly horticulture, favoring specific crops such as cassava but within 200 km of the coastline. More than 8.4 million people in threatening harvest diversity (Echeverri 2009). Climate change and LAC live in the path of hurricanes, and roughly 29 million live in climate variability is also apparent in the ‘decoupling’ of ecological low-elevation coastal zones where they are highly vulnerable to markers from seasonal changes (whereby seasons appear to occur sea-level rise, storm surges, and coastal flooding (McGranahan et al. significantly late or early), affecting livelihood decision-making in 2007; UNEP 2007). For example, several countries have a large particular for indigenous peoples. This may affect the credibility section of their urban population living in areas where elevation of elders and traditional leaders, as their authority to predict the is below five meters above sea level (CIESIN 2011). In Belize, the natural seasonality is challenged (Kronik and Verner 2010). Bahamas, Antigua and Barbuda, and Suriname, between 15 and In the indigenous Andes, rising temperatures can increase 62 percent of the urban population live below five meters above demand for water. At the same time, higher evapotranspiration sea level (Table 3.3). This low elevation significantly increases the rates and glacial retreat are expected to reduce the water supply; urban population’s exposure to sea-level rise, storm surges, and restricting pasture land availability in the dry season, and poten- modified tropical storm patterns. tially provoking conflict over land use (Kronik and Verner 2010). Furthermore, human activities such as overfishing, marine In the Bolivian Altiplano, however, Aymara communities have pollution, and coastal development have eroded the ecosystems declared that high-elevation zones have now become productive, in many coastal areas to a level where they no longer provide as changing climatic conditions have turned the area into arable buffers to climate extremes. Climate change and variability are land (Kronik and Verner 2010). likely to compound the damage to ecosystems and to human Indigenous populations in the Andes are not only subjected to settlements, directly through more intense and frequent storms biophysical vulnerability. In the rural Andes, social marginalization and sea-level rise and indirectly through the further degradation and social determinants that limit the ability to improve terms of of the ecosystems (Trab Nielsen 2010). labor, education and access to technical assistance, undermine Coastal communities at greatest risk from climate change the adaptive capacity of the indigenous population (McDowell and variability are generally those that rely on natural resources and Hess 2012). In Palca (Bolivia), for example, farmers are not for a living, occupy marginal lands, and have limited access to only vulnerable due to the retreat of the Mururata glacier and the the livelihood assets that are necessary for building resilience to resulting impact on water supply but also to historical margin- climate change. They include communities that rely on coastal alization due to the lack of official identity cards, land titles, or tourism and on fisheries. They also include much of the region’s access to bilingual (Aymara-Spanish) basic education (McDowell large population of urban slum dwellers (Trab Nielsen 2010). and Hess 2012). More than 50 percent of the Caribbean population lives along In the Caribbean and Central America, an increase in the the coastline, and around 70 percent live in coastal cities (Mimura frequency of some natural disasters (e.g., hurricanes) could limit et al. 2007; UNEP 2008). Many economic activities (e.g., tourism) the access of indigenous populations to key crop, forest, and fish are also concentrated in coastal areas (UNEP 2008). Pressures resources (Kronik and Verner 2010); slow onset changes, meanwhile, arise on the islands over limited land resources as people are could decrease the productivity of traditional varieties of maize, dependent on these natural resources for economic development generating pressure to switch to more commercial varieties (Kronik and their livelihoods. The GDP of the region is generated mainly and Verner 2010). Given that rural areas are mainly populated by from two sectors—tourism and agriculture. Both are highly vul- indigenous groups—especially those that are most remote—means nerable to climate-induced hazards, including flooding, sea-level that they are the most likely to be affected. This is exacerbated by rise, storms, and coastal erosion (Karmalkar et al. 2013). Small the strong dependence of indigenous groups on natural resources islands are especially vulnerable to extreme events (UNEP 2008). as well as by their reliance on traditional farming techniques. The east coast of Mexico and Central America, and the In contrast to the situation of rural indigenous populations, the Caribbean, are strongly affected by wind storms and cyclones 42 LATI N AME R I CA A ND THE CA RIBBEA N Table 3.3: Percentage of Latin American and Caribbean Population Living in Urban Areas and Below Five Meters of Elevation. URBAN POPULATION PERCENTAGE PERCENTAGE OF PERCENTAGE OF URBAN IN PERCENTAGE OF OF POPULATION LAND AREA BELOW POPULATION LIVING TOTAL POPULATION LIVING IN INFORMAL 5 METERS OF BELOW 5 METERS OF COUNTRIES (IN 2012) SETTLEMENTS (2005) ELEVATION ELEVATION (2010) Caribbean Countries Antigua and Barbuda 29.87 47.9 10.30 15.50 Bahamas, The 84.45 – 1.61 23.55 Barbados 44.91 – 0.92 0.92 Belize 44.59 47.3 0.56 17.36 Cuba 75.17 – 0.38 2.66 Dominica 67.30 – 1.39 3.05 Dominican Republic 70.21 17.6 0.20 0.90 Grenada 39.49 59.0 1.77 1.92 Haiti 54.64 70.1 0.20 2.44 Jamaica 52.16 60.5 2.05 3.08 St. Kitts and Nevis 32.11 – 9.25 9.46 St. Lucia 16.97 11.9 0.76 0.84 St. Vincent and the Grenadines 49.70 – 0.00 0.00 Trinidad and Tobago 13.98 24.7 1.68 2.85 Latin American Countries Argentina 92.64 26.2 0.07 3.29 Bolivia 67.22 50.4 0.00 0.00 Brazil 84.87 28.9 0.06 3.04 Chile 89.35 9.0 0.02 0.65 Colombia 75.57 17.9 0.09 1.35 Costa Rica 65.10 10.9 0.08 0.26 Ecuador 67.98 21.5 0.29 4.68 El Salvador 65.25 28.9 0.10 0.11 Guatemala 50.24 42.9 0.02 0.04 Guyana 28.49 33.7 0.22 11.81 Honduras 52.73 34.9 0.05 0.49 Nicaragua 57.86 45.5 0.03 0.31 Panama 75.78 23.0 0.13 1.90 Paraguay 62.44 17.6 0.00 0.00 Peru 77.58 36.1 0.02 0.81 Suriname 70.12 38.9 0.27 62.04 Uruguay 92.64 – 0.14 3.65 Venezuela, RB 93.70 32.0 0.16 2.63 Mexico 78.39 14.4 0.15 1.30 Source: Data from CIESIN (2011); UN-HABITAT (2013); and World Bank (2013b). 43 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL (Maynard-Ford et al. 2008). Coastal areas are prone to storm surge of Paraguay, northern Argentina, and southern Bolivia will see floods and sea-level rise (Woodruff et al. 2013). Floods and hur- more pronounced warming, up to 2.5°C in a 2°C world and up to ricanes present both a high risk of death (Dilley et al. 2005) and a 6°C in a 4°C world by 2071–2099. Similar levels of warming are threat to regional development and economic stability in Central projected for eastern Colombia and southern Venezuela. America and the Caribbean (Mimura et al. 2007). The normalized warming (that is, the warming expressed in terms of the local year-to-year natural variability—see Appendix) is 3.3 Regional Patterns of Climate Change plotted in the lower panels of Figure 3.4. The normalized warming indicates how unusual the projected warming is compared to the 3.3.1 Projected Temperature Changes natural fluctuations a particular region has experienced in the past, here the period 1951–1980 (Coumou and Robinson 2013; Hansen Figure 3.3 shows projected austral summer (December, January, Feb- et al. 2012; Mora and Frazier et al. 2013). The tropics will see the ruary—or DJF) temperatures for the LAC land area (see Appendix). strongest increase in normalized monthly summer temperatures By 2100, summer temperatures over the LAC land area will increase since historic year-to-year fluctuations are relatively small. In the by ~1.5°C under the low-emissions scenario (a 2°C world) and by eastern part of the equatorial region between 15°S and 15°N, monthly ~5.5°C under the high-emissions scenario (a 4°C world) compared temperatures will shift by 3–4 standard deviations in a 2°C world to the 1951–1980 baseline. This is about 0.5°C less than the projected and by 6–7 standard deviations in a 4°C world. A shift of 3–4 stan- global mean land warming which is typical for the Southern Hemi- dard deviations implies that an average monthly temperature in the sphere (see Figure 2.5 in World Bank 2013). in a 2°C world, warming future will be as warm as the most extreme monthly temperatures of 1.5°C (multi-model mean) is reached by mid-century. Summer experienced today (i.e., events in the tail of the current distribution). temperatures will continue to increase beyond mid-century under the A shift twice as large (i.e., 6–7 standard deviations) implies that high-emissions scenario, causing the multi-model mean warming for extremely cold summer months by the end of the 21st century will the 2071–2099 period to be about 4.5°C (Figure 3.3. and Figure 3.4). be warmer than the warmest months today. Thus, in a 4°C world, The regional maps (Figure 3.5) show rather uniform patterns monthly summer temperatures in tropical South America will move of summer warming, with regions in the interior of the continent to a new climatic regime by the end of the century. Subtropical generally projected to see a somewhat stronger temperature increase. regions in the south (northern Argentina) and the north (Mexico) Along the Atlantic coast of Brazil, Uruguay, and Argentina, the are expected to see a much less pronounced shift. Nevertheless, warming remains limited, with about 0.5–1.5°C in a 2°C world a shift by at least 1-sigma (in a 2°C world) or 2-sigma (in a 4°C and 2–4°C in a 4°C world. The central South American region world) is projected to occur here over the 21st century. 3.3.2 Heat Extremes Figure 3.3: Temperature projections for the Latin American Figure 3.5 and Figure 3.6 show a strong increase in the frequency and Caribbean land area compared to the 1951–1980 of austral summer months (DJF) warmer than 3-sigma and 5-sigma baseline for the multi-model mean (thick line) and individual (see Appendix) over LAC by the end of the century (2071–2099). The models (thin lines) under RCP2.6 (2°C world) and RCP8.5 (4°C tropics, which are characterized by relatively small natural variability, world) for the months of DJF. will see the largest increase in such threshold-exceeding extremes. Especially along the tropical coasts, including Peru, Ecuador, and Colombia, summer month heat extremes will become much more frequent, consistent with the large shift in the normalized temperature distribution here (see Figure 3.5). The 5-sigma events, which are absent under present-day climate conditions, will emerge in these countries even in a 2°C world, and are projected to occur in roughly 20 percent of summer months. At the same time, 3-sigma events, which are extremely rare today, will become the new norm (i.e., this threshold will be exceeded in roughly half of the summer months during 2071–2099). In a 4°C world, almost all summer months will be warmer than 3-sigma and, in fact, most will be warmer than 5-sigma as well (70 percent). Thus, under this scenario, the climate in tropical South America will have shifted to a new hot regime. Compared to the tropics, the subtropical regions in the north The multi-model mean has been smoothed to give the climatological trend. (Mexico) and south (Uruguay, Argentina, and southern Chile) 44 LATI N AME R I CA A ND THE CA RIBBEA N Figure 3.4: Multi-model mean temperature anomaly for Latin America and the Caribbean for RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) for the austral summer months (DJF). Temperature anomalies in degrees Celsius (top row) are averaged over the time period 2071–2099 relative to 1951–1980, and normalized by the local standard deviation (bottom row). are projected to see a more moderate increase in the frequency warmer than 3-sigma by 2071–2099 (i.e., this will have become of threshold exceeding extremes. In fact, in a 2°C world, 5-sigma the new norm). Furthermore, 5-sigma events will also emerge events will remain absent and 3-sigma events will still be rare (less and occur typically in about 20 percent of summer months over than 10 percent of summer months). In a 4°C world, however, a subtropical regions. substantial increase in frequency is projected. In most subtropi- The strong increase in frequency of summer months warmer cal regions, at least half of all summer months are expected to be than 3- and 5-sigma in the tropics, as reported here, is quantitatively 45 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 3.5: Multi-model mean of the percentage of austral summer months (DJF) in the time period 2071–2099 with temperatures greater than 3-sigma (top row) and 5-sigma (bottom row) for scenario RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) over Latin America and the Caribbean. consistent with published results from analyses using the full The duration of warm spells is projected to increase most in the CMIP5 dataset of climate projections (Coumou and Robinson 2013; tropics—already by 60–90 days in a 2°C world and by 250–300 days Sillmann et al. 2013a; b). In addition, minimum night-time and in a 4°C world. In the tropics, temperatures experienced during the maximum day-time temperatures in the summer are projected to 10 percent warmest summer nights of the 1961–1990 period will increase by 1–2°C in a 2°C world and by 5–6°C in a 4°C world occur most nights (50–70 percent) in a 2°C world and almost all (Sillmann et al. 2013b), in good agreement with the projected nights (90–100 percent) in a 4°C world by the end of the century seasonal mean temperatures in Figure 3.3. (Sillmann et al. 2013b). 46 LATI N AME R I CA A ND THE CA RIBBEA N Figure 3.6: Multi-model mean and individual models of exhibit substantial disagreement on the direction of change over the percentage of Latin American and Caribbean land area most land regions. With a more pronounced climatic signal (i.e., a warmer than 3-sigma (top) and 5-sigma (bottom) during 4°C world, RCP8.5), the models converge in their projections over austral summer months (DJF) for scenarios RCP2.6 (2°C most regions, but inter-model uncertainty remains over some areas world) and RCP8.5 (4°C world). (hatched shading in the maps). Nevertheless, a well-defined pat- tern of change in annual precipitation can be extracted for defined sub regions. For example, tropical countries on the Pacific coast (Peru, Ecuador, and Colombia) are projected to see an increase in annual mean precipitation of about 30 percent. This enhanced rainfall occurs year-round and can be detected in both the austral winter and summer seasons. Similarly, Uruguay on the Atlantic coast (and bordering regions in Brazil and Argentina) are projected to get wetter. Again, this increase in annual rainfall is year-round, though it is most pronounced during the summer (DJF). Regions which are projected to become drier include Patagonia (southern Argentina and Chile), Mexico, and central Brazil. These patterns indicate that, under climate change, most dry regions may get drier and most wet regions may get wetter in the future (but see Greve et al. 2014 for a discussion of this concept for past climate). The exception is central Brazil (i.e. the region from 0–20°S and 50–65°W), which contains the southeastern part of the Amazon rainforest. The annual mean precipitation here is projected to drop by 20 percent in a 4°C world by the end of the century. This drop in annual rainfall is entirely due to a strong and robust decrease in winter (JJA) precipitation (–50 percent), with essentially no change in summer (DJF) precipitation. In fact, this reduction in winter precipitation appears already in a 2°C world. These projected changes in annual and seasonal temperatures generally agree well with those provided by the IPCC AR5 based on the full set of CMIP5 climate models (Collins et al. 2013). However, there is one important difference in that the full set of CMIP5 models shows significant JJA Amazon drying over northern Changes in heat extremes in subtropical regions are less dramatic Brazil only. Over central Brazil, the multi-model mean of the full but nevertheless pronounced. In the Southern Hemisphere subtrop- set of CMIP5 models projects drying, as also seen in Figure 3.7, ics, the length of warm spells increases by roughly 0–15 days (2°C but the magnitude of change is small. Instead, significant drying world) or 30–90 days (4°C world). In the Northern Hemisphere over the full Amazon region primarily occurs during austral spring subtropics (Mexico) these values roughly double, but they are still (September-October-November). less than the increase in the tropics (Sillmann et al. 2013b). Night- time temperatures experienced during the 10 percent warmest austral 3.3.4 Extreme Precipitation and Droughts summer nights in 1961–1990 will occur in roughly 30 percent of nights (2°C world) and 65 percent of nights (4°C world). Analysis of the observational record since the 1950s indicates a robust increase in overall precipitation and in intensity of extreme 3.3.3 Regional Precipitation Projections precipitation events for South America, particularly over southern South America and the Amazon region (Skansi et al. 2013). Long- Projected changes in annual and seasonal precipitation (see Appen- term trends in meteorological droughts are not statistically robust dix) are plotted in Figure 3.7 for the LAC region for 2071–2099 over 1950–2010. Over the recent decade, however, two severe relative to 1951–1980. Note that projections are given as percent- droughts (2005 and 2010) have affected the Amazon, likely con- age changes compared to the 1951–1980 climatology and thus, nected to an anomalous warm tropical North Atlantic (Marengo especially over dry regions, large relative changes do not neces- et al. 2011; Zeng et al. 2008). sarily reflect large absolute changes. In general, in a 2°C world Dai (2012) finds a statistical significant increase in drought these changes are relatively small (+/–10 percent) and models conditions for Central America and the Caribbean for the 1950–2010 47 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 3.7: Multi-model mean of the percentage change in austral summer (DJF, top), winter (JJA, middle) and annual (bottom) precipitation for RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) for Latin America and the Caribbean by 2071–2099 relative to 1951–1980. Hatched areas indicate uncertain results, with two or more out of five models disagreeing on the direction of change. 48 LATI N AME R I CA A ND THE CA RIBBEA N period, although the significance of this trend depends on the refer- except the southern coast, southern Chile, and Central America, ence period and the formulation of the underlying drought index and in particular northern Mexico, is expected to be under severe (Trenberth et al. 2014). Fu et al. (2013) report a significant increase to extreme drought conditions relative to the present climate by in the length of the dry season over southern Amazonia since 1979. the end of the 21st century under the RCP4.5. These results are Using an ensemble of CMIP5 models, Kharin et al. (2013) inves- confirmed by a multi-model impact analysis under a 4°C scenario tigated extreme precipitation events based on annual maximum that also reveals a strong increase in drought risk in the Caribbean, daily precipitation with 20-year return values. In a 4°C world, although uncertainties remain substantial (Prudhomme et al. 2013). these events are found to intensify by about 25 percent over LAC The increase in future drought risk in Central America and the with a large uncertainty range.31 In addition, the return time of a Caribbean is generally related to an extension and intensification 20-year extreme precipitation event from the 1985–2005 period of the so-called midsummer drought period (Rauscher et al. 2008). would reduce to about 6 years by the end of the 21st century While an overall reduction in precipitation during the dry season is (2081–2100) in a 4°C world (Kharin et al. 2013). robustly projected by regional and global models alike (Campbell These increases are not, however, homogeneous over the full et al. 2011; Karmalkar et al. 2013; Taylor et al. 2013), it is not clear continent. This is consistent with the variable seasonal precipita- if this will lead to an increase in meteorological drought conditions tion projections in Figure 3.7. While little-to-not statistically sig- (e.g., an increase in the number of consecutive dry days) (Hall nificant, an increase in frequency is projected for the Caribbean, et al. 2012). This illustrates the added value of a comprehensive Meso-America, Southern Argentina, and Chile, and hotspots with impact model analysis, as undertaken by Prudhomme et al. (2013), extreme precipitation increases of more than 30 percent are pro- that accounts for the full change in the regional water cycle in jected in the Serra do Espinhaco in Brazil, the Pampas region in investigating future drought risk. Argentina, and the Pacific coastline of Ecuador, Peru, and Colombia Changes over the Amazon basin and eastern Brazil are found (Kharin et al. 2013). The latter may be related to an increase in to be particularly pronounced during the dry season (from July to frequency of future extreme El Niño events (Cai et al. 2014; Power September), which amplifies the risk of large-scale forest degradation; et al. 2013). These regions are also found to show the strongest this contrasts with Central America, Venezuela and southern Chile, rise in compound maximum 5-day precipitation (which is most where the drought risk is projected to increase during the austral sum- relevant for flooding events) by the end of the 21st century in a 4°C mer (Prudhomme et al. 2013). Drought risks are found to be strongly world (Sillmann et al. 2013b). Increases in extreme precipitation scenario-dependent and to be much less pronounced in a 2°C world, in southern Brazil and northern Argentina are in line with results in particular for Meso-America and the Caribbean; a substantial risk from regional climate models (Marengo et al. 2009) and might remains for South America under this scenario (Prudhomme et al. 2013). be dominated by intensification of the South American monsoon In the Amazon region, climate change is not the only anthropo- system (Jones and Carvalho 2013). Projections of extreme precipi- genic interference expected over the decades to come; deforestation tation for Meso-America and the Caribbean discussed above do will be at least equally important. The link between large-scale not comprehensively account for the risk of extreme precipitation deforestation and reduced precipitation is well established (e.g., related to tropical cyclones (that are discussed in Section 3.3.6, Davidson et al. 2012; Medvigy et al. 2011; Runyan 2012), and Bagley Tropical Cyclones/Hurricanes). In a 2°world, changes in heavy et al. (2014) used a regional climate model to demonstrate that precipitation would be greatly reduced and barely significant over deforestation might have amplified the severe droughts over the last most parts of the continent. decade. However, none of the projections given above accounts for While an increase in extreme precipitation represents a poten- the possible adverse effects of deforestation and forest degradation on tial threat for some regions, increase in duration and intensity of the climate of the Amazon region, which, in the presence of possible droughts might represent the bigger threat over all of Latin America self-amplifying feedbacks between reduced forest cover and extreme and the Caribbean. An increase and intensification in meteorological droughts, represent a substantial risk of large-scale Amazon dieback droughts is projected for large parts of South and Central America (see Section 3.4.5, Amazon Rainforest Dieback and Tipping Point). in a 4°C world (Sillmann et al. 2013b), although large model 3.3.5 Aridity uncertainties remain in particular for Central America (Orlowsky and Seneviratne 2013). A more comprehensive analysis of future Apart from a reduction in precipitation, warming can also cause droughts accounting for the effects of runoff and evaporation as a region to shift toward more arid conditions as enhanced surface well as local soil and vegetation properties was undertaken by Dai temperatures trigger more evapotranspiration—thereby drying the (2012). He found that the Amazon basin, the full land area of Brazil soil. This long-term balance between water supply and demand is captured by the aridity index (AI), which is shown in Figure 3.9 for 31 The lower and upper limits of the central 50 percent inter-model range are 14 and the Latin American region. The AI is defined as the total annual 42 percent respectively. precipitation divided by the annual potential evapotranspiration 49 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 3.8: Multi-model mean of the percentage change in the annual-mean of monthly potential evapotranspiration for RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) for Latin America and the Caribbean by 2071–2099 relative to 1951–1980. Hatched areas indicate uncertainty regions with two or more out of five models disagreeing on the direction of change. (see Appendix); it fundamentally determines whether ecosystems model uncertainty in these regions and no robust statements can and agricultural systems are able to thrive in a certain area. A be made about whether conditions will become more or less arid. A decrease in the value of the AI thus indicates that water becomes prime reason for this is that both annual precipitation and potential more scarce (i.e., more arid conditions), with areas classified as evapotranspiration in these regions have upward trends, and it is hyper-arid, arid, semi-arid, and sub-humid as specified in Table 3.4. the relative magnitude of these trends which determines whether Potential evapotranspiration is a measure of the amount of water a region becomes more or less arid. In other words, it is unclear a representative crop type would need over a year to grow, (i.e. whether or not warming-driven drying outpaces the increase in a standardized measure of water demand; see Appendix). Under annual precipitation projected for these regions. most circumstances, changes in potential evaporation are governed Outside these uncertain regions, the LAC land area is projected to by changes in temperature. become more arid other than for an isolated region in the southern Changes in annual-mean potential evapotranspiration (Fig- tip of Chile. Again three major drying regions can be identified: ure 3.8) broadly follow those of absolute warming (Figure 3.4), (1) Northern Hemisphere subtropics (Mexico); (2) the interior of with regions in the continental interior generally experiencing the South American continent (southern Amazonia, Bolivia, and the strongest increase. Thus, though potential evapotranspiration Paraguay); and (3) central Chile and Patagonia. Over the first two depends on several meteorological variables, it seems primarily regions, the AI is projected to decrease by up to 20 percent in a 2°C driven by future temperature changes. The signal is weak in a 2°C world and up to 40 percent in a 4°C world. For the third region, world, with relative changes in potential evapotranspiration smaller the decrease in AI is especially pronounced in a 4°C world, drop- than 20 percent everywhere except for some isolated regions in the ping up to 60 percent (note that this region is already arid today). continental interior. In a 4°C world, changes become much more The shift in AI in Figure 3.9 causes some regions to be clas- pronounced, with countries like Paraguay and Bolivia projected to sified in a different aridity class (see Table 3.4). The total area of see an increase in potential evapotranspiration of up to 50 percent. land classified as either hyper-arid, arid, or semi-arid is projected Consistent with this result, these regions are also projected to see to grow from about 33 percent in 1951–1980 to 36 percent in a 2°C the strongest absolute warming (Figure 3.4). world (i.e., an increase of nearly 10 percent) and to 41 percent in Over almost all of the LAC land area, the multi-model mean a 4°C world (i.e., an increase of nearly 25 percent). projects more arid conditions under future climate change. Still, over extended areas, notably near the equator, at the Pacific tropical 3.3.6 Tropical Cyclones/Hurricanes coast (Peru), and at the sub-tropical Atlantic coast (southern Brazil, Uruguay, and northern Argentina), the AI changes little and mod- In Central America and the Caribbean, tropical cyclones occur els disagree on the direction of change. Thus, there is substantial regularly and have severe impacts, especially when making landfall; 50 LATI N AME R I CA A ND THE CA RIBBEA N Figure 3.9: Multi-model mean of the percentage change in the aridity index under RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) for Latin America and the Caribbean by 2071–2099 relative to 1951–1980. Hatched areas indicate uncertain results, with two or more out of five models disagreeing on the direction of change. Note that a negative change cor- responds to a shift to more arid conditions.32 Table 3.4: Multi-model mean of the percentage of land area gas concentrations and/or aerosol concentrations, increases the in Latin America and the Caribbean which is classified as amount of heat that is absorbed by the atmosphere and can increase hyper-arid, arid, semi-arid and sub-humid for 1951–1980 and the potential for tropical cyclones to form. Changes in the frequency 2071–2099 for both the low (2°C world, RCP2.6) and high (4°C and intensity of tropical storms are modulated, however, by other world, RCP8.5) emissions scenarios. factors, including vertical wind shear and humidity. Of particular importance is the vertical wind shear (i.e., the difference between 2071–2099 2071–2099 1951–1980 (RCP2.6) (RCP8.5) wind speeds near the surface and higher up in the troposphere). High wind shear disrupts the process of tropical cyclone forma- Hyper-Arid 8.6 10.1 12.8 tion and intensification, so that increases in wind shear counter Arid 10.3 11.2 12.7 increases in sea surface temperature—impacting tropical cyclone Semi-Arid 14.3 14.8 15.9 formation and intensity. El Niño events (see Section 2.3.2, El-Niño/ Sub-Humid 5.5 5.8 6.0 Southern Oscillation) tend to enhance wind shear over the Gulf of Mexico and the Caribbean Sea and thus suppress Atlantic tropical cyclones (Aiyyer and Thorncroft 2011; Arndt et al. 2010; Kim et al. 2011). On the other hand, El Niño events have been shown to the impacts on marine ecosystems, transport, and infrastructure increase tropical cyclone activity in the eastern North Pacific can also be severe. Strobl (2012), for instance, derived an average (Kim et al. 2011; Martinez-Sanchez and Cavazos 2014). Obser- 0.83 percent drop in economic output after tropical cyclones strikes vational evidence, however, suggests atmospheric patterns tend in this region, with big variations between countries.32 to steer tropical cyclones away from the Mexican coast during The energy of tropical cyclones is derived from the ocean surface El Niño years (and toward the coast in La Niña years), so that and lower atmosphere. Warming, along with increased greenhouse the net effect on the Pacific coastlines of the Americas remains unclear. In both regions, changes in the El-Niño/Southern- 32 Some individual grid cells have noticeably different values than their direct neighbors Oscillation (ENSO) due to climate change and the associated (e.g., on the border between Peru and Bolivia). This is due to the fact that the Aridity uncertainties affect tropical cyclone projections. In addition to index is defined as a fraction of total annual precipitation divided by potential evapo- such dynamic changes, thermodynamic processes alone can also transpiration (see Appendix). It therefore behaves in a strongly non-linear fashion, and thus year-to-year fluctuations can be large. Since averages are calculated over a work to suppress tropical cyclone formation and intensification relatively small number of model simulations, this can result in these local jumps. (Mallard et al. 2013). 51 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL These factors make projecting changes in tropical cyclone Short-term variability in tropical cyclones is large and GCM frequency and intensity difficult. The recent IPCC WGI AR5 report resolution too low to resolve high-intensity tropical cyclone struc- found in relation to observed changes that “there is low confidence tures. Projections therefore rely on proxies of tropical cyclone char- in attribution of changes in tropical cyclone activity to human influ- acteristics, or a cascade of low- to high-resolution models. Using ence owing to insufficient observational evidence, lack of physical CMIP5 models (50 percent uncertainty range across 17 GCMs), understanding of the links between anthropogenic drivers of cli- Villarini and Vecchi (2013) projected that the Power Dissipation mate and tropical cyclone activity, and the low level of agreement Index would increase by 100–150 percent in a 2°C world over the between studies as to the relative importance of internal variability, North Atlantic. A considerably larger increase and a much wider and anthropogenic and natural forcings” (Bindoff et al. 2013). range of about 125–275 percent were projected for a 4°C world. Observational records show little or no historical global trend Bender et al. (2010) used a variety of models to initialize a very in tropical cyclone frequency or intensity, in particular in light of high-resolution operational hurricane-prediction model, noting an uncertainties resulting from the potential undercounting of tropical increase of 80 percent in the frequency of the strongest category 4 cyclones in early parts of the record predating satellite observations and 5 Atlantic tropical cyclones in a 4°C world (compared to the (before about 1970). The North Atlantic, however, is an exception. present.) Knutson et al. (2013) also found an 80 percent increase Tropical cyclone frequency has increased in the North Atlantic in the strongest category tropical cyclones for the same scenario sharply over the past 20–30 years, but uncertainty is large over and class of models and around a 40 percent increase for a lower longer time-periods (Bindoff et al. 2013). Emanuel (2008) noted an emissions scenario and for the most recent generation of global increase in the Power Dissipation Index (a combination of frequency models included in IPCC AR5 (2013b) at roughly 1.5–2.5°C warm- and intensity) of North Atlantic tropical cyclones over a 1949–2004 ing (early and late 21st century RCP4.5). The largest increase (see observational period. Using a new record of observations, Kossin Figure 3.10) occurred in the Western Atlantic, north of 20°N (i.e., et al. (2013) showed a strong and statistically significant increase to the north of Haiti), a pattern confirmed by Emanuel (2013). in lifetime maximum intensity of tropical cyclones over the North Using a very different statistical downscaling method, Grinsted et al. Atlantic of 8 m.s–1 per decade, over the period 1979–2010, particu- (2013) projected a twofold to sevenfold increase in the frequency larly for mid- to high-intensity storms. This can be compared to the of “Katrina magnitude events” regarding storm surge (not wind median lifetime maximum intensity of around 50 m.s–1 of tropical speed) for a 1°C rise in global temperature. For context: in terms cyclones across the region in this historical time series. Such observed of wind speeds, the 2005 Gulf of Mexico tropical cyclone Katrina changes were shown to be linked to both anthropogenic climate was a class 5 tropical cyclone. change and internal climate variability (Camargo et al. 2012; Villarini The eastern North Pacific is less well represented in the scien- and Vecchi 2013; Wang and Wu 2013). Differential warming of the tific literature. Based on variations of one high-resolution model, tropical Atlantic, with historically observed warming higher than Murakami et al. (2012, 2011) projected no significant trends for average for the tropics, tends to enhance tropical cyclone intensi- this region under future climate change. By contrast, Emanuel fication in this region (Knutson et al. 2013). No significant trends (2013), using an ensemble of 6 different CMIP5 models, projected have been observed over the Eastern North Pacific (Kossin et al. an increase in frequency of tropical cyclones along the Pacific coast 2013). In general however, tropical cyclones haven been observed of Central America (particularly large near the coast of southeast to migrate polewards (Kossin et al. 2014). Mexico); the author notes, however, that the method does not In the long term, model simulations from a range of models capture well the currently observed storm frequency in this region. lead to the expectation that tropical cyclone frequency will not With projected increased intensity and frequency of the most be affected much by continued global warming but that mean intense storms, and increased atmospheric moisture content, intensity, as well as the frequency of the most intense tropical Knutson et al. (2013) estimated an increase of 10 percent in the cyclones, are projected to increase (Knutson et al. 2010; Tory et al. rainfall intensity averaged over a 200 km radius from the tropical 2013). IPCC AR5 WGI found that: cyclone center for the Atlantic, and an increase of 20–30 percent for the tropical cyclone’s inner core, by the end of the 21st century “Projections for the 21st century indicate that it is likely for roughly 2.5–3.5°C global warming. This confirms the earlier that the global frequency of tropical cyclones will either results of Knutson et al. (2010) reported in IPCC AR5 WGI (2013b). decrease or remain essentially unchanged, concurrent This effect would greatly increase the risk of freshwater flooding with a likely increase in both global mean tropical cyclone from tropical cyclones making landfall. maximum wind speed and rain rates . . . The influence of The projections above focus on changes in frequency of tropi- future climate change on tropical cyclones is likely to vary cal cyclones and wind and rainfall intensity. Colbert et al. (2013) by region, but there is low confidence in region-specific projected a shift in tropical cyclone migration tracks in the tropi- projections. The frequency of the most intense storms cal North Atlantic, with more frequent ocean recurving tropical will more likely than not increase substantially in some cyclones and fewer cyclones moving straight westward toward land. basins.” (Stocker et al. 2013) 52 LATI N AME R I CA A ND THE CA RIBBEA N Figure 3.10: Change in average rate of occurrence of Category 4 and 5 tropical cyclones per hurricane season (August–October) at about 2.5°C warming globally above pre-industrial levels by the end of the 21st century compared to the present-day. Source: Knutson et al. (2013). © American Meteorological Society. Used with permission. Together with changes in genesis area, but assuming constant tropi- on the climate-driven factors of local heat uptake of the ocean, cal cyclone frequency, this lead to a projected increase in tropical ocean current changes, and the far-reaching influence of changing cyclones per season in the central Atlantic and a decrease in the gravity from the ice sheets. In LAC, regional sea-level rise largely Gulf of Mexico and the Caribbean. Murakami and Wang (2010), reflects the rise in global mean sea level. Still, consistent regional however, found no change in trajectories. Current literature does features exist across both the 1.5°C and 4°C world scenarios in not provide evidence for a (change in) risk of synchronized landfall both the median and high estimate (Figure 3.11, Table 3.5). These of different tropical cyclones (e.g., in Central America from both features are more pronounced for stronger overall sea-level rise. the Pacific and the Atlantic). While individual tropical cyclones may not be strong, their compound impact may be more severe. Moreover, any increase in trend of Pacific and Atlantic storms Table 3.5: Sea-level rise between 1986–2005 and 2081–2100 (not necessarily cyclones) making landfall simultaneously would for the RCP2.6 (1.5°C world) and RCP8.5 (4°C world) in selected locations of the LAC region (in meters). potentially entail more damaging impacts than increasing frequency of any individual Pacific or Atlantic cyclone alone. RCP2.6 (1.5°C WORLD) RCP8.5 (4°C WORLD) In summary, observations show historical positive trends in Acapulco 0.38 (0.23, 0.61) 0.6 (0.42, 1.01) tropical cyclone frequency and strength over the North Atlantic Antofagasta 0.37 (0.22, 0.58) 0.58 (0.42, 0.98) but not over the eastern North Pacific. While Atlantic tropical Barranquilla 0.39 (0.22, 0.65) 0.65 (0.43, 1.12) cyclones are suppressed by the El Niño phase of ENSO, they are enhanced in the eastern North Pacific. Under further anthropogenic Buenos Aires 0.34 (0.24, 0.52) 0.56 (0.45, 0.97) climate change, the frequency of high-intensity tropical cyclones Cristobal 0.39 (0.22, 0.65) 0.66 (0.44, 1.07) is generally projected to increase over the western North Atlantic Guayaquil 0.39 (0.25, 0.62) 0.62 (0.46, 1.04) by 40 percent for 1.5–2.5°C global warming and by 80 percent in Lima 0.38 (0.24, 0.61) 0.6 (0.45, 1.02) a 4°C world. Global warming around 3°C is associated with an Port-au-Prince 0.38 (0.21, 0.61) 0.61 (0.41, 1.04) average 10 percent increase in rainfall intensity averaged over a 200 km radius from the tropical cyclone center. Although there is Puerto Williams 0.27 (0.19, 0.37) 0.46 (0.38, 0.65) some evidence from multiple-model studies for a projected increase Recife 0.39 (0.23, 0.65) 0.63 (0.41, 1.14) in frequency of tropical cyclones along the Pacific coast of Central Rio de Janeiro 0.37 (0.24, 0.61) 0.62 (0.46, 1.11) America, overall projections in this region are currently inconclusive. Tumaco 0.38 (0.24, 0.6) 0.61 (0.44, 1.01) Valparaiso 0.35 (0.21, 0.54) 0.55 (0.41, 0.91) 3.3.7 Regional Sea-level Rise Numbers in parentheses indicate low and high bounds (see Section 6.2, Regional sea-level rise will vary in a large and geographically diverse Sea-Level Rise Projections for an explanation of the 1.5° world). region such as Latin America and the Caribbean, and will depend 53 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 3.11: Patterns of regional sea-level rise. Median (left column) and high (right column) estimates of projected regional sea-level rise for the RCP2.6 scenario (1.5°C world, top row) and the RCP8.5 scenario (4°C world, bottom row) for the period 2081–2100 relative to the reference period 1986–2005. Associated global mean rise is indicated in the panel titles. Representative cities are denoted by black dots and discussed in the text with numbers provided in Table 3.5. Sea-level rise is projected to be higher at the Atlantic coast estimate: 0.62 m for a 4°C world), where a large number of people than at the Pacific coast. Valparaiso is projected to benefit from are at risk (Nicoldi and Mueller Petermann 2010). this effect (median estimate: 0.55 m for a 4°C world), where Sea-level rise is enhanced at low latitudes due to both increased southeasterly trade wind intensification over the Southern Pacific ocean heat uptake in the region (Figure 3.12; middle) and the and associated upwelling of cold water (Merrifield and Maltrud gravity-induced pattern of ice sheets and glaciers (Figure 3.12 bot- 2011; Timmermann et al. 2010) lead to below-average thermoste- tom). As an example, Guayaquil on the Pacific Coast of Ecuador is ric sea-level rise (Figure 3.13). In contrast, Recife on the Atlantic projected to experience a sea-level rise of 0.62 m (median estimate; coast of Brazil is projected to experience above-average sea-level low estimate: 0.46 m; high estimate: 1.04 m) of sea-level rise in rise (median estimate: 0.63 m for a 4°C world). A similar rise a 4°C world (Figure 3.13). Guayaquil is among the most vulner- is projected for the Rio de Janeiro region (Figure 3.11) (median able coastal cities in terms of relative GDP losses (Hallegatte et al. 54 LATI N AME R I CA A ND THE CA RIBBEA N Figure 3.12: Regional anomaly pattern and its contributions in 2013). In contrast, Puerto Williams (Chile) at the southern tip the median RCP8.5 scenario (4°C world). of the South American continent is projected to experience only 0.46 m (median estimate for a a 4°C world) (low estimate: 0.38 m; high estimate: 0.65 m). The upper bound differs between the two locations since it is largely determined by the risk of high ice-sheet-driven sea-level rise. Port-Au-Prince (Haiti) is projected to experience 0.61 m (low estimate: 0.41 m, high estimate: 1.04 m) of sea-level rise in a 4°C world (Figure 3.13); it serves as a typical example for sea-level rise in other Caribbean islands. The sea-level rise at the conti- nental Caribbean Coast exceeds the projection for the Caribbean islands (Barranquilla, median estimate: 0.65; low estimate: 0.43; and high estimate: 1.12 in a 4°C world). The difference may be linked to a weakening of the Caribbean Current that is connected to the Atlantic meridional overturning circulation33 (Pardaens et al. 2011). The high upper bounds are due to the strong influence of the Antarctic ice sheet. The ocean is predicted to warm, with high rates in the Southern Ocean off Buenos Aires (Kuhlbrodt and Gregory 2012). The coastal waters are, however, dominated by the cold Malvinas Current, a branch of the Antarctic Circumpolar Current (ACC). Since minor warming is projected for the ACC and the Malvinas Current, the strong warming signal in the Southern Ocean does not lead to additional sea-level rise at the Rio de la Plata estuary—which is projected to rise slower than the global mean (Buenos Aires, median estimate: 0.56 m; low estimate: 0.42 m; and high estimate: 1.03 m in a 4°C world). Sea-level rise in the region will be influenced by the future strength of the Brazil and Malvinas Currents and the position of their confluence zone (Lumpkin and Garzoli 2011). 3.4 Regional Impacts 3.4.1 Glacial Retreat and Snowpack Changes 3.4.1.1 Topography of Glaciers in the Andes The Andes are the longest continental mountain range in the world, stretching about 7,000 km along the coast of South America. There are major ice masses in the Patagonian Andes and on Terra del Fuego in the Southern Andes. A much smaller amount of ice (about 217 Gt), covering an area of about 4,900 km2, is stored in the Cen- tral Andes; this region hosts more than 99 percent of the world’s glaciers that are located in tropical latitudes (Tropical Andes). These glaciers exist at high altitudes between 4,000 and 6,500 m above sea level and are of crucial importance for the livelihood of the local populations as they act as critical buffers against highly seasonal precipitation and provide water during the dry season for Total sea-level rise (top), steric-dynamic (middle), and land-ice (bottom) contributions to sea-level rise, shown as anomalies with respect to the global mean sea-level rise. Global mean contributions to be added on top of the spatial anomalies are indicated in the panel titles. 33 This ocean current system transports a substantial amount of heat energy from the tropics and Southern Hemisphere toward the North Atlantic, where the heat is then transferred to the atmosphere. Changes in this ocean circulation could have a profound impact on many aspects of the global climate system. 55 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 3.13: Sea level projections for selected cities. Time series for sea-level rise for the two scenarios, RCP2.6 (1.5°C world, blue) and RCP8.5 (4°C world, green). Median estimates are given as full thick lines and the lower and upper bound given as shading. Full thin lines are global median sea-level rise with dashed lines as lower and upper bound. Vertical and horizontal black lines indicate the reference period and reference (zero) level. 56 LATI N AME R I CA A ND THE CA RIBBEA N domestic, agricultural, and industrial use. In addition, the water Observed Glacier Recession and energy supplies (see Section 3.4.11) of the capital cities of As a general trend, the Andean glaciers are shrinking. This is caused Lima (Peru), La Paz (Bolivia), and Quito (Ecuador) depend on by increased melt rates, decreased accumulation, changes in the the glacial melt water (see Section 3.4.2, Water Resources, Water ice dynamics, and/or a combination of all these factors. For the Security, and Floods). Andean snowpack is also a crucial natural period 1980–2011, Giesen and Oerlemans (2013) calculated, for the water resource, particularly in the semi-arid regions of Southern tropical glaciers, a relative volume change of 7.3 percent (39 Gt) South America (Masiokas et al. 2012, 2013). with respect to a total glaciated area of 4,940 km2 (99 percent of which is located in the Central Andes). Over a much longer 3.4.1.2 Current Situation and Observed Changes historical period (1901–2009) and for a larger region (account- Characteristics of Tropical Glaciers ing for 82 percent of all tropical glaciers), Marzeion et al. (2012) Tropical glaciers are particularly threatened by climate change due estimated an area reduction of glaciers of about 79±2 percent to their high altitude, the high level of radiation, and the tropical (15,900±500 km2), which corresponds to a volumetric ice loss climate dynamics. Observations have shown that the variability of 90 percent (1,740 Gt). The Southern Andes contain a much of the surface temperature of the Pacific Ocean is the govern- larger ice mass of 11,430 Gt (1980) extending over a glaciated ing factor, explaining the dramatic glacier recession of the 20th area of 33,700 km2. This huge ice mass decreased in volume by century, although the precipitation trend has not been significant 6.1 percent (695 Gt) between 1980–2011 (Giesen and Oerlemans during that period. The impact of the ENSO phenomenon (see 2013). Over the 20th century (1901–2009), Marzeion et al. (2012) Section 2.3.2, El-Niño/Southern Oscillation) on the inter-annual infer a reduction of about 32 percent in area (15,500±200 km2), mass balance is consequently high in the tropical glacier zone, which can be associated with a volume loss of 22±5 percent with low temperatures, high precipitation, high wind speeds, high (1,340±290 Gt). albedo, and a nearly balanced or positive mass balance during La These global projections, however, rely on coarse resolution Niña events and a strongly negative mass balance during El Niño models that cannot adequately simulate glacier dynamics in the events (Chevallier et al. 2011). steep topography of the narrow mountain chain of the Andes. In particular, the strong recession rate of the comparably small tropical glaciers is likely overestimated by the global-scale methodology. Regional models, in contrast, can provide more comprehensive analyses with resolutions of up to a few hundred meters. Figure 3.14: Compilation of mean annual area loss rates for Rabatel et al. (2013) reviewed the various studies on the current different time periods for glaciated areas between Venezuela state of Tropical Andes’ glaciers, considering a variety of different and Bolivia. measurement techniques (e.g., monitoring of the mass balance, aerial photography, and remote sensing). Generally, a clear change in glacier evolution can be seen after the late 1970s, accelerating in the mid-1990s and again in the early 2000s (Figure 3.14). This is different from the glaciers located at mid or high latitudes, where accelerated melting started in the 1990s. The glaciers in the tropi- cal Andes appear to have had more negative mass balances than glaciers monitored worldwide. In the Peruvian Andes, glacial areas have been well documented and multiple reports found on average a retreat of 20–35 percent between the 1960s and the 2000s; most of that retreat occurred after 1985 (Vergara et al. 2011). A similar pattern of glacial reces- sion is found in the Bolivian Andes. A rapid decline has also been reported for the Ecuadorian Andes, where glaciers on Chimborazo shrunk by 57 percent during the period 1962–1997, while glaciers on the Cotopaxi and Antisana volcanoes shrunk in area during the period 1979–2007 by 37 percent and 33 percent respectively. For the Andes of Colombia, a moderate glacier area loss of 11 percent has been documented in the period of the 1950s to the 1990s, with a fourfold acceleration in retreat during the period of the 1990s to 2000s. In Venezuela, glacial retreat has been even more dramatic, The grey box around the average represents the uncertainty corresponding to ±1 standard deviation. Source: Rabatel et al. (2013), Figure 4. with a loss of about 87 percent between 1953–2003. 57 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Although the conditions in the Southern Andes are very dif- and Southern Patagonia, Ivins et al. (2011) inferred ice loss rates ferent (e.g., in terms of climate or sun angle) the trend in glacier of 26±6 Gt per year between 2003–2009, which explains a total retreat is obvious here as well (Figure 3.15). Lopez et al. (2010) loss of about 154±36 Gt over six years. Jacob et al. (2012) reached investigated changes in glacier length in 72 glaciers in the Chilean similar estimates using the same technique, with mass balance rates Southern Andes (Northern and Southern Patagonian Ice Field for the period 2003–2011 of 23±9 Gt per year in the Patagonian and Cordillera Darwin Ice Field) between 1945–2005, based on glaciers and of 6±12 Gt per year in the rest of South America aerial photographs and satellite images (ASTER, Landsat). They (including the tropical glaciers). However, the spatial resolution concluded that the observed general trend in glacial retreat is of about 300 km is extremely coarse and difficulties arise in dis- likely controlled by atmospheric warming. In the Northern Pata- tinguishing signals from hydrological storage and glacial isostatic gonian Ice Field, glaciers retreated in length by 4–36 percent, in adjustment (Gardner et al. 2013). the Southern Patagonian Ice Field by 0–27 percent, and further south in the Cordillera Darwin Ice Field by 3–38 percent. However, Snowpack and Snow Cover Changes glacial length fluctuations provide only limited insight into the In the tropical glacier region, due to the high solar radiance, with the imbalance of glaciers, and the large heterogeneity of glacial retreat sun close to the zenith, albedo appears to be a major determinant is very much influenced by such local conditions as exposition, in attenuating the melting process. Consequently, the frequency basin geometry, glacier dynamics, and response times. and intensity of snowfall plays a major role in determining the net A different way of measuring glacier mass loss rates is by radiation over the entire year, modulated by wet and dry seasons space gravimetry (GRACE)—by measuring the changing gravity (Rabatel et al. 2013). In the subtropical Andes of Chile and western field from satellites in regions with large continuous ice extent (a Argentina, where snowpack has been monitored for more than method available since 2003). For the large ice caps of Northern 50 years (1951–2004), there is no significant trend over this period (Masiokas et al. 2006, 2012). However, the data display a marked inter-annual variability ranging from 6–257 percent around the Figure 3.15: Ice loss from outlet glaciers on the Patagonian 1966–2004 mean, with a clear influence from the warm phases Ice Field in southern South America since the Little Ice Age. of ENSO (El Niño). Studies about snowpack in the Southern Andes are rare. It can generally be stated that changes in snowpack extent magnify changes in the seasonality of the water availability by a reduction of the flows in dry season and an increase in flows in wet seasons (Vicuña et al. 2013). 3.4.1.3 Projections of Glacial Change As the IPCC confirms with high confidence, glaciers worldwide are out of balance with current climatic conditions. Furthermore, it is very likely that anthropogenic forcing played a statistically significant role in the acceleration of the global glacier loss in the last decades of the 20th century (Bindoff et al. 2013). Various model projections for different future emissions sce- narios indicate that glaciers will continue to shrink in the future, even without further temperature increases. Confidence in these models is supported by their ability to reproduce past observed glacier changes using corresponding climate observations as forc- ing. Model validation is challenging, however, due to the scarcity of independent observations (currently available for only a small fraction of well-observed glaciers). In a 2°C world, Marzeion et al. (2012) expect a reduction in ice volume of tropical glaciers by 78–94 percent based on the period 1986–2005. This signal is less drastic in the Southern Andes, with an expected 21–52 percent volume reduction of the 4,700 Gt ice mass by 2100 for the same warming level. Marzeion et al. (2012) project the amount of tropical glaciers to be lost in a 3°C world Source: Glasser et al. (2011), Figure 1. at 82–97 percent, very similar to the 2°C world scenario. For the 58 LATI N AME R I CA A ND THE CA RIBBEA N much larger glaciers in the Southern Andes, the same study expects dammed behind the moraines of the last maximum extent of the a loss of 33–59 percent in a 3°C world. For the 21st century, with Little Ice Age in the mid-1800s. Carey et al. (2012) performed an a warming of 3°C above pre-industrial levels by 2100, Giesen and interdisciplinary case study on a glacial lake outburst flood (GLOF), Oerlemans (2013) estimate a volumetric loss of 66 percent in the which happened in Peru’s Cordillera Blanca mountain range in tropical glaciers (325 Gt) and of 27 percent in the Southern Andes 2010. Based on their analysis, they provide advice for effective (2,930 Gt). Marzeion et al. (2012) estimate an almost complete glacier hazard management. Hazard management is of high con- deglaciation (91–100 percent) of the remaining 280 Gt tropical cern, as projected warming will continue to promote glacial lake glaciers in a 4°C world (Figure 3.16.) This signal is less drastic formation (see Box 3.4: Glacial Lake Outbursts). in the Southern Andes, where a 44–72 percent deglaciation is estimated. An almost complete deglaciation for the 195 Gt tropi- 3.4.1.5 Synthesis This section indicates that glacial recession in South America cal glaciers (93–100 percent) in a 4°C world is also projected by has been significant. The tropical glaciers in the Central Andes in Radic´ et al. (2013). However, for the Southern Andes, the response particular have lost major portions of their volume in the course is much slower, and Radic ´ et al. (2013) expect 50 percent glacial of the 20th century. Regional studies show that the retreat has losses (3,080 Gt). All these models use a scaling methodology accelerated, with the strongest recession rates after 1985. A clear which may overestimate the recession of the small remnant tropi- trend of glacial retreat is also visible for glaciers in the southern cal glaciers. Regarding the Patagonian ice fields in the Southern Andes, which have lost about 20 percent of their volume. Regional Andes, Schaefer et al. (2013) estimate a glacial volume loss for studies highlight that the individual recession rate is very much the Northern Patagonian Ice Field of 590±50 Gt with 4°C global influenced by local conditions, which cause a large heterogeneity of warming with respect to pre-industrial levels. Their projections rates. Space gravimetry (GRACE) confirms that the declining trend of the future surface mass balance of the Northern Patagonian Ice in glacier volume has continued in the last decade. Monitoring Field predict a strong increase in ablation (refers to all processes of snow cover in the high altitudes of Chile and Argentina since that remove snow, ice, or water from a glacier or snowfield) from 1950 shows no significant trend (i.e., possible trends are hard to 2050 onward and a decrease in accumulation from 2080, both due identify in the records, since the inter-annual variability is large, to increasing temperatures. and clearly modulated by ENSO). The accelerated melting will lead to increasing runoff; when The recession of the tropical glaciers in the Central Andes the glacier reservoirs disappear, runoff will tend to decrease, par- will continue as rapidly as it has in recent decades. Even for low ticularly in the dry season (see Section 3.4.2, Water Resources, or intermediate emissions scenarios inducing a global warming Water Security, and Floods). Following the trend in the tropical of 2–3°C above pre-industrial levels, two comprehensive stud- Andes (Poveda and Pineda 2010), this peak is expected within the ies consistently project a glacial volume loss of 78–97 percent next 50 years (Chevallier et al. 2011) if it has not already occurred (Marzeion et al. 2012; Radic ´ et al. 2013). Both studies predict an (Baraer et al. 2012). almost complete deglaciation (93–100 percent) for a 4°C world. 3.4.1.4 Glacial Hazards In contrast, Giesen and Oerlemans (2013) project a loss of only Glacier hazards are a serious risk to populations in mountain 66 percent of the glacial volume of the year 2000 in the Central regions worldwide, where a general trend of glacial retreat has Andes with a global warming of about 3°C by 2100. Thus, irre- supported the formation of glacial lakes that were precariously spective of the temperature evolution in the next decades, large parts of the glaciers of the tropical Andes will be gone before the end of the century. In the Southern Andes, the model spread for Figure 3.16: Cumulative regional surface mass balance the 2–3°C global warming ranges from 22–59 percent; a com- relative to the 1986–2005 mean from the model forced with parison for individual scenarios is difficult. In a 4°C world, three CMIP5 projections up to the year 2100. SLE = Sea-level models project a glacier volume retreat of 44–74 percent by 2100. equivalent An important research gap is the lack of reliable projections for snowpack and snow cover changes in the Andes. 3.4.2 Water Resources, Water Security, and Floods LAC has abundant overall water resources, but their distribution is temporally and regionally unequal (Magrin et al. 2007). ENSO- related rainfall anomalies play a major role in many areas and Source: Modified after Marzeion et al. (2012), Figure 21. determine much of the inter-annual discharge variability (Baraer 59 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL In many parts of the region there is no clear trend in future Box 3.3: Water Security in the Mexico discharges due to the uncertainties in rainfall projections in different City Metropolitan Area GCMs (see also Section 3.3.3, Regional Precipitation Projections) and the diverging results of different impact models (Bravo et al. The Mexico City Metropolitan Area (MCMA) faces frequent climate- 2013; Davie et al. 2013; Döll and Schmied 2012; Hidalgo et al. 2013; related hazards. These include extremes of water and heat, with Imbach et al. 2012; Krol and Bronstert 2007; Malhi et al. 2009; floods on the one side and heat waves and droughts on the other. Rowell 2011; Schewe et al. 2013). The most common extreme events from 1980–2006 were floods resulting from heavy precipitation events (Romero Lankao 2010). 3.4.2.1 Central America and Mexico Up to 42 percent of the population in MCMA was estimated to be Milly et al. (2005) modeled a decrease in river runoff for Central vulnerable to climate change and natural hazards. Forty percent of America of up to 10 percent for the 20th century. Hidalgo et al. (2013) those live in so-called “high-risk areas” which are characterized by projected mean annual runoff to decrease by 10–30 percent by the very steep slopes of over 15 degrees where landslides can occur end of the 21st century with a median of 3°C regional warming. The after heavy precipitation events (Baker 2012). Exponential popula- same tendencies were shown in Imbach et al. (2012), who found tion growth in MCMA during the 20th century also contributed to the high vulnerability to flooding and water shortages (Brun 2007). decreases in annual runoff in 61–71 percent of the area (notably in Currently water is provided from the Mexico City basin aquifer, central Yucatan Peninsula, the mountains of Nicaragua, Honduras, with one-third transferred from external water basins (Romero and Guatemala) in a 2°C warmer world, depending on the sub- Lankao 2010). Overextraction of groundwater in combination with region and the emissions scenario. Increases were projected only locally unfavorable soil conditions (heavily saturated clay) have for just one percent of the area, mainly along the southern edge. caused parts of the city to subside and thus suffer more frequent Evapotranspiration was projected to increase more than 20 percent flooding (Baker 2012; Romero Lankao 2010). Another problem for in more humid areas (e.g., Costa Rica, Panama) whereas northern long-term water security is related to groundwater contamination of areas were projected to experience no change. Fabrega et al. (2013) the Mexico City basin aquifer, most probably due to surface waste- found precipitation increases by 5 percent or more for most regions water (Brun 2007) of Panama—but with no statistically significant changes in total The projection of temperature increases, more frequent and runoff in a 3°C world. Maurer et al. (2009) modeled the inflow of prolonged dry spells, and a (probable) precipitation decrease will two reservoirs in the Rio Lempa basin. They found decreases of harm water security and increase water dependencies in the grow- 13 percent in total annual reservoir inflow in a 2°C world and of ing MCMA. The water security of the water-providing external areas will also be affected (Magrin et al. 2007; Romero Lankao 2010; 24 percent in a 4°C world, implying potential reductions in hydro- Sosa-Rodriguez 2013). power capacities. Low flow years might occur more frequently, especially under a higher warming (Maurer et al. 2009). Global studies mostly confirm this picture. Milly et al. (2005) projected a decrease in river runoff in Central America of 5–20 per- et al. 2012; Cortés et al. 2011; Krol and Bronstert 2007; Mata et al. cent for the middle of the 21st century with 3°C global warming. 2001; Poveda 2004; Ronchail et al. 2005; Shi et al. 2013; Vicuña Nakaegawa et al. (2013) also found that the total annual runoff et al. 2010; Vuille et al. 2008). Large parts of the region are char- decreases for Mexico and Central America from 2075–2099 with 3°C acterized by inter-annual/seasonal rainfall variability through the warming. For the Rio Grande, annual discharge decreases by more oscillation of the Intertropical Convergence Zone (Garreaud et al. than 20 percent. In a 3°C world, Mesoamerica also experiences a 2009). Due to the unreliable rainfall, groundwater resources and strong decrease in discharge in the study of Schewe et al. (2013). water from glacier and snowmelt play a crucial role in supply- Portmann et al. (2013) reported a mean decrease across several ing local water (Chevallier et al. 2011; Hirata and Conicelli 2012; GCMs of more than 10 percent in groundwater recharge in a 4°C Vuille et al. 2008). world for Central America. The projected changes were much LAC suffers from widespread floods and landslides (Maynard- less pronounced when assuming lower global mean temperature Ford et al. 2008) which result from different origins (Dilley et al. increases of 2–3°C. 2005). Heavy precipitation events in the context of ENSO or tropical cyclones can lead to disastrous floods, especially in regions with 3.4.2.2 Caribbean steep terrains such as in the Andes and Central America (IPCC 2012; The assessment of water resources in the Caribbean relies more on Mata et al. 2001; Mimura et al. 2007; Poveda et al. 2001). Coastal assumptions and extrapolations from climatological data than on areas in the Caribbean and Central America suffer from flooding long-term hydrometric measurements, especially for the smaller as a result of storm surges and tropical cyclones (Dilley et al. 2005; islands (Cashman 2013; FAO 2003). Water provisioning is espe- Woodruff et al. 2013). In the Andes, glacial lake outbursts pres- cially difficult on islands which rely mainly on a single source of ent a permanent hazard for Andean cities (Chevallier et al. 2011). water (such as groundwater in Barbados, Bahamas, Antigua and 60 LATI N AME R I CA A ND THE CA RIBBEA N Barbuda, and Jamaica, or surface water in Trinidad and Tobago, Grenada, St. Vincent and the Grenadines, St. Lucia, Dominica, Box 3.4: Glacial Lake Outbursts and elsewhere)(Cashman 2013; Gencer 2013). The lack of long-term measured stream flows in the Caribbean Glacial lake outburst floods (GLOFs) originate from various causes. First, increasing glacier melting raises the water levels of lakes, renders the evaluation of hydrological models in the region dif- eventually resulting in an overflow of water or the breaking of dams. ficult, and future projections of runoff have only low confidence Second, ice instability may cause an avalanche of seracs into a lake, (Hidalgo et al. 2013). A combination of lower precipitation, high leading to suddenly higher water levels and the breaking of dams. abstraction rates, and sea-level rise may lead to intrusion of saline Ice instability might increase with increasing temperatures. Third, sea water into coastal groundwater aquifers (Cashman et al. 2010; glacier retreat may trigger major rock slides. It is important to note Cashman 2013). Another hydrological hazard regarding climate that the Andes belong to a region of high seismic activity, which can change is more severe flooding events related to tropical cyclones contribute to GLOFs. (Chevallier et al. 2011; Kaser et al. 2003). (Cashman et al. 2010). 3.4.2.3 Northern South America (Colombia) this is around 30 percent higher than what was measured during Restrepo et al. (2014) found significant discharge increases for the the 1930–2009 period (Baraer et al. 2012). Mulatos, Magdalena (at Calamar), Canal del Dique, and Fundación Kinouchi et al. (2013) simulated the glacier melt and runoff rivers, especially from 2000 to 2010. Regional studies of climate- in a headwater catchment of the Cordillera Real in Bolivia. They related hydrologic impacts are limited, as access to and quantity of applied different 1–1.5°C temperature increase scenarios by 2050 observational climate data is limited (Hoyos et al. 2013). Reanalysis and found only small changes in annual runoff. The seasonal or simulated/reconstructed datasets have been used, but Hoyos variation was, however, modified significantly. Under their projec- et al. (2013) reported substantial differences between climatologi- tions, streamflow during the dry and early wet season was reduced cal datasets and observed values. Nakaegawa and Vergara (2010) (e.g., because of snowmelt decrease, and during the wet season found a trend of decreasing mean annual river discharge due to it increased, especially in January and February). Baraer et al.(2012) increased evapotranspiration in the Magdalena Basin in a 3°C projected that once the glaciers have melted, the average dry season world. Monthly mean river discharge decreased significantly in discharge may decrease more than 60 percent in Parón and Llanga- April, October, and November at Puerto Berrio. It is important, nuco and up to 70 percent at La Balsa—with serious implications however, to note that mean precipitation was overestimated by for water supplies during the dry season. Juen et al. (2007) found about 35 percent for the GCMs used. little changes in total annual discharge by 2050 and 2080 in the Llanganuco catchment, but they did find a bigger amplitude of 3.4.2.4 Andes discharge seasonality (with a risk of very low flows during the In mountainous regions, winter precipitation accumulated as snow dry season). They concluded that a smaller glacier size causes and ice, similar to groundwater reserves, helps to buffer water short- decreasing glacier melting, but that this decrease is supplemented ages resulting from little or seasonal rainfall (Masiokas et al. 2006; by an increase in direct runoff from non-glaciered areas. Wet sea- Viviroli et al. 2011; Vuille et al. 2008). Downstream regions with son discharge was projected to increase from 10–26 percent and low summer precipitation in particular benefit from this temporal dry season discharge to decrease from 11–23 percent for warming water storage (Masiokas et al. 2013; Viviroli et al. 2011). Glacier >1.5°C in 2050 and >2°C in 2080 depending on the emissions retreat thus endangers water security in these areas (Vuille et al. scenario and timeframe (see Figure 3.17). For the northern half 2008). Current accelerated melting rates, however, imply a short- of the Andes, a very likely increase in flood frequency in a 4°C term local surge in water—and higher river flow peaks can cause world was projected (Hirabayashi et al. 2013). landslides and floods. Massive flood events have been associated with glacial lake outburst (Chevallier et al. 2011) (see Box 3.4). 3.4.2.6 Central Andes The Andes in Central Chile and central western Argentina are 3.4.2.5 Tropical Andes characterized by a direct relationship between the amount of Baraer et al. (2012) analyzed historical streamflow records for the snow accumulated in winter and river discharge released during Cordillera Blanca over the period 1990–2009; they showed that spring-summer (Masiokas et al. 2006), and around 85 percent of discharge was decreasing annually and during the dry season. the observed river flow variance over 60 years in the area can The trends were attributed to glacier retreat. Meltwater contrib- be explained by snowpack records (Masiokas et al. 2010, 2013). utes 10–20 percent to the total annual discharge of the Río Santa The inter-annual variability of snowpack extent is enormous, (Cordillera Blanca), but may rise to over 40 percent during the varying from zero percent to over 400 percent of the long-term dry season (Baraer et al. 2012). In the period from 1990–2009, the mean (Masiokas et al. 2006, 2010, 2013). Freshwater availability overall glaciered area was decreasing by 0.81 percent annually; is thus strongly dependent on mountain snowpack. In very dry 61 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 3.17: Changes in seasonal total runoff in 4 IPCC Box 3.5: Water Security in Quito, La climate-change scenarios with respect to the 1961–1990 Paz, Bogotá, and Lima mean monthly runoff. Precipitation patterns are different throughout the tropical Andes. The Pacific slopes in Colombia receive on average more than 8000 mm per year of rainfall, whereas large parts of the highlands of Bolivia and the Peruvian coast get less than 100 mm per year. Water security is a pressing issue because the capital cities experienced population growth rates of 11.9 percent in Quito and 20.6 percent in La Paz for the period 2000–2010. Bogotá and Quito are situated on steep mountain terrain at 2650 m and 2850 m above sea level respectively. High popula- tion densities provoke local water stresses, and both cities require inter-basin water transfers from the wet Amazonian slopes to meet their water demand. Sixty-two percent of Quito’s water is currently provided by the Amazonian basin. Lima, meanwhile, is the second- Source: Juen et al. (2007). largest desert city in the world and its water is supplied almost solely by the western slope of the Andes. Competition exists with other water users, including the agricultural sector (Buytaert and De Cortés et al. (2011) found that, for the period 1961–2006, river Bièvre 2012). regimes in the dry north were driven by snowmelt whereas those further south were more rainfall-dominated. The southern river systems (to the south of 35°) have displayed consistently earlier timing in peak annual flow rates. Vicuña et al. (2013) found indica- Box 3.6: Water from the Cordillera tions of a shift in the last 30 years to an earlier annual snowmelt Blanca season of around 15 days for the Mataguito basin. More high-flow discharges were also observed during the last The arid coastal area of Peru is home to approximately half of the 10 years. They occurred mostly during autumn months when high country’s population. Water is provided mainly by rivers coming rainfall and high minimum temperatures decreased the fraction down from the western slopes of the Cordillera of the Andes. Con- of precipitation falling as snow and cause a faster rainfall-runoff tributors to runoff during the rainy season are rainfall, groundwater, and glacier melt. In contrast, during the dry season rivers are fed by response (Vicuña et al. 2013). Annual low-flow levels during groundwater and glacier melting at higher than 5000 m (Chevallier spring and summer decreased significantly in the Mataguito basin et al. 2011). (Vicuña et al. 2013) and for some stations in the Limay River The water from the Cordillera Blanca supports human activities Basin (Seoane and López 2007). This trend in river flow variability at different altitudes. Irrigated agriculture is practiced between might endanger electrical power generation in the region, as the 2000–4000 m and at the foot of the Andes. Below 2000 m, Limay River basin contains many hydropower stations which yield electricity is generated (Chevallier et al. 2011; Kaser et al. 2003). around 26 percent of Argentina’s total electrical power generation To maintain the full capacity of electricity production, a discharge (Seoane and López 2007). of 60 m3 per second is required. As the minimum flow of the Río Projections of the mean number of snowy days decrease by Santa usually falls below that level, water management is needed to about 9 percent in a 2°C world, and 26 percent in a 4°C world, in guarantee the minimum discharge. the Mataguito basin (Demaria et al. 2013). In addition, the center The population of the Andes area is increasing. Due to this timing of mass of annual flow was projected to occur earlier, by population trend, and to the possibility of expanding cultivation into higher areas of the Andes under increasing temperatures, water 12 days in a >1.5°C world and by 16 days in a >3°C world. In demand is expected to increase. This might lead to conflicts with the Limarí basin, reductions in annual streamflow are probably hydroelectric power generation (Juen et al. 2007; Mark et al. 2005). intensified by increased evapotranspiration because a 19 percent rainfall decrease resulted in a 21 percent streamflow decrease in a 4°C world (Vicuña et al. 2010). Vicuña et al. (2010) also found years, with no or little solid precipitation in the upper water- increasing winter flows (28.8–108.4 percent), decreasing summer shed, glacier melting gains in importance, although there is little flows (–16.5 to –57.8 percent), and earlier center timing of mass information about the contribution of ice masses to river flow of annual flows for different sub-basins of the Limarí basin in a (Masiokas et al. 2013). >3°C world. For northeastern Chile, Arnell and Gosling (2013) 62 LATI N AME R I CA A ND THE CA RIBBEA N 3.4.2.7 Amazon Basin Box 3.7: Water Security in the Espinoza Villar et al. (2009) found significant decreasing mean and Central Andes minimum annual runoffs from 1990–2005 for southern Andean rivers (Peru, Bolivia) and increasing mean and maximum annual In the Andes, between latitudes 30 and 37° lie two major cities, runoffs for northern Andean Rivers (Ecuador) draining into the Santiago de Chile (Chile) and Mendoza (Argentina). Human activi- Amazon. For the southern Amazon, Li et al. (2008) found more ties in these cities are almost completely dependent on meltwater, dry events during 1970–1999. Espinoza Villar et al. (2009) reported especially in the drier Argentinean foothills which receive only around decreasing mean annual discharge and monthly minimum dis- 200 mm of annual precipitation (Masiokas et al. 2013). charge from 1974–2004 for Tapajós in the southeastern Amazon, Snowmelt is very important for water supply, hydroelectric the Peruvian Amazon Rivers, and the upstream Madeira. generation, and viniculture in large parts of Chile (Demaria et al. Guimberteau et al. (2013) analyzed the impacts of climate change on 2013). The central valley of Chile contains the majority of the coun- try’s reservoir storage and supplies water to several large towns. extreme streamflow over several Amazonian sub-basins by the middle Water demand results as well from agriculture, as 75 percent of of this century for a 2°C global warming scenario. They found that low the irrigated area in Chile is located here (Demaria et al. 2013). The flows would become more pronounced. The trend is significant at the Central Valley suffers from high inter-annual rainfall variability, with Madeira and Xingu rivers, with JJA precipitation decreases of 9 percent conditions being wetter during El Niño years and drier during El and 22 percent respectively. At Porto Velho, the decrease in median Niña years (Cortés et al. 2011). In years with above-average rainfall, low flows is about 30 percent; at Altamira, it is about 50 percent. In farmers irrigate annual crops (e.g., orchards, vineyards) with surface addition, Tosiyuki Nakaegawa et al. (2013) found total annual runoff water, whereas in below-average rainfall years they are forced to use decreases in the southern half of the Amazon River in a 3°C world. groundwater. The use of groundwater has recently reached unsus- Average annual runoff varied from –72 percent to +6 percent in a 3°C tainable levels, and the Chilean water authorities have therefore world for the Bolivian part of the Amazon (Alto Beni), assuming no restricted water extraction. Decreasing annual rainfall as a result of land use change (Fry et al. 2012). Nevertheless the projected ground- climate change would put even more pressure on agriculture and water recharge was consistently negative (–96 percent to –27 percent) water resources (Arumí et al. 2013). because potential evapotranspiration increases. Malhi et al.(2009) found an increase in dry-season intensity in eastern Amazonia in a 3° world and seasonal increased water stress because of climate change and deforestation. Langerwisch Box 3.8: Water Security and Glacial et al. (2013) found shifts in flood patterns in a 3° world. The Melt in La Paz and El Alto, Bolivia duration of flooding at the end of the 21st century was projected to be 0.5–1 months shorter than for 1961–1990. The probability of La Paz and El Alto receive 80 percent of their water from the Tuni three successive extreme wet years decreased by up to 30 percent Condoriri range. The contribution of glacier ice melt could be from 30–40 percent (World Bank 2008) up to 60 percent (Painter 2007). (Langerwisch et al. 2013). Since water demand has risen in recent years, the water manage- Median high flows in the western part of the Amazon basin ment of both cities is very much challenged (Jeschke et al. 2012; increase by 5–25 percent by the middle of this century for a warming Shi et al. 2013). Almost the entire energy supply of La Paz is sup- of 2°C; this trend is not, however, significant (Guimberteau et al. plied by hydroelectric power which comes mainly from two glacier 2013). In a 2°C world, the increase in high flow was projected to ranges—in the Zongo valley and Charquiri (Painter 2007). The be lower than in a warmer than 3°C world; low flows increase glaciers of the Cordillera Real encompass 55 percent of the Bolivian 10–30 percent under a 4° warming scenario (Guimberteau et al. glaciers. Between 1963–2006 they lost more than 40 percent of 2013). The flood zone is consistently projected to increase with a their volume (Soruco et al. 2009) and they are further declining (Liu 2–3 month longer inundation time in a 3°C world over several GCMs et al. 2013). It has been postulated that water demand might soon (Langerwisch et al. 2013). The average runoff and the maximum surpass water supply in El Alto (Shi et al. 2013). Future water and runoff increased in two subcatchments of the Paute basin for a 2°C energy supply will be increasingly critical due to rising demand, in global warming scenario from 2045–65 (Mora and Campozano et al. combination with decreasing tropical glacier volumes (Rabatel et al. 2013; Vuille et al. 2008). 2013). Exbrayat et al. (2014) also found increases in annual runoff for a catchment in the Ecuadorian Andes by 2100; they also showed a high variability of runoff projections depending on the choice of simulated a decreasing mean annual runoff for warming higher GCM, emissions scenario, and hydrological model. Similarly, Buytaert than 1°C over 21 GCMs. Projections by Döll (2009) also showed et al. (2009) showed that, due to the wide range of GCM projections, a reduced groundwater recharge for the central Andes region by the projected average monthly discharges diverge considerably in the 2050s with 2°C warming. the Paute River system in Ecuador under a 1°C increase by 2030. 63 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL For the northernmost Amazon and the river mouth region, from January–May at Corrientes on the Paraná River—but not at river flow and runoff coefficients decrease with a global warm- Posadas which lies further upstream. Nóbrega et al. (2011) found ing of 2°C in 2045–2065 (Guimberteau et al. 2013). In the same that for the Río Grande, a tributary of the Paraná, for every 1°C study, median low flows decrease by 20 percent for the Japura temperature rise annual flow increased by 8–9 percent in relation and Negro river and 55 percent at the Río Branco. to 1961–1990. Assuming 2°C warming, mean river flow ranged At the main stem of the Amazon River the runoff coefficient from –20 percent to +18 percent. is projected to slightly decrease at Óbidos (the last station of the Camilloni et al. (2013) projected an increase in frequency and Amazon before the mouth) with a warming of 2°C by 2050 duration of river flooding in a >3°C world in the Uruguay and (Guimberteau et al. 2013). Median low flow is projected to decrease Paraná basins. Hirabayashi et al. (2013) showed a decrease in the by 10 percent, but this trend is uncertain. In a 4°C world, Guim- 20th century 100-year return period for floods for the Parana in a berteau et al. (2013) projected that low flows and high flows would 4°C world, but there was little consistency across the 11 GCMs used. each increase by five percent at Óbidos. Döll and Schmied (2012), Besides river floods, storm surge floods present a major hazard however, projected the mean river discharge of the downstream for Buenos Aires. Barros et al. (2005; 2008) found a greater inland part of the Amazon to increase for under a 2°C warming by 2050 reach of recurrent storm surge floods by 2070 under a 3°C global in one GCM but decrease in another GCM. warming scenario. Assuming no changes in population distribu- tion, permanent coastal flooding due to sea-level rise will play a 3.4.2.8 Northeast Brazil minor role and will affect rather sparsely populated areas at the Krol and Bronstert (2007) found that a decrease in precipitation coast of Buenos Aires and its surroundings. In contrast, Pousa et al. by the end of the 21st century would significantly decrease the (2013) projected that sea-level rise could aggravate the impact of runoff of the Jaguaribe River and the stored volume in the Ceará storm surge floods in Buenos Aires. reservoir. In contrast, an increase of precipitation by 50 percent did not significantly increase river runoff because of an accompany- 3.4.2.10 Southernmost South America ing increase in water demand. Döll and Schmied (2012) projected Milly et al. (2005) simulated a decrease in mean relative runoff of the seasonality of river discharge in northeastern Brazil to remain up to 10 percent, which is in agreement with observed 20th century stable but also that mean river discharge would decrease by the trends for southernmost South America. They projected a decrease middle of the 21st century under a 2°C global warming scenario. in mean relative runoff of 10–30 percent for southernmost South Due to uncertainty in the GCM projections, there is no clear America for the middle of the 21st century with 3°C global warm- signal about the relative change of annual discharge for northeastern ing. Schewe et al. (2013) found similar results for a >2°C world. South America under 2°C warming (Schewe et al. 2013). Portmann et al. (2013) projected both strong decreases and increases in mean 3.4.2.11 Synthesis groundwater discharge for northeastern Brazil in a 4°C world ENSO-related rainfall anomalies play a major role in many areas depending on the GCM. Assuming different warming scenarios in LAC and determine much of the inter-annual discharge variabil- with varying levels of decreasing rainfall, Montenegro and Ragab ity. In Central America, there is a high agreement on decreasing (2010) projected strong decreases in groundwater recharge of up mean annual runoff and discharge, although the magnitude of to 77 percent and streamflows of up to 72 percent for a subcatch- the change varies (Arnell and Gosling 2013; Hidalgo et al. 2013; ment of the Sao Francisco River Basin. Imbach et al. 2012; Maurer et al. 2009; Milly et al. 2005; Nakae- gawa et al. 2013; Schewe et al. 2013). The trend seems to be more 3.4.2.9 Río de la Plata pronounced for the northern than for the southern part of Central The Río de la Plata region experienced a 10–30 percent increase America (Hidalgo et al. 2013; Imbach et al. 2012). Therefore, water in river runoff during the 20th century (García and Vargas 1998; stress may increase, especially in arid areas with high population Jaime and Menéndez 2002; Menéndez and Berbery 2005; Milly densities and during the dry season. et al. 2005). There is no consensus, however, among river runoff The Caribbean lacks long-term measured stream flow data. projections for the Río de la Plata and its tributaries because the Runoff projections are therefore of low confidence (Cashman 2013; projected direction of rainfall trend varies among GCMs. Milly FAO 2003; Hidalgo et al. 2013). However, freshwater availability et al. (2005) modeled an increase in mean relative runoff for the may decrease for several reasons. Sea-level rise (Mimura et al. Rio de la Plata region of 20–50 percent for the middle of the 21st 2007) may lead to an intrusion of sea water into coastal aquifers century and using a 3°C warming scenario. River flow projec- (Cashman et al. 2010; Cashman 2013) and summer precipitation tions in the upper Paraguay River basin varied from ±10 percent is projected to decrease (Mimura et al. 2007). Regionally the by 2030 for a 1.5° warming scenario and by ±20 percent by risk of flooding and mudslides with high mortality rates is high 2070 for a 2°C warming scenario (Bravo et al. 2013). For a 3°C (Cashman 2013; Edwards 2011; Williams 2010). Although floods world, Nakaegawa et al. (2013) projected increasing discharge often seem to be associated with land-use change, more severe 64 LATI N AME R I CA A ND THE CA RIBBEA N flooding events may also occur in the context of climate change phase (Hatfield et al. 2011). Generally, warmer temperatures act (Cashman et al. 2010; IPCC 2012). to decrease the development phase of perennial crops, resulting In Northern South America (Colombia), there are only a in earlier crop flowering and reduced seed sets (Craufurd and limited number of regional hydrological impact studies available Wheeler 2009). When temperatures increase above the maximum, for northern South America, and rainfall projections are uncertain. plant growth and yields can be drastically reduced (Ackerman and Conclusions about projected hydrological impacts are therefore Stanton 2013; Berg et al. 2013; Luo 2011). of low confidence. The optimum seasonal average temperature for maximum In the Andes, higher discharge seasonality is projected for the grain yield is 15°C for wheat, 18°C for maize, 22°C for soybeans, Tropical Andes. Streamflows during the dry season may decrease and 23°C for rice (Hatfield et al. 2011; Lobell and Gourdji 2012). because of ongoing glacier retreat (Baraer et al. 2012; Juen et al. Hatfield et al. (2011) also identified average temperatures leading 2007; Kinouchi et al. 2013). Lower dry season discharge has already to a total crop failure: 34°C for wheat, 35°C for maize, 39°C for been observed during the past two decades (Baraer et al. 2012). soybeans, and 35°C for rice. Short intervals of a few days above the However, streamflow during the wet season may increase (Juen optimum average temperature can lead to strong yield decreases et al. 2007; Kinouchi et al. 2013). The region has a high flood risk (Ackerman and Stanton 2013). Teixeira et al. (2013) project an (e.g., due to accelerated glacier melting; see Box 3.4: Glacial Lake increasing occurrence of heat stress for maize, rice, and soybeans Outbursts) (Carey 2005; Hirabayashi et al. 2013). For the Central in Latin America. Lobell and Gourdji (2012) estimate global yield Andes, more streamflow was observed and projected to occur at declines of 3–8 percent per °C of temperature increase based on earlier dates locally (Cortés et al. 2011; Vicuña et al. 2013; Demaria a literature review. It is, however, important to note that there et al. 2013). Lower dry season discharges may cause significant are numerous knowledge gaps concerning plant reactions to tem- water supply problems in urban areas. peratures above their optimum averages (Craufurd and Wheeler Amazon Basin: Runoff and discharge projections for most parts 2009; Porter et al. 2014). Moreover, plants are somewhat capable of the Amazon basin are diverging, especially for the southern of adapting to changing climatic conditions—and it is unclear if and eastern areas. The main reasons for this are the high vari- climate change will alter growing conditions too fast for crops to ability of rainfall projections using different GCMs and uncertain- adapt on their own (Ackerman and Stanton 2013). ties introduced by hydrological impact models. However, for the western part of the basin a likely increase in streamflow, runoff, 3.4.3.2 Plant Diseases flood zone, and inundation time was projected (Guimberteau et al. How pests and diseases will spread under future climate condi- 2013; Langerwisch et al. 2013; Mora and Campozano et al. 2013). tions, and how severe the effects will be on yields and production Northeast Brazil: The direction of discharge and groundwater quantities, is unclear. Already today crop diseases are respon- recharge trends vary due to diverging rainfall projections under sible for losses of 10 percent or more of global food production different GCMs (Döll and Schmied 2012; Krol and Bronstert 2007; (Chakraborty and Newton 2011; Ghini et al. 2011; Luck et al. 2011). Portmann et al. 2013; Schewe et al. 2013). Climate change is expected to alter the geographic distribution Río de la Plata: There are no consistent river runoff projections of insects and diseases in much of the world (Porter et al. 2014). for the basin because the directions of rainfall projections vary The knowledge on climate change and plant diseases, however, is among the GCMs (Bravo et al. 2013; Milly et al. 2005; Nakaegawa still very limited (Ghini et al. 2011; Luck et al. 2011). The impacts et al. 2013; Nóbrega et al. 2011). differ greatly between crops and pathogens as do the interactions Southernmost South America: A decrease of mean runoff among hosts, pathogens, microorganisms, and the climate (Ghini was projected with a high confidence (Milly et al. 2005; Schewe et al. 2011; Bebber et al. 2014). The existing knowledge base is et al. 2013). inadequate to make generalizations about the behavior of crop diseases under a changing climate (Luck et al. 2011). Factors that 3.4.3 Climate Change Impacts on Agriculture are most likely to influence the development of plant diseases are increasing atmospheric CO2, increasing winter temperatures, and 3.4.3.1 Temperature Sensitivity Crop Thresholds increasing humidity (Luck et al. 2011). Agriculture is one of the most climate dependent human activities, One recent example of the impact of plant diseases on agricul- and the development and growth of plants is affected to a very ture in the LAC region is the outbreak of coffee leaf rust (Hemileia large extent by temperature. Every plant has a range between a vastratix), considered the most destructive coffee disease, in Central maximum and minimum temperature in which the plant can exist America during the 2012–13 growing season. Around 50 percent of and an optimum temperature at which growth is at its optimal rate the roughly one million ha under coffee production in the region (Hatfield et al. 2011). Crops often require different temperatures were affected by the disease, reducing the production quantity in their numerous development stages and are very sensitive to by an estimated 17 percent in comparison to the previous year temperatures above the optimum, especially during the pollination (Ghini et al. 2011; ICO 2013). The outbreak devastated small holder 65 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL In Chile, even when including the CO2 fertilization effect, Box 3.9: Surface Ozone Concentrations yields could be reduced by 2050 by 5–10 percent for maize and 10–20 percent for wheat in comparison to 1971–2000 levels with Surface ozone concentrations have negative impacts on agricultural 2.7°C global warming if no adaptation measures are implemented yields. The impact on crop yields strongly depends on the seasonal (Meza and Silva 2009). In Argentina, yields for wheat, maize, and and regional distribution of surface ozone, as it is not distributed soybeans are projected to decline by 16, 24, and 25 percent respec- evenly in the atmosphere (Teixeira et al. 2011). Declines in yield tively by 2080 under a 3.5°C global warming scenario without levels currently range from 7–125 percent for wheat, from 6–16 per- CO2 fertilization (ECLAC 2010). Yield declines are less pronounced cent for soybeans, from 3–4 percent for rice, and from 3–5 percent with only 2.7°C global warming, with declines of 11 percent for for maize; wheat and soybeans are especially sensitive to surface ozone (Van Dingenen et al. 2009; Teixeira et al. 2011). Jaggard et wheat, 15 percent for maize, and 14 percent for soybeans; includ- al. (2010) noted that the impact of ozone on crop yields has been ing CO2 fertilization increases yields slightly (ECLAC 2010). In neglected in many climate impact projections and found that the southern Brazil, bean and maize productivity would decline by benefits of the CO2 fertilization effect (see Box 2.4) could be offset by 15–30 percent in comparison to 1971–2000 levels under a global the negative effect of increased ozone concentrations on C3 plants mean warming of 2°C by 2050 and by 30–45 percent with 4°C (and even lead to a yield reduction of five percent in C4 plants). By warming by 2080 without CO2 fertilization but with technological 2030, increasing surface ozone could lead to yield declines in Latin progress (Costa et al. 2009). Including CO2 fertilization for beans America by up to 7.8 percent for wheat, 2.9 percent for maize, and leads to productivity increases of up to 15 percent (Costa et al. 7.5 percent for soybeans depending on the emissions levels of 2009). Because maize is a C4 crop, including CO2 fertilization has ozone precursors (Avnery et al. 2011). only a limited impact and productivity keeps decreasing (Costa et al. 2009). Rain-fed sugarcane yields could increase by 15–59 percent with global warming of 1.5–2.3°C by 2050, including CO2 fertil- ization and technological improvement (Marin et al. 2012). In coffee growers and possibly contributed to rising coffee prices the Brazilian Amazon, soybean yields decline by 44 percent with globally (NYT 2014). Coffee leaf rust, together with soybean rust 4°C mean global warming by 2050 and by 1.8 percent with a 2°C (Phakopsora pachyrhizi), are expected to move further south and temperature increase (Lapola et al. 2011). On average and over all affect South American countries with global temperatures increas- analyzed crop types, yields are projected to decline by 31 percent ing by approximately 3.5°C by 2080 compared to pre-industrial when temperatures increase by 4°C without CO2 fertilization and levels (Alves et al. 2011). increase by 14 percent when temperatures increase by 2°C with CO2 fertilization (Lapola et al. 2011). 3.4.3.3 Projected Changes in Crop Yields In Ecuador, ECLAC (2010) projects yield declines of 53 percent Climate change impacts on crop yields vary depending on crop for maize, 9 percent for beans, 41 percent for bananas, 36 percent type and location. Fernandes et al. (2012) projected changes in for sugarcane, 23 percent for coffee, and 21 percent for cocoa; crop yields in 2050 (compared to 1989–2010) under global warming ECLAC also projects yield increases of up to 37 percent for rice for scenarios of between 1.7°C and 2.3°C. Table 3.6 and Figure 3.18 the year 2080 with 3.5°C warming. Colombian agriculture, mean- present some of their key results. It is important to note that, while, is projected to be severely impacted by climate change. Up when considering adaptation measures, yield declines are less to 80 percent of agricultural crops currently cultivated in Colombia pronounced but still negative for wheat, soybeans, and maize. in 60 percent of the cultivation areas of the country would be nega- Yield projections for rice show a different picture. With the tively affected by 2–2.5°C global temperature increases by 2050 if exception of Brazil, Mexico, and the Caribbean, where temperatures no adaptation measures are introduced (Ramirez-Villegas et al. are already high, rice yields could increase by up to 12 percent 2012). Perennial crops (notably such high-value crops as tropical by 2020 and by 17 percent by 2050 as average conditions for rice fruit, cocoa, bananas, and coffee) could be particularly affected by photosynthesis would improve with increasing temperatures climate change (Ramirez-Villegas et al. 2012). Coffee farming might (Fernandes et al. 2012). have to migrate to higher altitudes or other cultivation regions to Nelson, Rosegrant and Koo et al. (2010) project yield changes maintain present yields, a problem also relevant in other parts of for different crops in LAC with a 1.8–2.5°C global temperature Latin America (Camargo 2010; Laderach et al. 2011; Zullo et al. 2011). increase by 2050. Their key results, shown in Table 3.6, show that In Panama, yield changes for maize range from –0.8 to yields generally decline without CO2 fertilization; this is most pro- +2.4 percent for global warming of 1.7–1.9°C in 2055, and from nounced for irrigated maize, soybeans, and wheat. CO2 fertilization +1.5 to +4.5 percent for global warming of 2.2–3.3°C in 2085 increases yields for rice, soybean, and maize by over 10 percent including CO2 fertilization (Ruane et al. 2013). Accelerated crop besides irrigated maize (Nelson, Rosegrant, Koo et al. 2010). development helps to complete the grain-filling phase before the 66 LATI N AME R I CA A ND THE CA RIBBEA N Table 3.6: Projected Changes in Yields and Productivity Induced by Climate Change. YIELD OR SOURCE SCENARIO TIME HORIZON REGION CROP PRODUCTIVITY EFFECT Fernandes et al. (2012) A1B / B1 2050 Brazil Soybeans –30 to –70 % Brazil, Ecuador, Maize up to –60 % Brazil Wheat –13 to –50 % LAC Rice up to +17 % Meza and Silva (2009) A1F1 2050 Chile Maize –5 to –10 % Wheat –10 to –20 % Costa et al. (2009) A2 2050 Brazil Beans –15 to –30 % Maize –15 to –30 % Ruane et al. (2013) A2 / B1 2050 Panama Maize –0.8 to +2.4 % A2 /B1 2080 Panama Maize +1.5 to +4.5 % Lapola et al. (2011) A2 2050 Brazilian Amazon Soybeans –1.8 to –44 % Marin et al. (2013) A2 / B2 2050 Southern Brazil Sugarcane +15 to +59 % ECLAC (2010) A2 2080 Ecuador Maize –53 % Beans –9 % Bananas –41 % Sugarcane –36 % Coffee –23 % Cocoa –21 % Rice +37 % A2 / B2 Argentina Wheat –11 to –16 % Maize –15 to –24 % Soybeans –14 to –25 % Nelson et al. (2010) A2 2050 LAC Maize –3.0 to +2.2 % Rice –6.4 to +12.7 % Soybeans –2.5 to +19.5 % Wheat –5.6 to +12.2 % beginning of dry periods with high levels of water stress (Ruane A significant positive relationship between crop yield change and et al. 2013). In Mexico, wheat yields decline with global tempera- temperature is revealed, however, when CO2 fertilization is con- tures rising between 1.6–2.1°C by 2050 across several crop models sidered (see Table 3.7 and Figure 3.19), although the beneficial and GCMs (Rosenzweig et al. 2013b). Yield declines are more effects of CO2 fertilization are highly uncertain (Ainsworth et al. pronounced with stronger warming, but they remain relatively 2008)(see Box 2.4) The interpretation of these results therefore small because CO2 fertilization reduces the negative yield effect requires some caution, as model assumptions made regarding in the crop models (Rosenzweig et al. 2013b). CO2 fertilization may not hold in an actual crop production envi- A meta-analysis of the impacts of climate change on crop yields ronment. If the effects of CO2 fertilization are not considered, for the LAC region (see Section 6.3, Meta-analysis of Crop Yield the relationship remains significant but becomes negative, with Changes with Climate Change,) reveals no significant influence of increasing temperature leading to considerable yield declines (see temperature increase over crop yields across all available studies. Figure 3.19). 67 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 3.18: Aggregate impacts on crop yields in the LAC region with adaptation, computed by the AZS-BioMA platform under 2020 and 2050 NCAR GCM for A1B scenario. Coarse grains Oil seeds XSM URY COL PER ECU CHL CHL MEX PER ARG URY 2050 XSM 2050 CAC 2020 ECU 2020 MEX CAC ARG COL BRA BRA –40 –30 –20 –10 0 –40 –30 –20 –10 0 10 20 Rice Wheat COL PER ECU URY PER MEX URY CAC BRA BRA XSM 2050 CHL 2050 CAC 2020 ECU 2020 ARG XSM MEX ARG CHL COL 0 5 10 15 20 –60 –50 –40 –30 –20 –10 0 10 Source: Fernandes et al. (2012), Figure 4.1. Table 3.7: Summary of Crop Yield Responses to Climate Figure 3.19: Meta-analysis of crop yield reductions. Change, Adaptation Measures, and CO2 Fertilization. SLOPE R2 T-STAT P-VALUE Full dataset 0.0023 0 0.1255 0.9 Crop yield change with 0.07 0.266 2.81 0.009** effect of CO2 fertilization Studies not considering –0.065 0.24 –2.65 0.0145* the effects of adaptation measures or CO2 fertilization Results of a general linear model applied to all studies with reported values for changes in yield and changes in temperature, to studies considering the effect of CO2 fertilization, and to studies not considering the effects of adaptation measures nor those of CO2 fertilization. Significance levels: *P<0.05, **P<0.01, ***P<0.001. Best-fit line for LAC studies not considering the effects of adaptation mea- sures or those of CO2 fertilization (blue line) and for studies considering the effects of CO2 fertilization (but no adaptation, orange) and their 95 percent confidence intervals of regressions consistent with the data based on 500 bootstrap samples (patches). 68 LATI N AME R I CA A ND THE CA RIBBEA N To conclude, the possible effects of climate change on crop 2020, by 16.2 percent by2050, and by 22.1 percent by 2080 with yields in the region are very diverse. Yield impacts differ among 2.9°C regional warming (ECLAC 2010). regions and crops and also among different GCMs, emissions sce- narios, and crop models (see Table 3.6) (Berg et al. 2013). Most of 3.4.3.6 Climate Change Impacts on Food Security the effects of rising temperatures are expected to be negative, even Nelson, Rosegrant and Koo et al. (2010) project that international if lessened CO2 fertilization (which introduces large uncertainties crop prices will increase significantly even when ignoring climate into the impact projections). For some crops, however, increasing change—mainly driven by population growth, income growth, temperatures might have positive effects, such as increasing yields and demand for biofuels. The price of wheat is projected to for rice and sugarcane. increase by 39 percent, rice by 62 percent, maize by 63 percent, and soybeans by 72 percent. Including climate change with global 3.4.3.4 Climate Change Impacts on Livestock mean temperature increasing by 2.5 °C by 2050, and without CO2 The livestock sector in the LAC region is of high economic fertilization, would accelerate price increases by an additional importance, especially in major livestock producing and export- 94–111 percent for wheat, 32–37 percent for rice, 52–55 percent for ing countries Brazil and Argentina (ECLAC et al. 2012), and the maize, and 11–14 percent for soybeans. Including CO2 fertilization impacts of climate change on livestock systems in developing would lead to less severe price increases by 2050 (Nelson, Rose- countries are diverse (Thornton et al. 2009). Climate change can grant, Koo et al. 2010). These results are confirmed by the IPCC severely impact the quantity and quality of feed, as rising tem- AR5 report: Increasing food prices as a consequence of changing peratures, increasing atmospheric CO2 concentrations, and changes climatic conditions are to be expected by 2050 without taking the in precipitation patterns influence the availability of nutrients, CO2 fertilization effect into account; including elevated CO2 will the productivity of grasslands, and the composition of pastures. temper price increases (Porter et al. 2014). Furthermore, heat stress directly affects livestock productivity. Climate change poses great risks to the economic develop- Cattle, in particular, are susceptible to high temperatures. Heat ment of Latin America and the Caribbean; it not only threatens stress is known to reduce food intake and milk production and economic growth but also poverty reduction and food security also to affect reproduction, growth, and cattle mortality rates (ECLAC 2010). Without climate change, calorie availability would (Porter et al. 2014). Higher temperatures are also closely linked to be expected to increase by 3.7 percent, up to 2,985 calories per growing water demand for livestock, increasing the competition capita in 2050 in LAC (Nelson, Rosegrant, Koo et al. 2010). How- and demand for water in water-scarce regions. More scientific ever, with climate change and without CO2 fertilization, per capita research is needed, meanwhile, on the effects of climate change calorie availability in 2050 is expected to drop below the value on livestock diseases and livestock biodiversity. for the year 2000 (2,879 calories per capita) (Nelson, Rosegrant, Koo et al. 2010). These projections show that climate change 3.4.3.5 Projected Impacts on Livestock threatens food security, especially for people with low incomes, With a 2.7°C warming by 2060, livestock species choice (i.e., the as access to food is highly dependent on income (FAO 2013). The adoption of new livestock) is projected to decline across Argen- cascading impacts of warming that reduce productivity in other tina, Brazil, Chile, Colombia, Ecuador, Uruguay, and Venezuela sectors apart from agriculture can further reduce economic output by 3.2 percent for beef cattle; by 2.3 percent for dairy cattle; by and negatively affect incomes (Porter et al. 2014). Results from 0.9 percent for chicken; and by 0.5 percent for pigs. Meanwhile, a Brazilian study (Assad et al. 2013) on climate change impacts the adoption of sheep species is projected to increase by an aver- on agriculture to 2030 project that Brazil could face a reduction age of 7 percent across the region, and by more in Colombia of approximately 11 million hectares of high quality agricultural (11.3 percent), Chile (14.45 percent), and Ecuador (19.27 percent) land as a result of climate change with the South Region (current (Seo et al. 2010). grain belt) being the worst impacted losing ~5 million ha of ‘low With a lower warming of 1.3–2.3°C by 2060, the pattern of climate risk’ crop land. The increase in climate risk in the south declining livestock species choice for beef cattle, dairy cattle, could be partially offset by transferring grain production to the chicken, and pigs, and the increasing choice of sheep, remains the central region currently occupied by low productivity pastures same but is less pronounced (Seo et al. 2010). According to Seo et al. (sub regional reallocation). Intensification of livestock and pasture (2010), the choice of sheep increases with increasing temperatures systems will also offset projected losses due to climate change. and decreasing precipitation because sheep are better adapted to In general, however, the production declines can be expected to these conditions than other livestock species. In Paraguay, beef impact prices, domestic demand, and net exports of most crops/ cattle production is projected to increase by 4.4 percent by 2020, livestock products. Simulations from this study across all the but then decline by 7.4 percent by 2050 and 27.1 percent by 2080 climate change scenarios suggest that rising staple and export in a scenario leading to 3.5°C regional warming (ECLAC 2010). crop and beef prices could double the agricultural contribution Beef cattle production is projected to decline by 1.5 percent by to Brazil’s economy. 69 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL after which climate change becomes increasingly important for Box 3.10: Critical Ecosystem species loss (MEA 2005; Vuuren et al. 2006). Services of High Andean Mountain 3.4.4.2 Impacts of Future Climate Change Ecosystems on Biodiversity Forecasts of future changes in biodiversity are generally alarming A highly critical ecosystem service provided by the LAC region is (e.g., Bellard et al. 2012; Foden et al. 2013). Using a global meta- that of carbon storage. For example the ecosystems of the Andean analysis, MacLean and Wilson (2011) found a mean extinction mountains, including tropical montane cloud forests, the high-alti- probability of 10 percent by 2100 across taxa, regions, and warm- tude wetlands, and the páramos ecosystem, store large amounts of carbon. Despite the fact that they cover a mere 3 percent of global ing levels. Warren et al. (2013) found that, globally, 57 percent of land area they store about 30 percent of the global carbon stock of plants and 34 percent of animals will lose greater than 50 percent terrestrial ecosystems (Peña et al. 2011). Further, numerous large of their habitat in a 4°C world. cities (such as Quito, Bogota or La Paz) extract part of their water A comparative review of different model predictions across taxa supply from páramos areas. and regions revealed a large variability in the predicted ranges of biodiversity loss, especially at the local level (Bellard et al. 2012). One reason for this variability is that there is still high uncertainty about the capacity of species to buffer the effects of climate change 3.4.3.7 Synthesis (Moritz and Agudo 2013). Nonetheless, Scholes et al. (2014) state The results of the climate change impact projections on crop yields that there is “high confidence that climate change will contribute differ among studies, but most authors agree that climate change to increased extinction risk for terrestrial and freshwater species will very likely decrease agricultural yields of important food over the coming century.” crops in LAC (see Table 3.6) (ECLAC 2010; Fernandes et al. 2012; 3.4.4.3 Projections of Potential Future Shifts Nelson, Rosegrant, Koo et al. 2010). An exception is the possible in Ecosystems and Ecoregions yield development of rice in some regions (ECLAC 2010; Fernandes The G200 ecoregions (Olson and Dinerstein 2002) located in et al. 2012; Nelson, Rosegrant, Koo et al. 2010). Although studies Latin America and the Caribbean may experience severe climate on climate change impacts on livestock are scarce (Thornton et al. change in the future (Beaumont et al. 2010). Li et al. (2013) found 2009), the few studies that are available indicate that beef and strong local climatic changes in the ecoregions Coastal Venezuela dairy cattle production will decline under increasing temperatures, Montane Forests, Amazon River and Flooded Forests, and Atlantic as heat stress is a major influencing factor of cattle productivity Dry Forests for 2–4°C global warming. Further, 38.4 percent of the (Seo et al. 2010; Thornton et al. 2009). Sheep production could surface of the biodiversity hotspot of Tumbes-Choco-Magdalena become more important in the future, as sheep are better adapted to and 11.5 percent of the Mesoamerican biodiversity hotspot will be warmer and drier conditions than cattle and pigs (Seo et al. 2010). experiencing no-analogue climates in a warmer than 2°C world 3.4.4 Climate Change Impacts on Biodiversity (Garcia-Lopez et al. 2013). Heyder et al. (2011) find a range of small to severe eco- 3.4.4.1 Current Status and Current Threats system changes for the whole South American continent in to Biodiversity their projections for a 2°C and warmer world. In a 4°C world, Biodiversity, the diversity of genes, populations, species, communi- results of one dynamic vegetation model show severe ecosystem ties, ecosystems, and biomes, is the foundation for all ecosystem changes for more than 33 percent of the area in 21 out of 26 processes (MEA 2005). Climate change is a major threat to biodi- distinct biogeographic regions in South America (Gerten et al. versity, as species have evolved to live within specific temperature 2013). Warszawski et al. (2013), meanwhile, projected such ranges that may be surpassed faster than species are able to adapt. severe ecosystem changes in a 3°C world in South America South America is a biodiversity hotspot, particularly due to (notably in Amazon, Guyana moist forests, and Brazilian Cer- the large extent of tropical rainforests (MEA 2005; Myers et al. rado) when applying an ensemble of seven dynamic vegeta- 2000) and the continent’s long geographical isolation until approxi- tion models. Imbach et al. (2012) projected that such severe mately 3 million years ago—which together have nurtured a high ecosystem changes at global mean warming levels greater than number of endemic species. Habitat destruction and fragmenta- 3°C would lead to a considerable decrease in tree cover, indi- tion by land-use change as well as the commercial exploitation cated by a change in leaf area index of more than 20 percent of species groups are currently larger threats to biodiversity than across 77–89 percent of the area. Bellard et al. (2014) projected climate change (e.g., Hof et al. 2011). Land-use change is expected that out of 723 Caribbean islands, 63 and 356 of them will be to have a greater impact on plants than climate change by 2050, entirely submerged under one and six meters of sea-level rise, 70 LATI N AME R I CA A ND THE CA RIBBEA N respectively. They also found that 165 of the islands will be at turnover, and many in western South America and Central America least half-submerged (i.e., having lost more than 50 percent of to experience at least 50 percent species turnover, so that future their area) under one meter of sea-level rise, and 533 under communities would bear little resemblance to the currently estab- six meters of sea-level rise. While a six meter sea-level rise is lished ones (Lawler et al. 2009). In a greater than 3°C world, the not realistically expected to happen within this century, a one entire LAC region would experience high bioclimatic unsuitability meter sea-level rise is within the range of sea-level rise projected for amphibians in general (many grid cells between 50–80 percent under a global mean warming of 4°C at the end of this century loss). In the Northern Andes, 166 frog species (73 percent of local (see Section 3.3.7, Regional Sea-level Rise). frog fauna) and, in Central America, 211 species (66 percent of local salamander fauna), would lose their local climatic suitability 3.4.4.4 Projections of Habitat Changes, Species between 2070–2099 (Hof et al. 2011). Range and Distribution Shifts, and Extinction Risks Based on historical data, Sinervo et al. (2010) assume that if the for Species and Species Groups rate of change in maximum air temperature at 99 Mexican weather Microorganisms stations continues unabated by 2080, 56 percent of the viviparous Little is known about the consequences of future climate change on lizard species would go extinct by 2050 and 66 percent by 2080; of the microbial biodiversity due to the complex microbial feedback loops oviparous species, 46 percent would go extinct by 2050 and 61 percent within the climate system (Singh et al. 2010). The ratio between by 2080. By 2080, the predicted loss of suitable areas for the royal heterotrophic soil bacteria and fungi will likely be affected (Rinnan ground snake (Liophis reginae) is 30 percent (Mesquita et al. 2013). et al. 2007). Generally, temperature increase stimulates microbial Sea-level rise will affect the reproductive behavior of sea turtles, growth and accelerates decomposition, which leads to an increase which return to the same nesting sites every breeding season and in heterotrophic respiration (Davidson and Janssens 2006). therefore rely on relatively constant shorelines for laying their eggs. Invertebrates Fish et al. (2005) predict a 14/31/50 percent habitat loss of nest- Insects act as pollinators to ensure plant fertilization, but they may also ing sites for endangered sea turtles by 2050 under 0.2/0.5/0.9 m emerge as pests. Climate change affects temperature-driven reproductive sea-level rise respectively on Bonaire Island. Narrow and shallow cycles of many insect populations. In a 4°C world, Deutsch et al. (2008) beaches are predicted to be most vulnerable, but turtles seem to projected a range contraction of 20 percent for tropical insects, because prefer steep slopes which might to some extent alleviate climate tropical insects will face near-lethal temperatures much faster than change impacts at their preferred nesting sites. those in temperate climates. Estay et al. (2009) projected an increase Birds in insect population densities of grain pests in Chile of 10–14 percent Birds are most diverse in the tropics where they typically have smaller in a 3°C world and 12–22 percent in a 4°C world. home ranges than migratory birds in temperate zones (Jetz et al. 2007). This renders tropical bird diversity especially vulnerable to Amphibians and Reptiles extinctions caused by climate change and accompanying habitat Due to the difficulties in entangling the relative contributions destruction. Anciaes et al. (2006) projected that 50 percent of 49 of climate versus land use change, Scholes et al. (2014) stated neotropical manakin (passerine bird) species will have lost more that “due to low agreement among studies, there is only medium than 80 percent of their current habitat by around 2055 with mean confidence in detection of extinctions and attribution of Central global warming of 2°C. For a similar time frame and warming sce- American amphibian extinctions to climate change.” Amphibians nario, Souza et al. (2011) projected that 44 of 51 endemic Brazilian are particularly vulnerable because, due to their permeable skin, Atlantic forest bird species would lose their distribution area by 2050, they depend on constant water availability at least during some which corresponds to a habitat reduction of 45 percent of the original periods of their life cycle. Loyola et al. (2013) found that most of the area. The study assumes that the entire area of the Atlantic forest 444 amphibian species in the Atlantic Forest Biodiversity hotspot is suitable for these bird species. However, about 80 percent of the in Brazil could increase their range, while 160 species would face Atlantic forest is already deforested, and most remaining forest areas range contractions with 1.9°C global warming in 2050. A more are fragmented and isolated. Twenty-six Cerrado bird species face recent projection for 2050 includes different dispersal scenarios for 14–80 percent range contractions under a no-dispersal scenario, and the amphibians in this region and projects a majority of the 430 they face a 5 percent range increase to a 74 percent range decrease amphibian species would face range contractions accompanied under a full-dispersal scenario in a 3°C world (Marini et al. 2009). by an overall species loss with 1.9°C global warming (Lemes et al. 2014). Already in a 2°C world, 85–95 percent of species face Marsupials net loss in range size, and 13–15 percent of species would lose Most of the 55 marsupial species found in Brazil inhabit forested 100 percent of their current range depending on the modeled areas and are therefore exposed to both climate change and dispersal limitation (Lawler et al. 2009). In a 4°C world, most land use change caused by deforestation. Loyola et al. (2012) ecoregions are projected to experience at least 30 percent species found that marsupial species in Brazil face range contractions of 71 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL 67 percent of their original habitat with mean global warming of 2–5 percent of mammal species, 2–4 percent of bird species, and approximately 2°C by 2050. 1–7 percent of butterfly species in Mexico, as well as 38–66 per- cent of plant species in the Brazilian Cerrado, would go extinct Mammals (Thomas et al. 2004). At a warming of 1.8–2.0°C, these values Schloss et al. (2012) projected that up to 39 percent of the mammals increase to 2–8 percent, 3–5 percent, and 3–7 percent respectively. in the Neotropics would be unable to keep pace with climate-change With an increase in global mean temperature of greater than 2°C, velocity due to their limited dispersal abilities in a 4°C world. 44–79 percent of plant species in Amazonia are projected to go Torres, Jayat and Pacheco (2012) projected an at least 33 percent extinct (Thomas et al. 2004). habitat loss for the maned wolf (Chrysocyon brachyurus) in Central South America with 2°C global warming by 2050. Synthesis Climate change induced negative effects on biodiversity, from Plant Species range contractions to extinctions, are very likely in a warmer Plants are especially vulnerable to climate change because indi- than 2°C world. Climate change impacts on local biodiversity by vidual plants cannot migrate to avoid thermal stress. As a result, 2100 will depend on the balance between the number of species their dispersal mode will largely determine to what extent they abandoning an area and those facing local extinctions versus the may be able to adapt to changing climatic conditions. Moreover, number of species invading that same area due to thermal stress. plants directly respond to elevated atmospheric CO2 levels (see Species and species communities are possibly threatened by range Box 2.4)—but the degree to which some plant species may ben- contractions, extinctions, predator/prey disruptions, and phenology efit from rising CO2 levels is still being debated (Cox et al. 2013; changes due to climate and land use changes. Their opportunity Rammig et al. 2010). to survive in this changing environment lies in their capacity to Brazil is the country with the largest number of vascular plant adapt to these new conditions or to migrate to avoid them. As the species (>50,000) on Earth (ICSU-LAC 2010, p.57). Most future adaptive capacity of affected species and ecosystems is hard to projections paint a bleak picture for plant biodiversity, mostly due project or quantify, models need to use simplified approaches as to land-use change as a result of deforestation and, increasingly, the implemented in bioclimatic envelope models, species-distribution impacts of climate change. Simon et al (2013) projected a reduction models, and dynamic global vegetation models. in geographic distribution of 78 percent (±7 percent) in a >2°C One clear trend regarding future warming levels is that the world for 110 Brazilian Cerrado plant species. Feeley et al (2012) more temperatures are projected to increase, the more species projected a loss of suitable habitat area in the Amazon region of diversity is affected. Mountainous regions in the tropics (e.g., cloud between 8.2–81.5 percent in a 2°C world and 11.6–98.7 percent forests) are projected to become very vulnerable due to their high in a 4°C world, and a change in plant species richness between number of endemic and highly specialized species which might –4.1 percent to –89.8 percent in a 2°C world and –25.0 percent to face mountaintop extinctions. Most models do not take biotic complete loss for the studied species in a 4°C world. In Mexico, interactions (e.g., food-web interactions, species competition) and even common species are under threat, and great differences in resource limitations into account. Therefore, the realized ecologi- species response (0.1–64 percent loss) to regional warming above cal niche of species within an ecosystem might be much smaller 1.5°C in 2050 even among related tree species (e.g., oak trees) are than what is potentially possible according to climatic and other being projected (Gómez-Mendoza and Arriaga 2007). environmental conditions leading to shifts in ecological zones. Species Groups 3.4.5 Amazon Rainforest Dieback Most studies on range contractions focus on single species or spe- and Tipping Point cies groups; fewer studies have attempted to project the impact of future climate change at the community or biome level. Rojas- Old-growth rainforests in the Amazon basin store approximately Soto et al. (2012) projected a reduction of 54–76 percent in the 100 billion tons of carbon in their biomass (Malhi et al. 2006; extent of the Mexican cloud forest with 2°C global warming by Saatchi et al. 2011). Through evapotranspiration, Amazon rainfor- 2050. They concluded that this reduction forces tree communities ests recycle 28–48 percent of precipitation and contribute to local to move about 200 m to higher elevations. Similarly, Ponce-Reyes rainfall (van der Ent et al. 2010). A loss of these forest ecosystems et al. (2012) projected a 68 percent loss of suitable area for cloud due to climate change would release an enormous amount of forests in Mexico with a global warming of 3°C by 2080. Alarm- carbon into the atmosphere and reduce their evapotranspiration ingly, 90 percent of the cloud forest that is currently protected potential (thereby reducing atmospheric moisture); this would lead will not be climatically suitable for this ecosystem by 2080. As a to strong climate feedbacks (Betts et al. 2004; Costa and Pires 2010; consequence, climate change may lead to the extinction of 9 of Cox et al. 2004). These climate feedbacks, in combination with the 37 vertebrate species restricted to Mexican cloud forests. With large-scale deforestation, put the Amazon rainforest on the list of an increase of global mean temperature of 0.8–1.7°C by 2050, potential tipping elments in the Earth system (Lenton et al. 2008). 72 LATI N AME R I CA A ND THE CA RIBBEA N Factors Leading to Forest Dieback to cause large-scale forest dieback and to increase atmospheric and Potential Feedbacks CO2 concentrations. Observations of the Current Period Extreme drought events in combination with land use changes Current observations show that forests in Amazonia are adapted lead to an increased frequency in forest fires because the flamma- to seasonal drought (Davidson et al. 2012) mainly due to the bility of forests increases with a more open forest canopy (i.e., as ability to access deep soil water through deep rooting systems it allows more radiation to dry out the forest surface and enhance (Nepstad et al. 1994). It has been long debated, however, whether fire spread) (Ray et al. 2005). Fires were twice as frequent in 2005 the productivity of tropical rainforests during the dry season is as during the average of the previous seven years and they were more limited by precipitation or by cloud cover. Depending on the spatially concentrated in the arc of deforestation in the southern method used, remote sensing or modeling, seasonal droughts were Amazon (Zeng et al. 2008). Increasing fires resulting from defor- thought to enhance productivity either by more light entering the estation, pasture renewal, and other land-use-related activities canopy through reduced cloud cover, or by the combined effects increase the vulnerability of the Amazon rainforest to fire and of several interconnected processes (Brando et al. 2010; Huete et al. cause changes in forest composition and productivity (Brando et al. 2006). These findings have been challenged by remote sensing 2012; Morton et al. 2013). This interplay of factors is thought to experts who ascribe greening effects to saturation of the satellite initiate a positive bidirectional feedback loop between fire and sensor used (Samanta et al. 2011) or to changes in the optical forest which could initialize forest transformation into savannahs constellation of the sensor (near-infrared reflectance) (Morton and contribute to the Amazon tipping point (Nepstad et al. 2001, et al. 2014). Recent evidence from a large-scale and long-term 2008). Basin-wide measurements show that the combined effects experiment suggests that the feedbacks between climatic extreme of fire and drought can change the Amazon into a carbon source events such as droughts and forest fires increase the likelihood of (e.g., with 0.48 PgC emitted in 2010); it remains carbon-neutral, an Amazon dieback (Brando et al. 2014). meanwhile, during wet years (Gatti et al. 2014). Extreme weather events in the Amazon may have several causes. Deforestation is feared to influence the lateral moisture transport The drought events in 2005 and 2010 were not related to El Niño from coastal to inland areas because convective precipitation is but rather to high Atlantic sea-surface temperatures (Marengo et al. responsible for recycling precipitation locally. Walker et al. (2009) 2011). Cox et al. (2008) found the gradient between northern and showed that the current distribution of conservation areas in the southern tropical Atlantic lead to a warmer and drier atmosphere Amazon basin, which cover approximately 37 percent of the area, over the Amazon. Atypically low rainfall inflicted water stress on would be sufficient to maintain regional moisture transport and 1.9 million km² (2005) and 3.0 million km² (2010) of forest area recycling of precipitation when considering different deforesta- (Lewis et al. 2011). As a result, approximately 2.5 million km² tion rates. Influences of dynamic vegetation on water fluxes and (2005) and 3.2 million km² (2010) of forest area were affected by eventual carbon-climate feedbacks were not part of this study. increased tree mortality and reduced tree growth due to water Future Projections stress (Lewis et al. 2011). These two droughts are thought to have Water Stress reversed the currently assumed role of the intact forest as a carbon The Maximum Climatological Water Deficit (MCWD) (Aragão et sink and lead to decreased carbon storage of approximately 1.6 Pg al. 2007) is an indicator for drought intensity and plant water stress carbon (2005) and 2.2 Pg carbon (2010) compared to non-drought and correlates to tree mortality. For the period 2070–2099, 17 out years (Lewis et al. 2011; Phillips et al. 2009).34 The 2005 drought of 19 GCMs project increased water stress for Amazon rainforests reversed a long-term carbon sink in 136 permanent measurement in a 3°C world (which implies a mean regional warming of 5°C) plots (Phillips et al. 2009). (Malhi et al. 2009). Ten of 19 GCM projections passed the Two multi-year rainfall exclusion experiments in Caxiuanã and approximate bioclimatic threshold from rainforest to seasonal forest Tapajós National Forest generated remarkably similar results of (MCWD<–200 mm). Similarly, Zelazowski et al. (2011) projected water drought-induced tree mortality. These experiments demonstrated stress for forests to increase from 1980–2100 with an increase in global that once deep soil water is depleted, wood production is reduced mean temperature of 2–4°C above pre-industrial levels. They found by up to 62 percent, aboveground net primary productivity declines that humid tropical forests of Amazonia would retreat by 80 percent by 41 percent, and mortality rates for trees almost double (Brando for two out of 17 GCMs. Seven other models projected at least a et al. 2008; Costa and Pires 2010; Nepstad et al. 2007). Thus, an 10 percent contraction of the current extent of humid tropical forest. increase in extreme droughts in the Amazon region (medium con- Changes in Forest Cover fidence for Central South America in the IPCC SREX)(IPCC 2012) Hirota et al. (2010) simulated potential changes in Amazon for- or a prolonged dry season (Fu et al. 2013) may have the potential est cover. At a regional temperature increase of 2°C, forest cover decreased by 11 percent along with a 20 percent reduction in 34 The combination of reduced uptake of carbon due to the drought and loss of carbon precipitation in the Western Amazon forest (66°W). At a regional due to drought induced tree mortality and decomposition committed over several years. 73 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL temperature increase of 4°C, forest cover loss was 80 percent Biomass Loss independent of precipitation reduction. With fire included, tree Huntingford et al. (2013) showed that Amazon rainforest vegeta- cover was even further reduced. tion carbon generally increases in a 4°C world (and with regional Cook and Vizy (2008) projected a 69 percent reduction in temperature increases of up to 10°C). They conclude that there is rainforest cover extent in a 4°C world. Cook et al. (2012) showed evidence for forest resilience despite considerable uncertainties. that, with a regional warming of 3–4°C (which corresponds to Previous studies, however, projected considerable losses in a mean global warming of 3°C), soil moisture was reduced by biomass. In a 4°C world, Huntingford et al. (2008) found that eight percent, leaf area index (i.e., corresponds to forest cover) with a regional temperature increase of approximately 10°C from decreased by 12.6 percent, and the land-atmosphere carbon flux 1860–2100, vegetation carbon was reduced by about 7 kgC per m2. increased by about 27.2 percent due to fire from 2070–2099 com- Similarly, Fisher et al. (2010) simulated decreasing carbon stocks pared to 1961–1990. Cox et al. (2004) showed that a 4°C world of 15–20 kgC per m2 in 1950, 2.6–27 kgC per m2 in 2050, and could lead to a forest cover decrease of 10–80 percent. This find- 1–10 kgC per m2 in 2100 with a regional temperature increase of ing was recently challenged by Good et al. (2013), who showed about 2–5°C from 1900–2100. Galbraith et al. (2010) found that that an improved version of the Hadley model (HadGEM2-ES) vegetation carbon may either decrease or increase depending on projected only minimal changes in the Amazon forest extent due the emissions scenario and vegetation model. In their simulations, to forests surviving better in warmer and drier climates than pre- vegetation carbon changed by –10 to +35 percent relative to viously thought. However, about 40 percent of the difference in 1983–2002 with a regional warming of 4–8°C from 2003 to 2100. forest dieback projections was associated with differences in the Rammig et al. (2010) showed that including the CO2 fertiliza- projected changes in dry-season length in the new simulations, tion effect resulted in an increase in biomass in Eastern Ama- as a result of improvements in the simulated autotrophic plant zonia (EA) of 5.5–6.4 kgC per m2, in Northwestern Amazonia respiration, the soil moisture component, and a reduced gradient in (NWA) of 2.9–5.5 kgC per m2, and in Southern Amazonia (SA) of the tropical Atlantic sea surface temperatures. This does not mean 2.1–4.3 kgC per m2 in a 3°C world (see also Vergara and Scholz that the forest became more resilient in the updated model, but 2010). The probability of a dieback was zero percent for all regions rather that the improved vegetation-climate feedback mechanisms in this case. In contrast, climate-only effects without the buffering impose less stress on the simulated forests so that tree mortality CO2 fertilization effect resulted in biomass reduction in EA (–1.8 is reduced under warmer and drier climates (Good et al. 2013). In to –0.6 kgC per m2), NWA (–1.2 to 0.6 kgC per m2), and SA (–3.3 line with this, Cox et al. (2013) quantified a smaller risk (between to –2.6 kgC per m2) (Figure 3.20). The probability of biomass 1–21 percent) of Amazon dieback when constraining projections loss without CO2 fertilization was projected to 86.4 percent (EA), based on current observations of the atmospheric CO2 growth rate 85.9 percent (NWA), and 100 percent (SA). The probability of a and assuming that the CO2 fertilization effect is large. dieback (>25 percent biomass loss) was projected to be 15.7 percent Figure 3.20: Simulated precipitation changes in Eastern Amazonia from the 24 IPCC-AR4 GCMs with regional warming levels of 2–4.5 K (left panel). Simulated changes in biomass from LPJmL forced by the 24 IPCC-AR4 climate scenarios assuming strong CO2 fertilization effects (middle panel, CLIM+CO2) and no CO2 fertilization effects (CLIM only, right panel). Source: Calculated from Rammig et al. (2010). 74 LATI N AME R I CA A ND THE CA RIBBEA N (EA), 1.1 percent (NWA), and 61 percent (SA) (Rammig et al. 2010). 3.4.5.1 Synthesis These results imply that understanding the still uncertain strength Intensive research efforts over the past decades have enormously of the CO2 fertilization effect is critical for an accurate prediction improved the understanding and interaction of processes linking of the Amazon tipping point; it therefore urgently requires further climate, vegetation dynamics, land-use change, and fire in the empirical verification by experimental data from the Amazon region. Amazon. However, the identification of the processes and the Deforestation and forest degradation, for example from selec- quantification of thresholds at which an irreversible approach tive logging (Asner et al. 2005), are also factors which crucially toward a tipping point is triggered (e.g., a potential transition influence future changes in vegetation carbon. Gumpenberger et from forest to savannah) are still incomplete. al. (2010) found relative changes in carbon stocks of –35 percent Overall, the most recent studies suggest that the Amazon dieback to +40 percent in a protection scenario without deforestation and is an unlikely, but possible, future for the Amazon region (Good –55 percent to –5 percent with 50 percent deforestation in a 4°C et al. 2013). Projected future precipitation and the effects of CO2 world. Poulter et al. (2010) found a 24.5 percent agreement of fertilization on tropical tree growth remain the processes with the projections for a decrease in biomass in simulations with 9 GCMs highest uncertainty. Climate-driven changes in dry season length in a 4°C world. and recurrence of extreme drought years, as well as the impact of fires on forest degradation, add to the list of unknowns for which Large-Scale Moisture Transport combined effects still remain to be investigated in an integrative Several studies show that changes in moisture transport and regional study across the Amazon. A critical tipping point has been identified precipitation are strongly linked to deforestation. Costa and Pires at around 40 percent deforestation, when altered water and energy (2010) found in simulation runs with a coupled climate-vegetation feedbacks between remaining tropical forest and climate may lead model that precipitation was reduced in 9–11 of 12 months under to a decrease in precipitation (Sampaio et al. 2007). different deforestation scenarios (based on Soares-Filho et al. 2006). A basin-wide Amazon forest dieback caused by feedbacks Sampaio et al. (2007) performed simulations with a coupled climate- between climate and the global carbon cycle is a potential tipping vegetation model for different agricultural regimes and deforesta- point of high impact. Such a climate impact has been proposed tion scenarios of 20–100 percent for the Amazon basin (based on if regional temperatures increase by more than 4°C and global Soares-Filho et al. 2006). When replacing forest with pasture, they mean temperatures increase by more than 3°C toward the end of projected a 0.8°C increase in regional temperature and a 0.2 percent the 21st century. Recent analyses have, however, downgraded the reduction in precipitation at 20 percent deforestation levels. At 40 per- probability from 21 percent to 0.24 percent for the 4°C regional cent deforestation levels, regional temperatures increased by 1.7°C warming level when coupled carbon-cycle climate models are and precipitation was reduced by –2.2 percent. At 50–80 percent adjusted to better represent the inter-annual variability of tropical deforestation, regional temperatures increased by 1.8–2–1°C and temperatures and related CO2 emissions (Cox et al. 2013). This precipitation decreased by 5.8–14.9 percent. When replacing forest holds true, however, only when the CO2 fertilization effect is with soybeans at 50 percent deforestation, regional temperatures realized as implemented in current vegetation models (Rammig increased by 2.9°C and precipitation decreased by 4.6 percent. At et al. 2010). Moreover, large-scale forest degradation as a result 80–100 percent deforestation, regional temperatures increased by of increasing drought may already impair ecosystem services and 3.7–4.2°C and precipitation decreased by 19.2–25.8 percent. functions without a forest dieback necessarily to occur. Fire 3.4.6 Fisheries and Coral Reefs Studies projecting future fires in the Amazon are still scarce. Fires are projected to increase along major roads in the southern and 3.4.6.1 Vulnerability to Climate Change southwestern part of Amazonia with a 1.8°C global warming by Significant impacts of human origin, such as changes in tempera- 2040–2050 (Silvestrini et al. 2011; Soares-Filho et al. 2012). High ture, salinity, oxygen content, and pH levels, have been observed rates of deforestation would contribute to an increasing fire occur- for the oceans over the past 60 years (Pörtner et al. 2014). Such rence of 19 percent by 2050, whereas climate change alone would changes can have direct and indirect impacts on fishery resources account for a 12 percent increase (Silvestrini et al. 2011). Drought and food security (for example, as fish prey reacts sensitively to and anthropogenic fire incidences could significantly increase the ocean acidification or habitat is lost due to coral reef degradation) risk of future fires especially at the southern margin of the Amazon (Turley and Boot 2010). (Brando et al. 2014). If deforestation can be excluded in protected The Humboldt Current System off the coast of Peru and Chile areas, future fire risk would be evenly reduced, emphasizing the sustains one of the richest fisheries grounds in the world and management option to increase carbon storage when avoiding or is highly sensitive to climate variability such as that resulting reducing forest degradation (Silvestrini et al. 2011; Soares-Filho from ENSO. The Eastern Pacific region’s fishery is dominated by et al. 2012). catches of small pelagic fish which respond sensitively to changes 75 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL in oceanographic conditions. Peru and Colombia are among the of more than 2°C in 2100 and 851–1,370 µatm, which is associated eight countries whose fisheries are most vulnerable to climate with 4°C warming. They found differential responses for coral change (Allison et al. 2009; Magrin et al. 2014). species, with 38 percent and 44 percent of all species studied The Caribbean Sea and parts of the South Atlantic, in contrast exhibiting sensitivity to both scenarios. Echinoderms and mollusks to the Eastern Pacific, sustain vast coral reefs (see Section 3.4.6.2, both exhibit high sensitivity to ocean acidification as they have Coral Reefs). The Caribbean Sea sustains a more diverse but less low metabolic rates and depend on calcium carbonate for shell productive fishery (UBC 2011). formation. Wittmann and Pörtner (2013) stated that most species studied will be affected in a 4°C world, but that the effects will be Uncertain Climate Change Effects on the Intensity of Coastal Upwelling visible before that. In fact, nearly 50 percent of all species studied Several effects of climate change on upwelling and related ecosystem show sensitivity to a 2°C warming. While crustaceans appear functioning have been hypothesized—and they point in opposite to be relatively resilient, with about a third of species affected directions. One hypothesis is that a decrease in productivity may be in a 4°C scenario, the effects on fish are already significant in a driven by a globally warmer ocean in the future, as observed under relatively low CO2 concentration scenario; more than 40 percent El Niño conditions. However, sea-surface temperatures have not of species are affected. This figure nearly doubles for the high been observed to increase in the Humboldt Current System over the scenario. However, Wittmann and Pörtner (2013) stress that their last 60 years (Hoegh-Guldberg et al. 2014). The hypothesis is further investigation into fish species´ sensitivity is biased toward reef fish. contradicted by data indicating higher productivity during warm inter- Species Interaction and Ecosystem Effects glacial periods (Chavez and Messié 2009) and a projected weakening It is important to note that the effects of ocean acidification do not of trade winds and associated El Niño events (Bakun et al. 2010). act in isolation; rather, they act in concert with such changes as However, particularly the latter projections are highly uncertain. As rising sea-surface temperatures, changes in salinity, and decreas- described in Section 2.3.2, El-Niño/Southern Oscillation, projections ing nutrient availability due to enhanced stratification. These on the frequency and intensity of future El Niño events are uncertain. further interact with non-climatic pressures such as pollution and An increase in productivity, in turn, has been hypothesized to overfishing (Hoegh-Guldberg et al. 2014). For example, increasing occur due to a stronger land-sea temperature gradient—the land temperatures may lead to a drastic narrowing of species’ thermal surface warms faster than the ocean waters—leading to stronger tolerance window—with effects such as delayed spawning migra- winds driving stronger upwelling (Bakun 1990; Chavez and Messié tion or mortality (Pörtner and Farrell 2008). 2009). However, analysis has been unable to determine whether Sensitivity to ocean acidification can further narrow this toler- or not the recorded intensification of winds at the eastern sides ance window (Wittmann and Pörtner 2013). Such combined effects of the world’s oceans is due to inconsistencies in measurement are to date little understood and knowledge remains limited due to techniques over long time scales. Comparison with non-upwelling limits on experimental settings and the limited ability to discern regions, however, indicates that the wind intensification is more anthropogenic effects in a setting of high natural variability such pronounced in upwelling regions (Bakun et al. 2010). Analysis of as the Humboldt Current System (Hoegh-Guldberg et al. 2014). The changes in coastal biomass shows an upward trend of waters within expected synergistic effects of multiple pressures mean, however, the Humboldt Current System for 1998–2007, particularly for the that assessments remain conservative for a single or a subset of Peruvian coast. In contrast, the southern part of the California Cur- pressures (Wittmann and Pörtner 2013). rent System (south of 30°N off the coast of Baja California) shows A further potential risk factor for biological productivity and a negative trend (Demarcq 2009). Moreover, whether increased fisheries is the effect of climate-induced changes on species interac- upwelling would lead to higher productivity depends on nutrient tion, which can occur due to the differential responses of species availability—and changing physical conditions may disturb the to changing environmental cues. For example, phytoplankton and natural food web structure. zooplankton biomass changes may affect fish biomass (Taylor et al. Physiological Effects of Ocean Acidification 2012b). Different sensitivities to increasing CO2 concentration on Marine Species may lead to remarkable shifts in species composition (Turley and The IPCC states with high confidence that rising CO2 levels will Gattuso 2012). Similarly, asynchronous responses to warming increasingly affect marine organisms (Pörtner et al. 2014). A meta- may lead to mismatches in the predator-prey relationship. Such analysis of 228 studies revealed overall negative impacts of ocean effects have been detected to modify species interactions at five acidification not only on calcification but also on survival, growth, trophic levels (Pörtner et al. 2014). For the Humboldt Current development, and abundance (Kroeker et al. 2013). Wittmann and System, changes in the intensity of coastal upwelling add yet Pörtner (2013) conducted a meta-analysis of existing studies to another factor that may endanger the balance of various species determine responses for a range of future CO2 concentrations. The interactions. It has been hypothesized that the predator-prey ranges include 500–600 µatm, which is associated with warming relationship between phyto- and zooplankton could be disrupted 76 LATI N AME R I CA A ND THE CA RIBBEA N due to excessive offshore transportation of zooplankton. If in such a scenario overfishing would not allow for small pelagic fish to Box 3.11: Freshwater Fisheries— control phytoplankton growth, sedimentation of organic matter Vulnerability Factors to Climate may contribute to hypoxia, red tides, and the accumulation of Change methane (Bakun et al. 2010). Projections of Changes to Coastal Upwelling The freshwater fishery of the Amazon River is an important source The direction and magnitude of upwelling changes remains of protein for the local population. According to the FAO, annual per uncertain, particularly for the Humboldt Current System. Wang capita fish consumption in the Amazon basin may exceed 30 kg, which is significantly higher than in areas remote from freshwater et al. (2010) showed that results diverge for different models. sources where consumption has been estimated at around 9 kg per With approximately 1.5°C global warming in 2030–2039, projec- person per year (FAO 2010). As such, the river and its hydrological tions show an overall increase in the decadal averaged upwell- network provide an important source of proteins and minerals for ing index (July) for the California Current when compared to the local population. 1980–1989 for most of the GCMs analyzed. For the Humboldt Those resources may be under threat from climate change Current System, however, there is very little agreement among (Ficke et al. 2007) as rising water temperatures may exceed species’ models both in terms of direction and magnitude of change. temperature tolerance window. In addition, warmer waters are asso- Wang et al. (2010) observed that the factors driving coastal ciated with higher toxicity of common pollutants (e.g., heavy metals) upwelling systems are too local to be captured by the coarse and lower oxygen solubility; this may negatively affect exposed resolution of global models. organisms. In addition, “blackwaters” such as the Amazon varzèa lakes, depend on seasonal flooding for nutrient replenishment and Projected Changes to Fisheries Catch Potential for toxins to be flushed out. Reduced river flow and a reduced size In response to changing oceanic conditions, including seawater of the flood plain may further lead to a reduced habitat for spawning temperature and salinity, fish stocks have been observed in, and (Ficke et al. 2007). are further expected to shift to, higher latitudes (Perry et al. 2005). This ultimately affects local fisheries in the tropics and subtropics. A further climate impact on the productivity of fisheries is the fish catch is projected to decrease by up to 30°percent, but there reduction in productivity at the base of the food chain due to the are also increases towards the south. stronger stratification of warming waters (Behrenfeld et al. 2006). There are, however, inherent uncertainties in the projections In fact, primary productivity has been shown to have declined by presented here. Cheung et al (2010) pointed out that important 6 percent since the early 1980s (Gregg 2003). Declining pH levels factors, such as expected declines in ocean pH (ocean acidifica- and increasing hypoxia may further negatively impact fisheries tion), direct human pressures, and local processes which escape (Cheung et al. 2011). the coarse resolution of global models, are not taken into account. No regional projections of future fishery catches appear to Incorporating the effects of decreasing ocean pH and reduced oxygen exist. A global study that considers the habitat preference of availability in the northeast Atlantic yields catch potentials that 1,066 commercially caught species and projects changes to pri- are 20–30 percent lower relative to simulations not considering mary productivity computes the expected changes in fish species these factors (Cheung et al. 2011). distribution and regional patterns of maximum catch potential by Taking into consideration the effects of species interaction 2055 in a scenario leading to warming of approximately 2°C in on redistribution and abundance, Fernandes et al. (2013) report 2050 (and 4°C by 2100) (Cheung et al. 2010). latitudinal shifts in the North Atlantic to be 20 percent lower than Results of Cheung et al. (2010) for LAC indicate a mixed picture reported by the bioclimatic envelope model developed by Cheung (see Figure 3.21). Concurrent with the expectation of fish popula- et al. (2010). A further limitation of the studies is that results are tions migrating poleward into colder waters, the waters further given as 10-year averages and consequently do not take into con- offshore of the southern part of the Latin American continent are sideration abrupt transitions as observed under El Niño conditions. expected to see an up to 100 percent increase in catch potential. It should also be noted that not all captured species are included Catch potentials are expected to decrease by 15–50 percent along in these calculations, with small-scale fisheries possibly not taken the Caribbean coasts and by more than 50 percent off the Ama- into account (Estrella Arellano and Swartzman 2010). Finally, it zonas estuary and the Rio de la Plata. Furthermore, the Caribbean needs to be taken into account that local changes in fish population waters and parts of the Atlantic coast of Central America are distribution are likely to affect the small-scale sector most severely, expected to see declines in the range of 5–50 percent, with the as artisanal fishers will not have the means to capture the benefits waters around Cuba, Haiti, the Dominican Republic, and Puerto of higher productivity at higher latitudes further offshore. Rico, as well as Trinidad and Tobago, St. Lucia, and Barbados, For the Exclusive Economic Zone of the Humboldt Current particularly severely affected. Along the coasts of Peru and Chile, System, Blanchard et al. (2012) projected a 35 percent decline in 77 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 3.21: Change in maximum catch potential for Latin American and Caribbean waters. Source: Cheung et al. (2010), Figure 1a. phytoplankton and zooplankton density and similar magnitudes Vulnerability of Coral Reefs to Climate Change of change in the overall biomass of fish under 2°C global warming Coral reefs are particularly vulnerable to the double effects of by 2050. Comparing the impacts of climate change with fishing climate change on the oceans: rising temperatures and declin- pressure shows that climate impacts drive ecosystem change under ing pH levels. This vulnerability is particularly visible in events low fishing rates (0.2yr–1); under heavy fishing pressure (0.8yr–1), of coral bleaching, where external stresses cause corals to expel climate effects become secondary. their symbiotic algae (Hoegh-Guldberg 1999). Severe or prolonged Fisheries are in many cases at risk due to high fishing pressure, bleaching events are often followed by disease outbreaks and can with many commercially caught species already showing signs of cause coral mortality (Eakin et al. 2010). Bleaching events on a overexploitation. Climate change, by locally limiting productivity, large scale (“mass bleaching”) have been linked to unusually has the potential to further aggravate this situation. While sustain- high sea-surface temperatures, which exceed the temperature able fisheries management can significantly reduce the risks of threshold of affected species. Other factors exerting stresses on fisheries collapse, the uncertainty of climate impacts adds to the coral reef systems which have been identified as causes for coral challenge of establishing the quantity of fish that can be caught bleaching include pollution, overfishing, and the related shift in at sustainable levels (maximum sustainable yield). species composition (De’ath et al. 2012). A prolonged period of unusually high sea-surface temperatures 3.4.6.2 Coral Reefs across the Caribbean reefs for more than seven months in 2005 Coral reefs provide ecosystem services which are particularly caused the most extensive and most severe bleaching event recorded important at the local level for subsistence fisheries and tourism to date. Analyzing the concurrently high hurricane season during sector income (Hoegh-Guldberg et al. 2007). Healthy coral reefs that period, Trenberth and Shea (2006) found that about half of the also help to dampen the impact of coastal storm surges through observed sea-surface temperature anomaly was linked to global the reduction of wave energy (Villanoy et al. 2012). In the face warming. Following the warming events, bleaching continued in of rising sea-surface temperatures and declining pH-levels (see 2006 and was accompanied by disease and mortality. Mortality Section 2.3.8) as well as in concert with local stressors such as reached 50 percent in a number of locations, with the strongest pollution, coral reefs and the services they provide are particularly effects recorded in the northern and central Lesser Antilles and vulnerable to climate change. less severe cases in the waters of Venezuela. 78 LATI N AME R I CA A ND THE CA RIBBEA N Hurricanes, while posing a direct threat to coral reef structures, The study from Meissner et al. (2012) is based on a single Earth have also been found to cool the surrounding waters and thereby System Model. Van Hooidonk et al. (2013) used a large ensemble reduce the warming signal and the risk of severe bleaching (Eakin of climate models to analyze the onset of bleaching conditions et al. 2010). The effect of hurricanes on coral reefs may thus be for different emissions scenarios. With warming leading to a 2°C positive in the sense that vertical mixing and upwelling caused world, the median year in which bleaching events start to occur by tropical cyclones may reduce heat stress for coral reefs. This annually is 2046. While this median applies to most regions within effect was reconstructed for the 2005 anomaly, for which Carrigan the Caribbean, some parts experience bleaching 5–15 years ear- and Puotinen (2014) found that nearly 75 percent of the assessed lier. These include the northern coast of Venezuela and Colombia area experienced cooling from tropical cyclones. They estimated as well as the coast of Panama. With warming leading to a 4°C that this lead to around a quarter of reefs not experiencing stresses world, the median year in which annual bleaching starts to occur above critical thresholds, outweighing the negative effects of direct is 2040 (with no earlier onset in the Caribbean region). Generally, damage (e.g., through breakage). While they pointed out that the the reefs in the northern waters of the Caribbean Sea appear to be relatively frequent occurrence of tropical cyclones may have sup- less sensitive than those in the south. However, as Caldeira (2013) ported the development of relatively resistant coral reef species, it points out, those reefs at the higher latitude fringes of the tropical remains unclear whether such resistance will persist in the face coral range (both north and south of the tropics) are likely to be of a projected increase in the intensity and frequency of tropical more heavily affected by ocean acidification. cyclones, particularly as such developments would be concur- Buddemeier et al. (2011) computed coral losses in the Caribbean rent with changes to ocean chemistry deleterious to coral reefs. for three different scenarios, which would lead to a 2°C, 3°C, and Dove et al. (2013) point out that expected reductions in reef net 4°C world by 2100. Temperature trajectories diverge around 2050, calcification, associated with changes in ocean chemistry under a by which time warming reaches about 1.2°C. A comparison of high atmospheric CO2 concentration, will significantly hinder the all trajectories shows little difference between scenarios in terms recovery of coral reefs after damages related to extreme events. of coral cover. By 2020, live coral reef cover is projected to have halved from its initial state. By the year 2050, live coral cover is Projections of Climate Change Impacts on Coral Reefs less than five percent; in 2100, it is less than three percent, with Based on observations and laboratory experiments, thresholds no divergence among emissions scenarios. Notably, a 5–10 per- have been identified which enable the projection of risk of cent live coral cover is assumed as the threshold below which bleaching events in the future. The decrease in calcium carbon- the ecosystem no longer represents a coral reef (but is instead a ate saturation and concurrent pH levels poses another threat to shallow-water ecosystem that contains individual coral organ- reef-building corals (see Section 2.3.8). Taking into consideration isms). Assuming a scenario in which corals are able to adapt by both the projected decrease in the availability of calcium carbonate gaining an additional 1°C of heat tolerance, the loss of live coral and increase in sea-surface temperatures, Meissner et al. (2012) cover below five percent is prolonged by around 30 years. Bud- projected that most coral reef locations in the Caribbean sea and demeier et al. (2011) noted that results can be extrapolated for western Atlantic will be subject to a 60–80 percent probability of the wider Southeast Caribbean, albeit with the caveat that the annual bleaching events with 2°C warming by 2050, with areas at assumed high mortality rate of 50 percent was lower in regions the coast of Guyana, Suriname, and French Guiana being exposed outside the Virgin Islands. to a 100 percent probability. In contrast, under 1.5°C warming by The modeling projections of future bleaching events and loss 2050, most locations in the Caribbean sea have a comparably low of coral cover presented above are based on changes in marine risk of 20–40 percent probability of annual bleaching events, with chemistry and thermology. While the adaptive potential of coral the waters of Guyana, Suriname, French Guiana, and the north reef species remains uncertain, it should be noted that further Pacific being at slightly higher risk (up to 60 percent probability). impacts impeding the resilience of coral reefs are not included By the year 2100, almost all coral reef locations are expected in these future estimates. These include potential impacts that to be subject to severe bleaching events occurring on an annual are likely to change in frequency and/or magnitude under future basis in a 4°C world. Exceptions are major upwelling regions, warming, such as hurricanes and the variability of extreme tem- which experience a risk of 50 percent. Compared to impacts in the peratures. Taking these uncertainties into account, Buddemeier year 2050, the Caribbean sea experiences more locations under et al. (2011) concluded that the presented projections are likely risk in 2100 despite no significant further increase in emissions unduly optimistic, which leads the authors to predict that the and temperatures, highlighting the long-term impact of climate “highly diverse, viable reef communities in the Eastern Caribbean change on marine ecosystems even under emissions stabiliza- seem likely to disappear within the lifetime of a single human tion. A potentially limiting assumption made by Meissner et al. generation.” According to some estimates, a 90 percent loss of (2012) is that no changes in the frequency or amplitude of El Niño coral reef cover would lead to direct economic losses of $8.712 events is expected. billion (2008 value) (Vergara et al. 2009). 79 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Overall, while there are limitations to the projections of the leishmaniasis, and fascioliasis, and food- and water-borne diseases state of coral reefs in the future, a bleak picture emerges from the such as cholera and childhood diarrheal disease. Many of these available studies. Irrespective of the sensitivity threshold chosen, diseases have been found to be sensitive to changes in weather and indeed irrespective of the emissions scenario, by the year 2040 patterns brought about by ENSO. This indicates that disease Caribbean coral reefs are expected to experience annual bleaching transmission in LAC could prove highly responsive to changes in events. This is in accordance with Frieler et al. (2012) showing temperature and precipitation patterns induced by climate change. that, at the global scale, the global mean temperature at which Extreme weather events, including heat waves, hurricanes, floods, almost 90 percent of coral reefs are at risk of extinction is 1.5°C and landslides, also cause injuries and fatalities in Latin America, above pre-industrial levels. and these in turn can lead to outbreaks of disease. 3.4.6.3 Synthesis 3.4.7.1 Vector-Borne Diseases The rich fishery grounds of the Humboldt Current System in the Dengue fever is widespread in Latin America, with much of the Eastern Pacific react strongly to fluctuations in oceanic condi- region providing highly suitable climatic conditions to the primary tions related to the El Niño/Southern Oscillation (ENSO), during mosquito vector, Aedes aegypti. There has recently been a reemer- which the upwelling of nutrient-rich waters is suppressed by the gence and marked increase in the incidence of dengue fever and influx of warm surface waters. Together with ocean acidification dengue haemorrhagic fever in countries that had been declared and hypoxia, which are very likely to become more pronounced free of the illness following successful elimination programs in under high-emissions scenarios, the possibility of more extreme the 1950s and 1960s (Tapia-Conyer et al. 2009). El Niño events pose substantial risks to the world’s richest fishery Climate change is expected to play a contributing role in grounds. Irrespective of single events, the gradual warming of determining the incidence of the disease (Confalonieri et al. 2007), ocean waters has been observed and is further expected to affect although it is often difficult to separate the impact of climate fisheries particularly at a local scale. Generally, fish populations change from the impacts of urbanization and population mobil- are migrating poleward towards colder waters. Projections taking ity (Barclay 2008). In Brazil, the country with the largest number into consideration such responses indicate an increase of catch of cases in the world, a greater intensity of transmission of the potential by up to 100 percent in the south of Latin America. Off the disease has been observed during the hot, rainy months of the coast of Uruguay, the southern tip of Baja California and southern year (Teixeira et al. 2009). Between 2001–2009 in Rio de Janeiro, Brazil the maximum catch potential is projected to decrease by a 1°C increase in monthly minimum temperature was associated more than 50 percent. Caribbean waters may see declines in the with a 45 percent increase in dengue fever cases the following range of 5–50 percent, with the waters around Cuba, Haiti, the month, and a 10 mm increase in precipitation with a 6 percent Dominican Republic, and Puerto Rico, as well as Trinidad and increase (Gomes et al. 2012). Analysis from Mexico points to a Tobago, St Lucia, and Barbados, particularly severely affected. Along correlation between increases in the number of reported cases the coasts of Peru and Chile, fish catch is projected to decrease by and increases in rainfall, sea-surface temperature, and weekly up to 30°percent, but there are also increases towards the south. minimum temperature (Hurtado-Diaz et al. 2007). A study in It stands to reason that the communities directly affected by Puerto Rico, based on analysis of a 20-year period, likewise finds local decreases in maximum catch potential would not gain from a positive relationship between monthly changes in temperature catch potentials increasing elsewhere. They may also be the ones and precipitation and monthly changes in dengue transmission whose livelihoods would be most affected by the expected deleteri- (Johansson et al. 2009). Projections by Colon-Gonzalez et al ous effects of ocean acidification and warming on tropical coral (2013), holding all other factors constant, point to an upsurge in reefs. Irrespective of the sensitivity threshold chosen and indeed dengue incidence in Mexico of 12 percent by 2030, 22 percent by irrespective of the emissions scenario, by the year 2040 Caribbean 2050, and 33 percent by 2080 with a warming scenario leading to coral reefs are expected to experience annual bleaching events. a 3°C world in 2100; or 18 percent by 2030, 31 percent by 2050, While some species and particular locations appear to be more and 40 percent by 2080 with a warming scenario leading to a 4°C resilient to such events than others, it is clear that the marine world by 2100. In general, increases in minimum temperatures ecosystems of the Caribbean are facing large-scale changes with play the most decisive role in influencing dengue incidence, with far reaching consequences for associated livelihood activities as a sharp increase observed when minimum temperatures reach or well as for the coastal protection provided by healthy coral reefs. exceed 18°C (Colon-Gonzales et al. 2013). It appears, however, that climatic conditions alone cannot 3.4.7 Human Health account for rates of disease occurrence. Based on temperature-based mechanistic modeling for the period 1998–2011, Carbajo et al.(2012) The main human health risks in Latin America and the Caribbean found that temperature can estimate annual transmission risk include vector-borne diseases such as malaria, dengue fever, but cannot adequately explain the occurrence of the disease on a 80 LATI N AME R I CA A ND THE CA RIBBEA N national scale; geographic and demographic variables also appear the ENSO cycle and annual incidence of cutaneous leishmaniasis in to play a critical role. Colombia (Gomez et al. 2006) and visceral leishmaniasis in Brazil Malaria is endemic in Latin America, and rates of transmission (Franke et al. 2002). A study from Colombia of both types of the have increased over recent decades. This resurgence is associated disease also identified an increase in occurrence during El Niño and in part with local environmental changes in the region, such as a decrease during La Niña (Cardenas et al. 2006). These findings extensive deforestation in the Amazon basin (Moreno 2006). suggest that an increased frequency of drought conditions is likely Periodic epidemics have also been associated with the warm to increase the incidence of leishmaniasis (Cardenas et al. 2006). phases of ENSO (Arevalo-Herrera et al. 2012; Mantilla et al. 2009; Fascioliasis, a disease caused by flatworms and carried by snails Poveda et al. 2011). as an intermediate host, is a major human health problem in the It is possible that high temperatures could cause malaria to Andean countries of Bolivia, Peru, Chile, and Ecuador (Mas-Coma spread into high altitude cities (e.g., Quito, Mexico City) where 2005). Cases have also been reported in Argentina, Peru, Venezuela, it has not been seen for decades (Moreno 2006). Evidence shows Brazil, Mexico, Guatemala, and Cuba (Mas-Coma et al. 2014). The an increasing spread of malaria to higher elevations in northwest host infection incidence of fascioliasis is strongly dependent on Colombia during the last three decades due to rising temperatures, weather factors, including air temperature, rainfall, and/or potential indicating high risks under future warming (Siraj et al. 2014). evapotranspiration. Temperature increases associated with climate The connection between malaria and climate change, however, change may lead to higher infection and transmission rates and cause is unclear given the complexity of factors involved. Indeed, it is an expansion of the endemic zone, while increases in precipitation likely that the effect of climate change on malaria patterns will could, for example, increase the contamination risk window presently not be uniform. While incidence could increase in some areas, it linked to the November-April rainy season (Mas-Coma et al. 2009). is also possible that it may decrease in others—for example, in the Amazon, in Central America, and elsewhere where decreases in 3.4.7.2 Food- and Water-Borne Diseases precipitation are projected (Haines et al. 2006) (see Section 3.3.3, Cholera is transmitted primarily by fecal contamination of food Regional Precipitation Projections). and water supplies. Outbreaks are therefore often associated with Caminade et al. (2014) projected a lengthening of the malaria warm temperatures, flooding, and drought, all of which can aid transmission season in the highlands of Central America and contamination. Climatic variables have been shown to be decisive southern Brazil by the 2080s, but a shortening in the tropical in determining the extent of outbreaks (Koelle 2009). A recent regions of South America. This spatial differentiation is signifi- study of the relationship between rainfall and the dynamics of the cantly more pronounced with warming leading to a 4°C world cholera epidemic in Haiti, for example, shows a strong relationship than to a 2°C world. whereby increased rainfall is followed by increased cholera risk Earlier projections, however, offer mixed results. Béguin et al. 4–7 days later (Eisenberg et al. 2013). In South America, ENSO can (2011) projected an expansion of malarial area by 2050 in Brazil be a driving factor in cholera outbreaks in coastal areas because and isolated areas near the west coast of the continent under the El Niño phase provides warm estuarine waters with levels of approximately 2°C of global warming, although this is only if salinity, pH, and nutrients suitable for the blooming of the V. Chol- climatic and not socioeconomic changes are taken into account. erae pathogen (Martinez-Urtaza et al. 2008; Salazar-Lindo 2008). Van Lieshout et al. (2004), in contrast, found reductions in the Rates of childhood diarrheal disease have also been shown to size of the population exposed to malaria for at least three months be influenced by ambient temperature—and by ENSO in particular. of the year in all the scenarios they considered, and a reduction This was observed during the 1997–1998 El Niño event in Peru. in exposure to malaria for at least one month of the year in a 4°C During that particularly warm winter, in which ambient tempera- world (but not in a 3°C world). This study, meanwhile, projects tures reached more than 5°C above normal, hospital admissions for an expansion of the malarial zone southward beyond its current diarrheal disease among children increased by 200 percent over the southernmost distribution in South America—a finding consistent previous rate (Checkley et al. 2000). The relative risk of diarrheal with Caminade et al. (2014). disease in South America is expected to increase by 5–13 percent for Leishmaniasis is a skin disease carried by sandflies that takes the period 2010–39 with 1.3°C warming, and by 14–36 percent for the two main forms: cutaneous and visceral. Both are found through period 2070–99 with 3.1°C warming (Kolstad and Johansson 2011). much of the Americas from northern Argentina to southern Texas, 3.4.7.3 Impacts of Extreme Temperature Events excluding the Caribbean states (WHO 2014). A spatial analysis by Unusually high or low temperatures can potentially increase Valderrama-Ardilla et al. (2010) of a five-year outbreak of cutaneous morbidity and mortality, particularly in vulnerable groups such as leishmaniasis in Colombia beginning in 2003 identified temperature the elderly and the very young. A strong correlation has been found as a statistically significant variable. The study concluded, however, between unusually cold periods and excess deaths in Santiago, that climatic variables alone could not explain the spatial variation Chile, for example. In a time-series regression analysis, Muggeo of the disease. A positive association has been reported between 81 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL and Hajat (2009) estimated a 2.4 percent increase in all-cause to how changes in temperature and precipitation might affect the deaths among the above-65 age group for every 1°C decrease incidence of a particular disease in a particular location. below a cold threshold identified in their model. Cold-related risks Projections of how malaria incidence in LAC could be affected to human health would therefore be reduced if climate change by climate change over the rest of the century are somewhat results in a reduction in extreme cold events. inconsistent, with some studies pointing to increased incidence and Urban populations tend to be the most vulnerable to extreme others to decreased incidence. Such uncertainty also characterizes heat events due to the urban heat island effect, in which the studies of the relationship between climate change and malaria built environment amplifies temperatures. In northern Mexico, globally and reflects the complexity of the environmental factors heat waves have been correlated with increases in mortality rates influencing the disease. Little quantitative data is available on the (Mata and Nobre 2006); in Buenos Aires, 10 percent of summer future impacts of extreme weather events on human health, although deaths are associated with heat strain (de Garin and Bejaran studies based on historical data, such as that of Muggeo and Hajat 2003). Excessive heat exposure can cause or exacerbate a range (2009), have revealed a link between extreme temperatures and of health conditions, including dehydration, kidney disease, and increased rates of mortality in vulnerable sub-populations. cardiovascular and respiratory illnesses (Kjellstrom et al. 2010). Increased rates of hospital admissions of kidney disease patients 3.4.8 Migration have been documented during heat waves (Kjellstrom et al. 2010). While migration is not a new phenomenon in the region, it is Heat stress has been identified as a particular danger for work- expected to accelerate under climate change. There are many ers in Central America and one that coincides with high rates of areas in LAC prone to extreme events, including droughts, floods, kidney disease in some populations (Kjellstrom and Crowe 2011). landslides, and tropical cyclones, all of which can induce migra- 3.4.7.4 Impacts of Flooding and Landslides tion. Faced with severe impacts, migration might seem like the Torrential rain and resulting floods are among the main natural only option for finding alternative livelihoods (Andersen et al. hazards in the region and cause widespread injury and loss of 2010). However, migration typically causes an economic strain on life, livelihood, and property (Mata and Nobre 2006). Catastrophic both internal and external migrants (Raleigh et al. 2008), and not flooding has affected Mexico, Venezuela, Colombia, Brazil, Chile, everyone whose livelihood is threatened can afford to migrate. Argentina, and Uruguay in recent years (WHO/WMO 2012). Flood- The very poorest who do not have the necessary resources to ing can have multiple indirect health impacts, including the spread migrate can get trapped in a situation of ever-increasing poverty. of water-borne disease through water supply contamination and The transition from temporary to permanent migration resulting via the creation of stagnant pools that serve as habitats for disease from climate events can be facilitated by the existence of strong vectors such as malaria and dengue mosquitoes. Landslides and migration ties and networks (e.g., between LAC and the United mudslides can also be a consequence of flooding; these tend to States) (Deprez 2010). There are also important pull factors, which be exacerbated by factors such as deforestation and poor urban have been a major determinant of emigration in the past, including planning. Flash floods and landslides are a particular danger for more and better paid job opportunities and better access to services. informal settlements located on steep slopes and on alluvial plains They incentivize migration, particularly to North America or to (Hardoy and Pandiella 2009). countries within LAC with stronger economies. The question to Glacial lake outburst floods (see Box 3.4) also pose a risk to be assessed over time is whether climate change will make push populations located in the Andean region (Carey et al. 2012). The factors more important than pull factors. historical impacts of glacial lake outburst floods in Peru’s Cordillera The scientific literature on the interaction between migration and Blanca mountain range illustrate the potential for catastrophic loss of climate change is limited in terms of future projections. There is, life during periods of glacial retreat; many thousands of deaths have however, a growing body of literature on the demographic, economic, resulted from flooding, most notably in incidents in 1941, 1945, and and social processes of the interactions of climate and migration 1950 (Carey et al. 2012). The projection of further glacial melting in (Piguet et al. 2011; Tamer and Jäger 2010). Migration is considered the Andes (see Section 3.4.1, Glacial Retreat and Snowpack Changes) an adaptive response to maintain livelihoods under conditions of means that flooding continues to pose a risk to human populations. change. Assunção and Feres (2009) show that an increase in poverty levels by 3.2 percent through changes in agricultural productivity 3.4.7.5 Synthesis induced by regional warming of 1.5°C in 2030–2049 is reduced to The literature on the potential climate change impacts on human two percent if sectoral and geographic labor mobility is allowed health in the LAC region shows increased risks of morbidity for. This means that migration can reduce the potential impact of and mortality caused by infectious disease and extreme weather climate change on poverty (Andersen et al. 2010). events. Observed patterns of disease transmission associated with The projections of environmentally-induced migration agree different parts of the ENSO cycle seem to offer valuable clues as that most of the movement is likely to occur within the same 82 LATI N AME R I CA A ND THE CA RIBBEA N country or region (Deprez 2010). The largest trend in migration events to consider permanent domestic, regional, or international continues to be major movements from rural areas to urban areas. migration (ECLAC 2001). In 1998, Hurricane Mitch affected sev- Given the well-established migration channels between most LAC eral Central American countries and displaced up to two million countries and the United States, however, the impacts of climate people either temporarily or permanently. The impact was highly change may increase South-North migration flows. differentiated by country, with much lower displacement rates in There are no official statistics to show how many migrants Belize compared to Nicaragua, Honduras, and El Salvador, and a in LAC are moving in response to climate-related or other envi- 300 percent increase in international emigration from Honduras ronmental factors (Andersen et al. 2010). Although functioning (Glantz and Jamieson 2000; McLeman and Hunter 2011). Although as an adaptive strategy, environmentally-induced migration has the number of migrants has decreased over time, it has so far strong negative impacts on transitory areas and final destinations. remained above the level prior to the hurricane (McLeman 2011). For example, the International Organization for Migration reports that “rapid and unplanned urbanization has serious implications 3.4.8.3 Exacerbating Factors for urban welfare and urban services”, particularly in cities with The largest level of climate migration is likely to occur in areas “limited infrastructure and absorption capacity” (IOM 2009). For where non-environmental factors (e.g., poor governance, political areas of origin, the general consensus for LAC seems to be that the persecution, population pressures, and poverty) are already present impact of environmentally induced migration is overwhelmingly and already putting migratory pressures on local populations. In addi- negative (Deprez 2010). tion, poverty and an unequal geographical population distribution heighten people’s vulnerability to biophysical climate change impacts, 3.4.8.1 Drought thus compounding the potential for further migration (Deprez 2010). The effects of drought on migration have not been fully researched. Perch-Nielsen et al. (2008) explained that drought is the “most 3.4.8.4 Social Effects of Climate-Induced Migration complex and least understood natural hazard,” and that there Similar to traditional migrants, climate migrants with more edu- are a number of adaptive measures households might take before cation and skills are able to benefit the most from migration. resorting to migration. Nevertheless, scenarios based on projec- Benefits to migrants and their families can include, for example, tions of Mexico–United States migration rates (Feng et al. 2010) the possibility of finding better jobs. But migration can also have and of Brazilian internal migration (Barbieri et al. 2010) suggest a strong negative social impact on those who stay behind, par- that drought will lead to increased emigration along established ticularly on the poorest who typically do not have the resources migration routes and the depopulation of rural areas (Faist and to migrate and therefore risk being trapped in an adverse situation Schade 2013). Other examples of drought-induced migration with limited coping strategies (Andersen et al. 2010). In addition, include the flow of migrants from Brazil’s and Argentina’s north- climate-change-induced labor migration can have implications east regions to the state capitals and to the south-central regions on families left behind (e.g., challenges to children resulting (Andersen et al. 2010). Examples indicate that drought-induced from being raised in single parent homes with limited economic migration is already occurring in some regions. In Northeastern resources). In addition, climate change may induce greater levels Brazil, a primarily agricultural region, spikes in the rate of migra- of female migration; in the context of gender-based discrimination, tion to rapidly growing coastal cities or to the country’s central these women may face more challenges settling down and finding and southern regions have been observed following decreases in adequate housing and stable jobs (Deprez 2010). crop yields in years of severe drought (Bogardi 2008). Barbieri Labor migration can also provide benefits to migrants and et al. (2010) projected emigration rates in Brazil from rural areas their families. Migration can generate an increase in a family’s and found that depopulation is expected to occur—especially with financial assets, as work in the new location often pays better. increasing temperatures. This study, meanwhile, finds the biggest This contributes to better living conditions if the family is able to increase in migration coming from productive agricultural areas migrate together or generates remittances that can be sent back that support a large labor force. to help the family left behind. Despite some benefits, climate migrants face significant risks. 3.4.8.2 Sea-Level Rise and Hurricanes For example, the cost of migration (including travel, food, and Projections considering the impacts of sea-level rise on migration housing) can be very high and result in a worse financial situa- in Latin America and the Caribbean are sparse (Deprez 2010). tion for the family. There is also evidence that migrants’ working There is more research, however, on the impact of hurricanes. and housing standards can, in some cases, be very poor (such Although projections posit that migration resulting from hurricanes as in marginalized areas, informal settlements, and slums) with will continue to be mostly temporal and internal (Andersen et al. possible negative effects on health (Andersen et al. 2010). Further, 2010), stronger hurricane impacts in the Caribbean will increas- migrants who do not have networks or social capital in their new ingly drive households that have repeatedly suffered from these location can be socially isolated or discriminated against, resulting 83 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Several countries in the region have also faced political Box 3.12: Distress Migration during instability in the past few years. For example, Bolivia has seen Hurricane Mitch some internal social movements calling for the independence of some regions; and there are important challenges related to the A typical example of distress migration took place when Hurricane drug trade which has become increasingly violent, particularly in Mitch struck Central America in 1998. Honduras evacuated 45,000 Mexico (Necco Carlomagno 2012). In fact, the activities of criminal citizens from Bay Island. The government of Belize issued a red alert groups and organized crime syndicates in countries like Brazil and asked citizens on offshore islands to leave for the mainland. and Mexico are a major source of some of the most significant Much of Belize City was evacuated. Guatemala issued a red alert conflicts (Rubin 2011). as well. By the time Mitch made landfall those evacuated along Climate change could aggravate these situations, further the western Caribbean coastline included 100,000 in Honduras, increasing conflicts over the use of resources. Socioeconomic 10,000 in Guatemala, and 20,000 in the Mexican state of Quintana disparities could be exacerbated, and in the worst case scenario, Roo. Despite this, nearly 11,000 people were killed and more than government capacities could be insufficient to face these along 11,000 were still missing by the end of 1998. In all, 2.7 million were left homeless or missing. The flooding caused damage estimated with natural disasters and climate-related challenges (McLeman at over $5 billion (1998 dollars; $6.5 billion in 2008 dollars). Source: 2011). In some Latin American countries where criminal organi- Andersen et al. (2010). zations already have significant power, such security gaps can enable them to increase their influence, further weakening the capacity of the state (Carius and Maas 2009). It is important to note that in the past, environmental degradation has often in tensions or conflict. In addition, ties to families and networks been used as a pretext for conflicts that are in fact caused by in the communities of origin may deteriorate while they are away underlying ethnic tensions and injustices associated with an (Andersen et al. 2010). unequal geographical distribution of the population and income In the case of climate-related evacuations, the social effects inequality (Deprez 2010). are mostly negative (see Box 3.12: Distress Migration during Hur- In this context, Rubin (2011) suggests four ways in which ricane Mitch). These include serious damage to physical assets climate change could increase the risk of conflict in LAC: (e.g., housing and livestock), and to other natural resources in the community of origin. In many cases, natural disasters can also • More resource scarcity. Climate change is likely to exacerbate contribute to financial and health problems (Andersen et al. 2010). resource scarcities. Increasing scarcity of food, water, forests, Migration contributes strongly to structural and sociodemo- energy, and land could intensify competition over the remaining graphic change in LAC cities. Migrants tend to come from similar resources, triggering internal unrest and even border conflicts. locations and settle in the same areas which usually are marginal • More migration. The LAC region has important migration areas in urban areas where they might have social capital or dynamics that are being exacerbated by climate change as social networks (Vignoli 2012). This contributes to creating social households face increased resource scarcities, rising sea levels, vulnerability to climate change by increasing spatial segregation and more (and more intense) natural disasters. Larger flows at the destination, or by modifying social networks of migrant of migrants could potentially destabilize destination countries. households in their origin (Pinto da Cunha 2011; Vignoli 2012). As a result, immigrant populations may lack the knowledge • Increasing instability. Climate change and variability may of disaster risk management plans, especially if they are new undermine the capacity of the state by increasing the cost of to urban areas and had not encountered disaster management infrastructure in remote rural areas, limiting the reach of the plans in the rural areas where they migrated from (Adamo 2013). state. This could be aggravated by the rising costs of disaster management (e.g., an increase in the level of agricultural 3.4.9 Human Security subsidies needed to maintain adequate food production) as well as by the general need for increased adaptation spending. The LAC region is considered to be at low risk of armed conflict, These limits in state capacity might result in a weakening of with the incidence of armed conflicts declining substantially in the relationship between the state and its citizens. the last 15 years (Rubin 2011). With the downfall of many military • Increasing frequency and intensity of natural disasters. The regimes in the 1980s, and continued economic integration, the chaotic conditions that follow in the wake of natural disasters region has achieved relative stability (Rubin 2011). However, in may provide opportunities for rebel groups to challenge the the context of high social and economic inequalities and migration government’s authority. flows across countries, disputes regarding access to resources, The empirical literature is inconclusive regarding the linkages land, and wealth are persistent. between climate change and an increasing risk of conflict globally. 84 LATI N AME R I CA A ND THE CA RIBBEA N Rubin (2011) suggests, however, that while resource scarcity in isola- 3.4.10 Coastal Infrastructure tion from other socioeconomic factors does not necessarily increase the risk of conflict, it often acts as a catalyst or driver, amplifying the Coastal areas, infrastructure, and cities are all vulnerable to cli- existing (often traditional) causes of conflict. In this sense, Haldén mate change. This is particularly true for the Caribbean region (2007) notes that disparities in standards of living and income can due to its low-lying areas and the population’s dependence on be problematic for several reasons: (1) disparities and divisions coastal and marine economic activities (Bishop and Payne 2012). might by themselves impede growth and undermine adaptation Tropical cyclones and sea-level rise represent the main risks as strategies; (2) substantial inequality might also destabilize societ- their combination can severely affect economic development (and ies and increase the risk of conflict in the light of climate change have generated significant losses and damages in past decades). and variability; and (3) the differences between large segments of For example, category 5 Tropical Cyclone Ike generated approxi- the populations imply that climate change will have very unequal mately $19 billion in damages, including $7.3 billion in Cuba impacts on the population, further exacerbating tensions. This is alone (Brown et al. 2010). particularly relevant in the LAC region, which has one of the most Overall losses induced by climate change stressors such as unequal income distributions in the world (Fereira et al. 2013). increased wind speed, storm surge, and coastal flooding could Large inequalities among groups (differentiated along ethnic, amount to 6 percent of GDP in some Caribbean countries (CCRIF religious, political, or geographical lines) increase the risk of violent 2010). Climate change-related impacts, including local sea-level conflict and high individual income inequality is a driver of crime rise, increased hurricane intensity, and modified precipitation and (Dahlberg and Gustavsson 2005; Fajnzylber et al. 2002; Østby 2007). temperature patterns could increase current economic losses by In the case of Bolivia, for example, highly unequal distributions of 33–50 percent by the 2030s (CCRIF 2010). natural resources create tensions among regions. In response to the nationalization of gas and oil reserves, the resource-rich regions 3.4.10.1 Impacts of Sea-Level Rise on Coastal Cities of Santa Cruz, Tarija, Beni, and Pando (comprising 35 percent of Several studies (Brecht et al. 2012; Hallegatte et al. 2013; Hanson the Bolivian population) unsuccessfully sought autonomy. Violent et al. 2011) have recently estimated the potential costs of sea-level disputes erupted between the government and the regions seeking rise, and the modification of storm patterns and land subsidence autonomy (Rubin 2011); this has since subsided due to the 2010 (which is not induced by climate change), for coastal cities in LAC. Autonomies Law and other agreements. Hallegatte et al. (2013) found that, by 2050, coastal flooding could Relatively small populations can have a tremendous impact on generate approximately $940 million of mean annual losses in the the environment (Hoffman and Grigera 2013). There are examples– 22 largest coastal cities in the region with a sea-level rise of 20 cm, notably in the Amazon basin—where the rural poor have turned and about $1.2 billion with a sea-level rise of 40 cm (Table 3.8). to illicit extractive activities (e.g., illegal logging) because they The study likely underestimates the overall impact as it only lack legal or formal alternatives. The effects of climate change assesses the costs of climate change on the largest coastal cities. and environmental degradation, along with the rapid growth of the extractive industry, is expected to take the greatest toll on the 3.4.10.2 Impacts on Port Infrastructure most vulnerable—small-hold farmers, indigenous populations, and Port infrastructures are particularly vulnerable to the direct and the poor (Hoffman and Grigera 2013). The resulting increase in indirect consequences of climate change (Becker et al. 2013). resource competition, (particularly for water and land)—together Becker et al. (2013) identified sea-level rise, higher storm surges, with increasing market pressure on landholders with tenuous legal river floods, and droughts as the main direct impacts, and coastal tenure—is expected to exacerbate existing inequities and tensions erosion, which could undermine port buildings and construction, surrounding the proper and equitable allocation of the region’s as one of the indirect impacts of climate change. The potential natural wealth (Hoffman and Grigera 2013). increase in tropical cyclone intensity may increase ships’ port Climate change could also increase violence in small com- downtime and, therefore, increase shipping costs (Chhetri et al. munal or household settings. One example is gender-based vio- 2013; Esteban et al. 2012). lence, which is already widespread in Latin America (Morrison Port infrastructure is crucial to economic development as et al. 2004). Although there is little research on this topic, there international trade is principally channeled through ports. Fur- are some studies (e.g., Harris and Hawrylyshyn 2012) which thermore, in Caribbean countries, port infrastructure plays a very indicate that climate change (by transforming livelihoods and significant role as they often are the only vector for trade in goods social structures) could spur social violence in non-conflict and assets (Bishop and Payne 2012). Impacts on seaports will situations. Moser and Rogers (2005) showed how rapid socio- also have indirect consequences on local economies as import economic changes might have a destabilizing effect not only on disruptions generate price increases for imported goods and societies but also within families, leading to an increased risk export disruptions lead to revenues and incomes decreasing at of domestic violence. the national level (Becker et al. 2012). 85 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Table 3.8: Projected losses from sea-level rise under two different sea-level-rise scenarios and land subsidence in the largest LAC cities. 20 CM SEA-LEVEL RISE AND SUBSIDENCE (NO 40 CM SEA-LEVEL RISE AND SUBSIDENCE ADAPTATION) (NO ADAPTATION) MEAN INCREASE DUE MEAN INCREASE DUE TO TO SLR AND SUBSIDENCE SLR AND SUBSIDENCE MEAN ANNUAL COMPARED TO CURRENT MEAN ANNUAL COMPARED TO CURRENT URBAN AGGLOMERATION LOSS (M$) LOSSES LOSS (M$) LOSSES La Habana (Cuba) 9 5939% 21 13660% Port-au-Prince (Haiti) 8 1090% 11 1482% San Juan (Puerto Rico) 1.680 2365% 4.238 6118% Santo Domingo (Dominican 263 1166% 410 1880% Republic) Baixada Santista (Brazil) 274 3041% 467 5256% Barranquilla (Colombia) 87 1782% 102 2106% Belém (Brazil) 93 698% 586 4955% Buenos Aires (Argentina) 161 268% 592 1257% Panama City (Panama) 431 916% 451 962% Fortaleza (Brazil) 52 2762% 108 5814% Grande Vitória (Brazil) 2.643 1289% 10.096 5208% Guayaquil (Ecuador) 31.288 1012% 32.267 1047% Lima (Peru) 39 1009% 48 1254% Maceió (Brazil) 54 887% 283 5025% Maracaibo (Venezuela) 67 1086% 588 10238% Montevideo (Uruguay) 50 258% 180 1181% Natal (Brazil) 150 1505% 487 5100% Porto Alegre (Brazil) 71 641% 483 4918% Recife (Brazil) 259 1279% 970 5063% Rio de Janeiro (Brazil) 411 1088% 1.803 5108% Salvador (Brazil) 245 4903% 262 5248% San Jose (Costa Rica) 10 551% 67 4133% Total 2769.6 6164.4 Source: Hallegatte et al. (2013). 3.4.10.3 Impacts on Tourism Activities Beach tourism is particularly exposed to several direct and Tourism in the region, especially beach tourism in the Caribbean, is indirect climate change stressors, including sea-level rise, modi- projected to be affected by the impacts of climate change (Hyman fied tropical storm pattern, heightened storm surges, and coastal 2013). The total contribution of travel and tourism in the Caribbean erosion (Simpson et al. 2011). In a study comparing the vulner- was about 14 percent of the regional GDP and directly supported ability of four different tourist destinations in Jamaica, Hyman approximately 650,000 jobs (World Travel and Tourism Council 2013). (2013) found that coastal tourist resorts are two-to-three times As a result, the impact of climate change on tourism could detrimen- more exposed to climate change-related stressors than inland tally affect regional economic development (Simpson et al. 2011, 2010). touristic resorts. 86 LATI N AME R I CA A ND THE CA RIBBEA N 3.4.10.4 Impacts of Tropical Cyclones into account GDP and population projections but not potential Although projections on tropical cyclone frequency are still uncertain, adaptation measures. they indicate an augmentation of the number of Category 4 and 5 3.4.11 Energy Systems high-intensity tropical cyclone on the Saffir-Simpson scale (see Chap- ter 3.3.6, Tropical Cyclones/Hurricanes). The losses and damages Energy access is a key requirement for development, as many associated with tropical cyclones making landfall are also projected economic activities depend on reliable electricity access (Akpan to change (Hallegatte 2007; Mendelsohn et al. 2011). Quantifying the et al. 2013). At the individual and household level, electricity future impact of tropical cyclones and their associated costs is complex access enables income-generating activities, increases safety, and as it involves not only climate model projections but also projections contributes to human development (Deichmann et al. 2011). In of socioeconomic conditions and potential adaptation measures. 35 36 37 LAC, the population generally has extended access to electricity in In a scenario leading to a 4°C world and featuring a 0.89–1.4 m rural and urban areas (apart from Haiti, where only 12 percent of sea-level rise, tropical cyclones in the Caribbean alone could gen- the rural population and 54 percent of the urban population had erate an extra $22 billion and $46 billion in storm and infrastruc- access to electricity in 2010) (World Bank 2013z). ture damages and tourism losses by 2050 and 2100, respectively, Climate change is projected to affect electricity production and compared to a scenario leading to a 2°C world (Bueno et al. distribution both globally and in the region (Sieber 2013). This 2008). The sea-level rise assumed in this study is based on challenges the LAC countries, which will have to increase or at semi-empirical sea-level rise projections (Rahmstorf 2007) and is least maintain electricity production at the current level to support higher than the upper bound projected in Section 3.3.7, Regional economic development and growing populations. Sea-level Rise. Curry et al. (2009) project that cumulative losses The effects of extreme weather events and climate change induced by tropical cyclones in the Caribbean, Central America, could lead to price increases and/or power outages (Ward 2013). and Mexico are going to increase to about $110 and $114 billion Thermal electricity and hydroelectricity are projected to be most during the period 2020–2025. These numbers assume an increasing vulnerable. Three types of climate-change-related stressors could tropical cyclone intensity of 2 percent and 5 percent, respectively, potentially affect thermal power generation and hydropower gen- compared to average values from 1995–2006. The majority of the eration: Increased air temperature (which would reduce thermal costs, approximately $79 billion in cumulative losses, would incur conversion efficiency); decreased available volume and increased in Mexico (Table 3.9). The estimates of Curry et al. (2009) take temperature of cooling water; and extreme weather events (which could affect the production plants, the distribution systems, and grid reliability) (Han et al. 2009; Sieber 2013). Table 3.9: Cumulative loss for the period 2020–2025 for Latin American and Caribbean sub-regions exposed to tropical 3.4.11.1 Current Exposure of the cyclones under scenarios A1 (constant frequency, intensity LAC’s Energy Systems increased by 2 percent) and A2 (constant frequency, intensity LAC countries have a diverse energy mix (Table 3.10). The majority of increased by 5 percent). the South American countries heavily rely on hydroelectricity(almost 100 percent, for example, in Paraguay); Central American countries SCENARIO A1 SCENARIO A2 use thermal electric sources and hydroelectricity. Caribbean coun- SUB-REGION (IN MILLION US$) (IN MILLION US$) tries, meanwhile, rely on thermal electric sources for electricity Mexico 79.665 79.665 production. Between 91 percent (for Jamaica) and 55 percent (for Central America Cuba) of the electricity consumed is generated from these sources. and Yucatan35 5.128 5.847 With a projected change in water availability, from decreas- Greater Antilles 36 22.771 26.041 ing precipitation and river runoff and/or increasing seasonality Lesser Antilles37 1.813 2.073 and shrinking snow caps and decreasing snow fall in the Latin Bahamas, The 985 1.241 American mountainous regions, thermal electricity plant cooling systems may become less efficient and electricity production Total 110.362 114.867 could be affected (Mika 2013; Sieber 2013). Hydroelectric power The data and calculations are based on Curry et al. (2009). Please note generation is similarly affected (Hamududu and Killingtveit 2012). that the scenarios named here ‘A1’ and ‘A2’ are not SRES scenarios but based on Emanuel (2005) and Webster et al. (2005). 3.4.11.2 Impacts of Climate Change on Energy Supply There are limited studies specifically quantifying the impacts of 35 Belize, Costa Rica, El Salvador, Guatemala, Honduras, and Nicaragua. 36 Cuba, Dominican Republic, Haiti, Jamaica, and Puerto Rico. climate change on thermal electricity and hydroelectricity genera- 37 Antigua and Barbuda, Barbados, Dominica, Grenada, St. Kitts and Nevis, St. Lucia, tion in LAC. As the larger share of the electricity produced in the and St. Vincent and the Grenadines. 87 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Table 3.10: Electricity production from hydroelectric and thermoelectric sources, including natural gas, oil, coal, and nuclear in 2011 in the Latin American and Caribbean countries. ELECTRICITY ELECTRICITY PRODUCTION FROM PRODUCTION FROM ELECTRICITY ELECTRICITY POWER HYDROELECTRIC THERMOELECTRIC PRODUCTION FROM CONSUMPTION SOURCES SOURCES OTHER SOURCES COUNTRY OR REGION (KWH PER CAPITA) (% OF TOTAL) (% OF TOTAL) (% OF TOTAL) Caribbean Cuba 1326.6 0.56 54.89 44.55 Dominican Republic 893.31 11.79 87.99 0.21 Haiti 32.49 16.71 78.97 4.32 Jamaica 1549.23 1.96 91.81 6.22 Trinidad and Tobago 6331.94 – 100 – Latin America Argentina 2967.39 24.36 73.97 1.66 Bolivia 623.37 32.50 64.10 3.41 Brazil 2437.96 80.55 12.77 6.68 Chile 3568.08 31.97 60.40 7.63 Colombia 1122.73 79.06 17.64 3.30 Costa Rica 1843.94 72.56 8.78 18.66 Ecuador 1192.28 54.93 42.27 2.79 El Salvador 829.57 34.64 34.06 31.30 Guatemala 539.08 39.84 33.10 27.07 Honduras 707.76 39.50 56.51 3.99 Nicaragua 521.58 11.61 65.99 22.41 Panama 1829.01 52.16 47.55 0.29 Paraguay 1228.19 100.00 0.00 0.00 Peru 1247.75 55.00 43.13 1.87 Uruguay 2810.12 62.64 28.10 9.26 Venezuela, RB 3312.68 68.55 31.45 0.00 Mexico 2091.69 12.26 84.13 3.62 Sources: World Bank (2013e, f, g, h, i, j). – means not available. region originates from hydropower, general impacts of climate would result in a decrease in annual power output of approxi- change on thermal electric generation plants are discussed in mately 10 percent, from 1540 gigawatt hours (GWh) to 1250 GWh Section 4.4.6, Energy Systems. (Vergara et al. 2007). Hamududu and Killingtveit (2012) found that production will Hydropower increase by 0.30 TWh (or 0.03 percent) in the Caribbean compared Hydropower produces the larger share of electricity in the LAC to 2005 production levels, and by 0.63 TWh (or 0.05 percent) in region (see Table 3.10). The core natural resource for hydroelec- South America, under 2°C global warming by the middle of the tricity is river runoff, which has to be inter- and intra-annually 21st century. Maurer et al. (2009) projected the impacts of climate stable to allow hydropower installations to produce electricity most change on the Rio Lempa Basin of Central America (flowing through efficiently (Hamududu and Killingtveit 2012; Mukheibir 2013). In Guatemala, Honduras, and El Salvador and into the Pacific—see Peru it is estimated that a 50 percent reduction in glacier runoff Table 3.11). They concluded that an increase in frequency of 88 LATI N AME R I CA A ND THE CA RIBBEA N Table 3.11: Projected temperature and hydrologic changes basin, average river flow could be between –20 and +18 percent in the Rio Lempa River during the period 2040–2069 and with a global warming of 2.1°C depending on the different GCM 2070–2099 relative to the period 1961–1990 for hydrological chosen. This difference between the lowest and the highest esti- change and pre-industrial levels for temperature changes. mates highlights the limitations of the current models to project the potential hydropower production from dams built on this river IMPACT/PERIOD SCENARIO 2040–2069 2070–2099 basin and reflects that projections vary from one GCM to another Temperature increase B1 +1.8°C +2.2°C (Nóbrega et al. 2011). Popescu et al. (2014) showed an increase above pre-industrial levels A2 +2.1°C +3.4°C in the maximum hydropower energy potential for the La Plata Basin of between 1–26 percent with a global warming of 1.8°C Precipitation change B1 – –5% relative to 1961–1990 by 2031–2050 (see Table 3.12). There were also great disparities (median change) A2 – –10.4% in the sub-basin projections depending on the model used. The Reservoir inflow relative B1 – –13% La Plata basin is one of the most economically important river to 1961–1990 (median basins in Latin America (being part of Argentina, Bolivia, Brazil, change) A2 – –24% Paraguay, and Uruguay). It has a major maximum hydropower Frequency of low flow B1 +22% +33% energy potential, producing on average 683,421 GWh per year relative to 1961–1990 (median change) A2 +31% +53% during the period 1991–2010 and 76 percent of the 97,800 MW total electricity generation capacity of the five countries in the La Source: Maurer et al. (2009). Plata basin (Popescu et al. 2014). The results of these studies, however, need to be interpreted with care. For example, the significant decrease in hydropower capacity at the micro-level as projected by Maurer et al. (2009) low-flows in scenarios leading to a 2°C world and a 3°C world strongly contrasts with the results of Hamududu and Killingtveit implies a proportional decrease in hydropower capacity for the (2012), who projected an increase in hydropower generation at two main large reservoirs used for hydroelectricity generation in El the macro-level. The Hamududu and Killingtveit study may be Salvador (Cerron Grande and 15 Setiembre). Low-flow frequency limited for several reasons. First, it does not take into account is a key indicator of the economic viability of hydropower infra- seasonality and the impacts of climate change on the timing of structures as it determines the firm power, which is the amount river flows. Second, changes in hydrology and temperatures are of “energy a hydropower facility is able to supply in dry years” accounted for at the country level but not at the river basin level; (Maurer et al. 2009). The projected increase in low-flow frequency this does not take into account potential spatial variability and could therefore reduce the economic return from the existing facil- changes occurring over short distances. Third, the study does not ity and reduce the return on investments in future hydroelectric consider the potential impacts of floods and droughts, which have infrastructures (Maurer et al. 2009). very significant impacts on hydroelectricity generation and are For Brazil, de Lucena et al.(2009) project that average annual projected to occur more frequently and with a greater intensity in river flows will decrease by 10.80 percent with 2.9°C global warm- the coming decades (Marengo et al. 2012, 2013; Vörösmarty et al. ing, and by 8.6 percent with 3.5°C global warming, by 2071–2100. 2002) (see also Section 3.4.2, Water Resources, Water Security, This decrease in annual flow will lead to a decrease in firm power and Floods). Finally, the study does not consider the impacts on of 3.2 and 1.6 percent, respectively, during this time period com- river runoff from decreasing snow cover and snowfall in the Latin pared to production level for 1971–2000. For the Rio Grande river American mountainous regions (Barnett et al. 2005; Rabatel et al. Table 3.12: Maximum hydropower energy potential for the La Plata Basin with present, near future, and end-of-century climate conditions for two climate models (PROMES-UCLM and RCA-SMHI). PRESENT CLIMATE FUTURE CLIMATE 2031–2050 END OF CENTURY 2079–2098 SCENARIO 1991–2010 (1.8°C IN SCENARIO A1B) (3.2°C IN SCENARIO A1B) ENERGY VARIATION TO ENERGY VARIATION TO ENERGY (GWH/YEAR) (GWH/YEAR) PRESENT (GWH/YEAR) PRESENT PROMES-UCLM 688,452 1.01 715,173 1.05 683,421 RCA-SMHI 861,214 1.26 838,587 1.23 Source: Popescu et al. (2014). 89 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL 2013; Vuille et al. 2008). These limitations could potentially explain Table 3.13: Climate change-related stressors projected to why their results are different from those of Maurer et al. (2009), affect hydroelectricity generation. who use monthly precipitation rates to calculate annual inflows. Further research is needed to adequately inform decision mak- CATEGORY OF CLIMATE- CHANGE-RELATED CLIMATE-CHANGE- ers in the region on the impacts of climate change on hydropower STRESSORS RELATED STRESSORS generation. Similarly to Hamududu and Killingtveit (2012), de Long term trends or gradual Reduction in average precipitation Lucena et al. (2009) only accounted for the average behavior of changes induced by climate Increase in average precipitation flows and did not integrate potential change in seasonality or the change effects of extreme dry or wet events on hydropower generation. In Increase in average temperature this context, projections for hydropower production in Brazil by Increase in extreme climate Drought de Lucena et al. (2009) may underestimate the potential impacts variability Flooding of climate change. The projections by Popescu et al. (2014) only Indirect climate change impacts Water scarcity estimated a maximum hydropower potential, but this does not mean that more hydropower electricity will be produced from Siltation through land degradation existing or future installations. For example, the specifications Source: Mukheibir (2013). for existing dams (e.g., reservoir size, dam height, and so forth) may not be sufficient to efficiently manage projected excess flows. Despite these uncertainties, there are some clear climate change impacts on hydropower. Mukheibir (2013) inventoried the Oil and Gas climate-related stressors that are projected to affect hydroelectric- Some LAC countries, such as the Republica Bolivariana de Ven- ity generation. He separated climate-change-related stressors into ezuela, Brazil, and Mexico, benefit from significant oil and/or three categories: (1) long-term trends or gradual changes induced gas reserves. For example, Venezuela is the world’s tenth largest by climate change; (2) increases in extreme climate variability; exporter of oil and Mexico was the ninth biggest producer in and (3) indirect climate change impacts (Table 3.13). These 2013 (EIA 2014a). The production of gas and oil in LAC countries stressors could potentially reduce firm energy and increase vari- contributed 7.38 percent in 2012 and 12.01 percent in 2013 to ability and uncertainty of supply in the energy sector (Ebinger global production (Table 3.14). Furthermore, some countries in and Vergara 2011). the region have a very significant share of their GDP originating Table 3.14: Natural gas production for LAC countries in 2012 and oil production in 2013. NATURAL GAS PRODUCTION IN 2012 OIL PRODUCTION IN 2013 (IN (IN BILLION CUBIC FEET) THOUSANDS OF BARRELS PER DAY) Argentina 1,557.39 707.91 Bolivia 652.27 64.46 Brazil 910.77 2,712.03 Chile 42.98 15.57 Colombia 1,110.30 1,028.47 Cuba 38.00 48.73 Ecuador 54.39 527.03 Mexico 1,684.42 2,907.83 Peru 639.91 174.96 Trinidad and Tobago 1,504.74 118.12 Venezuela, RB 2,682.81 2,489.24 Total LAC region 10,878.68 10,851.42 LAC percentage of global production 7.38% 12.01% The LAC countries not displayed in the table produce little oil or natural gas. Source: EIA (2014a; b). 90 LATI N AME R I CA A ND THE CA RIBBEA N from oil and natural gas rents (defined as the difference between Solar energy installations are subject to two types of climate- value of natural gas or oil production and the total cost of produc- related impacts: reduced insulation induced by cloudiness, which tion); in Venezuela and Trinidad and Tobago, for example, about decreases heat or electricity output, and extreme weather events 30 percent of 2012 GDP in 2012 came from oil and gas rents (World such as windstorms or hail, which could damage production units Bank 2013e; f). The assessment of climate change impacts on oil and their mounting structures (Arent et al. 2014). According to and gas presented here focuses on direct impacts and does not Arent et al. (2014), gradual climate change and extreme weather consider the possible decrease in fossil fuel assets value induced by events “do not pose particular constraints to the future deploy- future mitigation policies, which would contribute to a reduction ment of solar technologies.” Studies estimating projected impacts in fossil fuel demand at the global level (IPCC 2014d). of climate change and extreme weather events on solar energy Off-shore platforms and on-shore infrastructures are susceptible outputs are not available. to climate related impacts, such as sea-level rise and coastal ero- sion that could damage extraction, storage and refining facilities 3.4.11.3 Impacts of Tropical Cyclones (Dell and Pasteris 2010) and also to extreme weather events such on Power Outages as tropical cyclones which could lead to extraction and production The Caribbean and Central American regions are particularly disruption and platform evacuation (Cruz and Krausmann 2013). exposed to the impacts of tropical cyclones and a higher frequency For example, in the aftermath of tropical cyclones Katrina and of high-intensity tropical cyclones is projected (see Section 3.3.6, Rita in 2005, 109 oil platforms and five drilling rigs were damaged Tropical Cyclones/Hurricanes). Strong winds, heavy precipitation, leading to interruptions in production (Knabb et al. 2005). Cozzi and floods associated with tropical cyclones have the capacity to and Gül (2013) identify two key climate related risks for the LAC disrupt and even damage essential power generation and distribu- region: sea-level rise and an increase in storm activity (particularly tion infrastructures leading to power outages. A growing number of for Brazil). These risks would mainly lead to an increase in the studies have developed models to estimate and forecast the risks shutdown time of coastal refineries and an increase in offshore of power outages to energy systems in order to improve disaster platform costs, which will have to be more resistant to high-speed assistance and recovery (Cao et al. 2013; Han et al. 2009; Nateghi winds associated with tropical cyclones (Cozzi and Gül 2013). et al. 2013; Quiring et al. 2013). However, studies and models spe- cifically quantifying or taking into account the projected effect of Wind and Solar Energy climate change on tropical cyclones intensity and frequency and Solar and wind energy sources play an important role in climate the potential disruptions to power generation and distribution in change mitigation strategies to reduce global emissions from fossil the Caribbean and Central American countries are lacking. fuel combustion. Even though wind and solar energy still play a very minor role in LAC, significant development of the sector is 3.4.11.4 Effects of Climate Change projected (Bruckner et al. 2014). In this context, a more precise on Energy Demand understanding of the effect of climate change on these energy Climate change will also affect energy demand. Increasing sources is of great significance. temperatures and heat extremes (see Sections 3.3.1, Projected For wind energy, the main climate change impact relates to Temperature Changes, and 3.3.2, Heat Extremes) lead to a higher changing wind patterns and how climate change will affect inter- demand for air conditioning (Cozzi and Gül 2013); on the other and intra-annual variability and geographical distribution of wind hand, demand for heating may decrease. At the global level, Isaac (Arent et al. 2014). Despite significant progress, GCMs and RCMs and van Vuuren (2009) estimated that by 2100 in a 4°C world the still do not produce very precise projections for inter-annual, sea- number of cooling degree days will rise from 12,800 during the sonal, or diurnal wind variability (Arent et al. 2014). Furthermore, period 1971–1991 to 19,451, (a 51.9 percent increase) while demand specific studies estimating the effects of climate change on wind for heating (measured in heating degree days) was projected to patterns and therefore wind energy in LAC are still missing (Pryor remain almost constant. At the regional level, they project that and Barthelmie 2013). However, drawing on conclusions from Pryor by 2100 demand for heating is going to decrease by 34 percent and Barthelmie (2013) for the United States and Europe stating compared to the 1971–1991 period—from 364 to 240 heating that “generally, the magnitude of projected changes over Europe degree days. They project the demand for cooling to increase by and the contiguous USA are within the ‘conservative’ estimates 48 percent, from 1802 to 2679 cooling degree days. embedded within the Wind Turbine Design Standards,” it can be assumed that future climate change may not significantly affect 3.4.11.5 Synthesis The assessment of the current literature on climate change impacts wind energy supply. Pryor and Barthelmie (2013) nonetheless on energy in LAC shows that there are only a few studies, most of highlight the need for more research in this area to better quantify which make strong assumptions about key issues such as seasonality the effects of long-term climate change and extreme events on of water supply for hydropower. These studies are more qualitative wind energy supplies. 91 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL than quantitative, and important gaps remain. However, in general will add water stress to an area in which agriculture has already substantial climate change impacts can be expected for the energy shifted toward more water-intensive irrigated crop production since sector. There is also a lack of studies with respect to the impacts the 1990s (Bury et al. 2011). As the glacier reservoirs gradually of climate change impacts on renewable energies. In general, the disappear, however, runoff will tend to decrease (particularly in impacts of climate change on energy demand is less well studied the dry season). The peak in runoff is expected to be reached in than those on energy supply—and, yet, demand and supply interact about 20–50 years from now (Chevallier et al. 2011) if it has not in a dynamic way. For example, the concomitant increase in energy yet peaked already (Baraer et al. 2012). demand during heat extremes and the decrease of energy supply Changes to river flow translate into risks to stable water supplies through reduced river flow and low efficiencies may put existing for much of the region. Diminishing downriver water flow will energy systems under increasing pressure in the future. also undermine hydropower generation; crucial to the economic development of the continent (Hoffman and Grigera 2013) (see 3.5 Regional Development Narratives also Section 3.4.11). Current studies project that by 2050, up to 50 million people in the vast lowland area fed by Andean glacial In this section, implications of climate change for regional devel- melt, will be affected by the loss of dry season water for drink- opment are discussed in order to relate climate change impacts to ing, agriculture, sanitation, and hydropower (Cushing and Kopas existing and future vulnerabilities in the LAC region. The develop- 2011). Deforestation and land degradation can furthermore alter ment narratives are split into overarching development narratives the water cycle and possibly endanger water availability (Buytaert across the region and in sub-regional development narratives. It et al. 2006; Viviroli et al. 2011). is important to note that each development narratives presents Changes to the seasonal cycle of water availability affect the only one of the many possible ways in which climate change can ecosystems that rely on a stable water supply. Consequently eco- put key development trajectories at risk. Table 3.15 summarizes system services are put at risk. For example, freshwater fisheries the key climate change impacts under different warming levels may be exposed to climate-related risks as decreasing river flows in the Latin American and the Caribbean region and Figure 3.22 reduce the floodplain for spawning and the natural seasonal flood- summarizes the key sub-regional impacts. ing of lakes is reduced. A projected decrease in annual precipita- tion and increasing risk of drought will in turn increase the risk 3.5.1 Overarching Development Narratives of large-scale forest degradation, not only in the Amazon, with a loss of associated ecosystem services. 3.5.1.1 Changes to the Hydrological Cycle Endanger In the Andes in particular, water stress will reduce pasture the Stability of Freshwater Supplies and Ecosystem land availability in the dry season and increase the potential for Services conflict over land use (Kronik and Verner 2010). Social conflicts An altered hydrological system due to changing runoff, glacial melt, over water rights and water access may increase in the Peruvian and snowpack changes will affect the ecosystem services that the Andes between farming communities and mining companies. rural population depends on, freshwater provisioning in cities, Moreover, higher river flow peaks can lead to landslides and devas- and major economic activities such as mining and hydropower. tating floods associated with glacial lake outburst (Chevallier et al. Throughout the 20th century the tropical glaciers in the Central 2011)—with direct consequences for human lives and settlements. Andes have lost large amounts of their volume (see Section 3.4.1, Cities are highly vulnerable as continuous urbanization and Glacial Retreat and Snowpack Changes). As land surface tempera- population growth increases water demand (Hunt and Watkiss 2011) tures rise, this trend is expected to accelerate possibly leading to and as they depend on ecosystem services provided by surround- an almost complete deglaciation of 93–100 percent in a 4°C world. ing areas. The high Andean moorlands (known as páramos—see In concert with decreasing snowpack, changes to precipitation pat- Box 3.10, Critical Ecosystem Services of High Andean Mountain terns, and higher evaporation, increasing glacial melt will impact Ecosystems), which are key ecosystems able to stock large amounts the timing and magnitude of river flows. In general, runoff is of carbon on the ground and act as water regulators, are threatened projected to increase during the wet season, increasing flood risk by temperature rise, precipitation changes, and increasing human (see Section 3.4.2, Water Resources, Water Security, and Floods). activity. Major population centers, such as Bogota and Quito, rely Accelerated melting rates may lead to a localized short-term surge on páramo water as a significant supply source. The melting of in water supply that might lead to unsustainable dependency (Vuille the Andean glaciers, increasingly unpredictable seasonal rainfall 2013). For example, in the area known as Callejon de Huaylas in the patterns, and the overuse of underground reserves are affecting the central highlands of Peru, climate-change-induced glacier retreat urban centers of the highlands (e.g., La Paz, El Alto, and Cusco), 92 LATI N AME R I CA A ND THE CA RIBBEA N Figure 3.22: Sub-regional risks for development in Latin America and the Caribbean (LAC) under 4°C warming in 2100 compared to pre-industrial temperatures. Central America & the Caribbean Higher ENSO and tropical cyclone frequency, Dry Regions precipitation extremes, drought, and heat waves. Risks of reduced water availability, crop yields, food security, and coastal safety. Caribbean Poor exposed to landslides, coastal erosion Central America with risk of higher mortality rates and migration, negative impacts on GDP where share of coastal tourism is high. Amazon Rainforest Increase in extreme heat and aridity, risk of Amazon Rainforest forest fires, degradation, and biodiversity loss. Risk of rainforest turning into carbon source. Shifting agricultural zones may lead to conflict Dry Regions over land. Risks of species extinction threatening traditional livelihoods and cultural losses. Andes Andes Glacial melt, snow pack changes, risks of 3RSXODWLRQ'HQVLW\ flooding, and freshwater shortages. >3HRSOHSHUVTNP@ In high altitudes women, children, and Southern Cone indigenous people particularly vulnerable; and  agriculture at risk. In urban areas the poor living ² on steeper slopes more exposed to flooding. ² Dry Regions ² Increasing drought and extreme heat events ² Falkland Islands (Islas Malvinas) leading to cattle death, crop yield declines, and  A DISPUTE CONCERNING SOVEREIGNTY OVER THE ISLANDS EXISTS BETWEEN ARGENTINA WHICH CLAIMS challenges for freshwater resources. THIS SOVEREIGNTY AND THE U.K. WHICH ADMINISTERS THE ISLANDS. Risks of localized famines among remote indigenous communities, water-related health problems. Stress on resources may lead to Data sources: Center for International Earth Science Information Network, Columbia University; United conflict and urban migration. Nations Food and Agriculture Programme; and Centro Internacional de Agricultura Tropical—(2005). Southern Cone Gridded Population of the World, Version 3 (GPWv3): Population Count Grid. Palisades, NY: NASA Socio- economic Data and Applications Center (SEDAC). This map was reproduced by the Map Design Unit of Decreasing agricultural yields and pasture The World Bank. The boundaries, colors, denominations and any other information shown on this map productivity, northward migration of agro- do not imply, on the part of The World Bank Group, any judgment on the legal status of any territory, or ecozones. any endorsement or acceptance of such boundaries. Risks for nutritious status of the local poor. Risks for food price increases and cascading impacts beyond the region due to high export share of agriculture. 93 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL which rely to some extent on glacial melt for dry season water has increased by roughly three percent per annum since the supplies and are already facing dire shortages. The arid coastal early 1990s (IFPRI 2012). In LAC, large parts of agriculture are plain of Peru faces similar challenges. Water shortage has become rain-fed and therefore very vulnerable to climatic variations such a huge risk and a source of tension in Lima, which is dependent as droughts and changing precipitation patterns. Only 10.5 mil- on water from the Andes. In Santiago de Chile, meanwhile, an lion ha of agricultural area are irrigated, amounting to roughly estimated 40 percent reduction in precipitation will impact water 0.6 percent of the total agricultural area. Of those 10.5 million ha supplies in a city that is expecting a 30 percent population growth of irrigated area, 3.5 million are located in Brazil, amounting to by 2030 (Heinrichs and Krellenberg 2011). Quito is another city 1.3 percent of Brazil’s agricultural area (FAOSTAT 2013; Oliveira that will face water shortages as a result of glacier retreat (Hardoy et al. 2009). Changing precipitation patterns and extreme events and Pandiella 2009). could therefore affect important parts of the economy. Moreover, Freshwater in coastal areas is particularly exposed to the Hoffman and Grigera (2013) calculate that for the Amazon and risks associated with sea-level rise. Here, a substantial section of Cerrado regions, shifting rainfall patterns and temperature rises due the population is exposed to repeated flooding, contamination of to climate change will lead to more frequent droughts and forest groundwater by salt water, and constraints on the availability and fires in the dry season and floods in the rainy season, threatening quality of drinking water (Magrin et al. 2007). Low-income groups the growth of monoculture agribusiness and the livelihoods of who already lack adequate access to water will be even less able small-holder farmers and ranchers. likely to obtain it unless there is a considerable improvement in Besides the implications of climate change for large-scale the provision of basic services. agriculture, there is also evidence that climate change will strongly affect small-to-medium-scale agriculture and regional 3.5.1.2 Climate Change Places at Risk Both Large- food security as well as indigenous communities. This is par- Scale Agricultural Production for Export and Small- ticularly true for rural communities who heavily rely on sub- Scale Agriculture for Regional Food Production sistence farming and the urban poor who are most hard-hit by Latin America and the Caribbean is a climatically highly heteroge- rising food prices. A projected 15–50 percent decline in fishery neous region. As a result, agricultural production systems and their catch potential along the Caribbean coast, and by more than outputs differ greatly among climatic zones and countries, as will 50 percent off the Amazonas estuary and the Rio de la Plata the impacts of climate change on agricultural production. Despite (Cheung et al. 2010), together with widespread coral reef loss, the relatively low contribution of agriculture (10–12 percent) to further adds to the challenge of maintaining a healthy diet for the total GDP, agriculture plays a vital role for the LAC economy, the poorest in the region. Coral reef loss and more frequent with 30–40 percent of the labor force engaged in the agricultural extreme events could also affect the viability of the tourism sector (IAASTD 2009). However, the numbers and proportion of industry, with significant implications for livelihoods across the sector comprised by subsistence and commercial agriculture, different socio-economic groups. differ greatly among the LAC countries. Overall undernourishment in the region has decreased. In Climate change is expected to have different impacts over 1990, roughly 65 million people (14.6 percent of the population) different timeframes (see Section 3.4.3, Climate Change Impacts were undernourished; by 2012, the number decreased to 49 mil- on Agriculture). In the short run, expected changes in agricultural lion people (8.3 percent of the population) (FAO 2012a). The LAC outputs in the region are likely to be heterogeneous, with some countries most affected by undernourishment are Haiti, Bolivia, regions and crops seeing gains and others losses (Table 3.15). In Guatemala, Nicaragua, and Paraguay. In all five countries more the long run, however, larger reductions in agriculture are expected than 20 percent of the population are undernourished (FAO 2012a). with important impacts on livelihoods (Calvo 2013; Sanchez and However, population growth and changing nutritional patterns Soria 2008; Samaniego 2009) despite uncertainties with regard are expected to increase global food demand by 60 percent by to the importance of CO2 fertilization and potential adaptation. 2050 (FAO 2012a). As a result, increased agricultural production Overall, the potential risks estimated for the agricultural sector is essential to maintain the currently positive trend of decreasing in LAC are substantial, particularly during the second half of the undernourishment. The expected negative effects of climate change 21st century. on agriculture (Table 3.15) will make the challenge of achieving By putting agricultural production at risk, climate change food security in LAC all the more difficult. According to one study, threatens an important regional export. LAC plays a vital role without adaptation measures climate change is likely to stall the in global agriculture (IFPRI 2012). The two biggest exporters of projected decline in child undernourishment in LAC by 5 percent agricultural products in Latin America are Brazil and Argentina by 2050 (Nelson et al. 2009). This study does not take into account (Chaherli and Nash 2013). Agricultural production in LAC countries the impacts on food resources other than agricultural crops; it 94 LATI N AME R I CA A ND THE CA RIBBEA N therefore may underestimate the multidimensional impacts that urban populations, including social marginalization and limited climate change has on food security. access to resources. However, conditions such as dense and poorly There are potentially direct consequences of climate change on constructed housing in urban areas or high, direct dependence the levels of poverty and food security in the region. According to on ecosystem services among rural indigenous populations result an exploratory modeling study by Galindo et al. (2013) an average in specific vulnerability patterns for different population groups. decline of six percent in agricultural production due to climate Despite these differences in vulnerabilities, climate impacts also change by 2025 would result in 22.6 percent and 15.7 percent act along an urban-rural continuum. For example cities depend fewer people overcoming the $1.25 and $2 per day poverty lines on the surrounding landscape to provide ecosystem services and respectively, given losses in livelihoods. This means a total of the rural population benefits from remittances sent from urban to 6.7–8.6 million people who would remain under the poverty line rural areas. However, the effectiveness of remittances in support- as a result of climate change impacts on agriculture. In addition, ing adaptive capacity under rising impacts is open to question, as important indirect effects resulting from reductions in agricultural both demand on the receiving end and exposure to climate risks yields include risks to agro-industrial supply chains. Given the on the sending end are expected to rise. exploratory nature of this study, however, exact numbers have to The rural poor are likely to feel the impacts of climate change be interpreted with care. and variability most directly given their dependency on rain-fed agriculture and other environmental resources (e.g., forests and 3.5.1.3 A Stronger Prevalence of Extreme Events fish) which are particularly susceptible to the effects of climate Affects Both Rural and Urban Communities, change in general and extremes in particular. Moreover, these Particularly in Coastal Regions populations have limited political voice and are less able to A changing frequency and intensity of extreme events, such as leverage government support to help curb the effects of climate drought, heat extremes, tropical cyclones, and heavy precipitation, change (Prato and Longo 2012; Hardoy & Pandiella, 2009). Rural will have strong implications for the urban and rural populations poverty in the LAC region has declined considerably over the past of the region, with particular vulnerability patterns shaping the two decades—both in terms of the numbers of people who live in risks of different population groups. poverty and the rate of poverty among rural populations—with The LAC region is heavily exposed to the effects of strong many countries in the region showing positive trends both in ENSO events, including extreme precipitation and disastrous poverty reduction and in a better distribution of income. That flooding, especially in the Andes and Central America where said, many rural people in the region continue to live on less than steep terrains are common (IPCC 2012; Mata et al. 2001; Mimura $2 per day and have poor access to financial services, markets, et al. 2007; Poveda et al. 2001). Glacial lake outbursts present training, and other opportunities. There is a strong concentra- a further permanent hazard for Andean cities (Chevallier et al. tion of extreme poverty among landless farmers and indigenous 2011). Along the Caribbean and Central American coasts, tropical peoples, particularly among women and children; indeed, close cyclones and rising sea levels expose the population to storm to 60 percent of the population in extreme poverty live in rural surges and coastal inundation (Dilley et al. 2005; Woodruff et areas (RIMISP 2011). al. 2013). Although the scientific evidence is limited, there are Extreme events will also strongly impact the urban poor as studies indicating an increase of 80 percent in the frequency urban areas are also a focal point of climate change impacts from of the strongest category 4 and 5 Atlantic Tropical Cyclones extreme events (Vörösmarty et al. 2013). In 2010, the urban popu- (Bender et al. 2010; Knutson et al. 2013) and a doubling in the lation accounted for 78.8 percent of the total population (ECLAC frequency of extreme El Niño events above 20th century levels 2014). National economies, employment patterns, and government (Cai et al. 2014). The latter two projections are especially wor- capacities—many of which are highly centralized—are also very risome as they concur with a sea-level rise up to 110 cm (see dependent on large cities; this makes them extremely vulnerable Section 3.3.7, Regional Sea-level Rise). A poleward migration to the effects of extreme events (Hardoy and Pandiella 2009). of tropical cyclones as recently observed (Kossin et al. 2014) Urbanization in the region includes unplanned, haphazard could potentially lead to less damage to tropical coasts but expansion of cities (ONU-Habitat 2012) over floodplains, mountain countries would also benefit less from the water replenishment slopes, or areas prone to flooding or affected by seasonal storms, that cyclone rainfall brings and areas currently less exposed to sea surges, and other weather-related risks (Hardoy and Pandiella tropical cyclones would face additional risks. 2009). Houses in informal settlements are frequently built with The vulnerability of people exposed to extreme events is inadequate materials, which make them damp and cold in the shaped by a multitude of non-climatic factors. Some socioeco- winter and very hot in the summer (Hardoy and Pandiella 2009). nomic factors shaping vulnerability are the same for rural and Hence, there are concentrations of low-income households at high 95 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL risk from extreme weather (Hardoy et al. 2001). For example, an protection provided by healthy coral reefs. Altogether these impacts estimated 1.1 million people live in the favelas of Rio de Janeiro that may augment impacts on coastal infrastructure (including beach sprawl over the slopes of the Tijuca mountain range, making them erosion), thus threatening transport, settlements, and tourism. particularly at risk from mudslides (Hardoy and Pandiella 2009). In combination with an up to 50 percent decrease in fish catch Moreover, most low-income people live in housing without air potential under a 4°C world (Cheung et al. 2010), damage to coral conditioning or adequate insulation; during heat waves, the very reefs threatens artisanal fisheries that support local livelihoods. young, pregnant women, the elderly, and people in poor health are Infrastructure in the Caribbean is already highly vulnerable particularly at risk (Bartlett 2008) (see also Section 3.4.7, Human to natural hazards, and important assets (including airports) are Health). In northern Mexico, heat waves have been correlated often low-lying. Further climate change impacts may affect the with increases in mortality rates; in Buenos Aires, 10 percent of condition of infrastructure, increasing failures and maintenance summer deaths are associated with heat strain; in Peru, records costs. The high vulnerability to hurricanes and tropical storms in show a correlation between excessive heat and increases in the low-lying states could further increase with a growing population, incidence of diarrhea (Mata and Nobre 2006). Such effects may putting growing numbers of assets at risk, exacerbating pervasive be compounded by a climate-change-related increase in the broad poverty/inequality and potentially leading to displacement of a geographic areas and microclimates in which certain vector-borne greater proportion of the population, as was observed in the wake diseases, such as malaria and dengue fever, can flourish (Costello of Hurricane Mitch (Glantz and Jamieson 2000; McLeman and et al. 2009). Hunter 2011). Ultimately, due to the small size of many Caribbean Adverse socioeconomic conditions in concert with exposure to islands, more frequent natural disasters may cause severe setbacks climate change impacts undermine the development of adaptive to the overall economy. However, not only coastal areas are at high capacity. People living in informal urban settlements without legal risk. In Central America and the Caribbean, the poor are often tenure rights—who are generally from poor and socially excluded living on steep slopes or close to rivers; they are therefore espe- communities (including marginalized ethnic groups)—in principle cially exposed to landslides and floods. Such hydro-meteorological have limited means or incentive to attempt to climate-proof their events may damage the poor quality (often informal) residential houses (Moser et al. 2010). A lack of accountability to the citizens structures of vulnerable communities, which could in turn lead to and a very limited scope for public participation in decision making higher mortality rates and population displacement. More generally, means that poorer areas are deprioritised for infrastructural upgrad- intensified non-climate stressors related to land use change and ing, and thus frequently have inadequate infrastructure (e.g., storm ecosystem degradation could impair the resilience and the ability drains) to cope with extreme events (Hardoy and Pandiella 2009). to cope with the impacts of extreme hydro-meteorological events. Furthermore, the impacts of extreme weather events are often more Climate extremes (e.g., drought, heat waves) in combination severe in areas that have been previously affected or have not yet with long-term decreases in precipitation may reduce crop yields been fully recovered from previous a previous extreme event; these in Central America and Caribbean countries and affect food secu- cumulative effects are difficult to overcome (Hardoy and Pandiella rity and market prices. This is particularly relevant in the case of 2009; Hardoy and Romero Lankao 2011). Damage to housing as a coffee crops, which are important for the livelihoods of workers result of extreme weather can lead to loss of key assets used in urban and small farmers in Central America. informal sector businesses (Moser et al. 2010), further undermining There are several studies projecting reduced runoff and ground- the buildup of resilience and increasing the risk of poverty traps. water recharge in a 4°C world (Table 3.15), which will reduce water availability. This may disproportionately affect the lives of 3.5.2 Sub-regional Development Narratives women responsible for managing household water resources, as well as the health and wellbeing of vulnerable members of poor 3.5.2.1 Central America and the Caribbean— households (e.g., infants, the chronically ill, and the elderly). Extreme Events as a Threat to Livelihoods Ultimately, water stress could increase conflicts over land, affect In a 4°C world, the Central American and Caribbean countries food security, and provoke climate-induced migration. Additionally, are projected to be at risk from higher ENSO and tropical cyclone Central America is heavily dependent on hydropower to generate frequency, drought and heat extremes, and precipitation extremes electricity; it is expected that energy security could become an issue. (Table 3.15). The impacts of tropical cyclones will be exacerbated by rising sea levels fostering storm surges. Moreover, by the year 3.5.2.2 The Andes—Changing Water Resources 2040, Caribbean coral reefs are expected to experience annual Challenge the Rural and Urban Poor bleaching events due to sea-level rise, ocean warming, and Climate change already affects and will further affect water resources sedimentation from flood events, which will diminish the coastal in the Andes (Table 3.15). These resources are already scarce as 96 LATI N AME R I CA A ND THE CA RIBBEA N a result of insufficient management and degradation of critical and forest-fringe communities would be at particular risk, which ecosystems (including the páramo and cloud forests). Increas- could, in turn, spur additional migration of affected groups to ing temperatures leading to higher evapotranspiration, changing cities, and could also open up forest regions to settlers, spurring precipitation patterns, and more extreme precipitation events all additional deforestation. directly affect river runoff. Moreover, glacial melt and snow pack Besides the risk of a tipping point, climate change is expected changes are important components of the regional hydrological to contribute to forest degradation and biodiversity loss. There balance. Increased glacial melt may increase water availability are variable rainfall patterns and significant differences in rain- in the next few decades while reducing it thereafter. Both glacial fall projections between the northern and southern zones of the melt and snowmelt affect the amount and seasonality of water Amazon. In a 4°C world, in the southern zone, winter annual flows. These changes threaten the water supply for hydropower, precipitation is likely to decrease while evapotranspiration and agriculture, and domestic use. This is particularly relevant because aridity is expected to increase (see Section 3.3). This puts the many large cities and populations are located at high altitudes or southern part of the forest at increased fire risk. in arid regions in the lowlands where alternative sources of water Increasing fires will not only lead to large emissions of CO2 are not abundant and where the urban poor are already suffering but, in combination with deforestation, declining rainfall, and from limited access to water. Moreover, the regional energy mix forest drying, may also draw the agricultural frontier northwards. strongly depends on hydropower; higher risks of power outages This threatens the livelihoods of forest-dependent communities may impact household and community welfare. In addition, and could lead to land-use conflicts between existing communi- subsistence farming and cattle herding in the highlands, as well ties and newly arriving farmers. In addition, timber harvest from as large-scale agriculture in the coastal areas, depend on water concessions could be negatively affected. Moreover, increasing coming from the mountains. While a decreasing water supply is fires in the Amazon threaten rural and urban settlements and the an important risk to food security and poverty levels in general, resulting smoke/haze could aggravate respiratory disease for both women and children who are often in charge of agriculture in high- forest dwellers and urban residents in central Brazil. altitude communities are at particular risk of increasing poverty Negative effects of climate change on biodiversity resulting from and water-scarcity-related conflicts. The same applies to indigenous habitat contractions and extinctions are very likely in a warmer people whose traditional water management systems are likely to than 2°C world. In combination with increasing forest degrada- be affected and whose livelihoods are already threatened. Other tion, changes in the range of certain species will affect resource water-dependent activities, such as large-scale or artisanal mining, availability for indigenous populations that are very reliant on may be affected as well. These stresses could exacerbate the cur- native plants and animals. This could increase malnutrition among rent urbanization trend, leading to further rural-urban migration children and the elderly and undermine traditional knowledge of and amplifying the risks to the urban poor. ecosystems, impacting the community social structure and the Besides these more gradual changes, extreme hydrological value placed on traditional knowledge. Altogether, these changes events (such as an intensification of ENSO, extreme precipitation, may push local communities to expand subsistence agriculture high flows, and glacial lake outburst floods) increase the risk for as an alternative livelihood strategy or to migrate to other forest natural disasters, erosion, and landslides. While such events may areas, thereby amplifying forest degradation and threatening generally decrease GDP, the impact across different layers of the existing protected areas. population is uneven, with the urban poor living on steep slopes typically at the highest risk. 3.5.2.4 Southern Cone—Risks to Export Commodities from Intensive Agriculture 3.5.2.3 The Amazon—Risk of Tipping Point, Forest The Southern Cone countries are currently a major grain and Degradation, and Biodiversity Loss Threatens Local livestock producing region for local and global markets (Chaherli Communities and Nash 2013). The region has experienced significant climate Despite an improved understanding of processes linking climate, shocks, mainly related to ENSO, which have resulted in floods vegetation, land-use change, and fire in the Amazon, the identifica- or droughts at critical phases of the crop cycle. Moreover, despite tion of the processes and the quantification of thresholds at which small and uncertain increases in precipitation, the region faces an irreversible approach toward a tipping point is triggered (i.e., increasing evapotranspiration rates under a 4°C world, (see Sec- a potential transition from forest to savannah) is still incomplete. tion 3.3). This highlights the high risks to agricultural production Overall the most recent studies suggest that forest dieback is an from climate change in a 4°C world; this is particularly true for unlikely, but possible, future for the Amazon region (Good et al. rain-fed agriculture, which is prevalent in more than 98 percent 2013). Should such die-back occur, the livelihoods of forest-dwelling of Brazil’s agricultural area (FAOSTAT 2013; Oliveira et al. 2009). 97 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL The results of agricultural modeling studies differ in the severity “drought polygon,” an area characterized by a semi-arid climate of the climate change impact, but most agree that climate change that suffers from recurrent droughts (Krol and Bronstert 2007). will very likely decrease agricultural yields for important food Parts of this region in Brazil have been identified as having socio- crops in Latin America in the absence of adaptation measures and climatic hotspots, given the naturally limited water availability, persistent CO2 fertilization (ECLAC 2010; Fernandes et al. 2012; a relatively low human development index, a high population Nelson, Rosegrant, Koo et al. 2010) (see also Section 3.4.3, Climate density (Torres and Lapola et al. 2012), and existing conflicts over Change Impacts on Agriculture). Moreover, while CO2 fertiliza- water (Araújo and Knight 2005; Krol et al. 2006). Especially in a tion may increases yields, there is some evidence of it decreasing 4°C world, dry regions in Mexico and Brazil face strong increases protein contents in major grains (Müller et al. 2014; Myers et al. in highly unusual heat extremes and aridity leading to more 2014). For sugarcane, there might be beneficial effects with yield intense and longer drought events (Table 3.15). Northeast Brazil increases (Table 3.15). Moreover, agro-ecozones in Brazil, includ- is particularly impacted by ENSO-related droughts; these may ing major grain belts, may move northward (to central Brazil) to become more frequent in a 4°C world. In dryland Brazil, urban already cleared lands in the Cerrado region; Assad et al. (2013) migration to rapidly growing coastal cities in the northeastern project displacement of poorly productive and degraded pasture states is highly likely to be the result of the loss of agricultural with intensive multicrop grain cropping and intensified pastures. income (Mendelsohn 2007). The impacts on agriculture and livestock (Table 3.15) may In these dry regions, increasing drought events may lead to lead to increasing food prices that could entail trade impacts and problems for urban water supply or widespread cattle deaths. In stresses on other regions’ food production systems and may alter addition, small-hold family farmers in rural areas may experience dietary patterns (especially of the poor). Alongside price risks, lower productivity or even lose entire harvests, threatening their reduced nutrient contents (especially protein) could also raise the livelihoods. A decline in agricultural productivity may cause local- risk of malnutrition in children. Opportunities may arise from the ized famines, especially among remote indigenous communities reshuffling of agricultural zones as plantation forestry, horticultural (particularly in Northern Mexico), and possibly result in long-term crops, and biofuel production from sugarcane may be able to impacts on the household nutritional status. expand on lands that become unsuitable for grain crops. Reduced Increases in irrigated agriculture, if not well integrated with crop and livestock productivity can be moderated via adaptation long-term water resource planning and management, pose another measures and climate-smart technologies (e.g., improved varieties risk as they will exacerbate issues of water availability and also and breeds, irrigation, conservation agriculture, liming, and fertil- concentrate wealth. A diminished drinking water supply in rural izers to enhance crop rooting depth). These intensification and communities may also lead to an increasing reliance on water trucks climate-smart innovations would, however, require a significant that occasionally deliver contaminated water (resulting in illness upgrading of knowledge and extensive field testing. and death). Moreover, the need to search for drinking water and the associated health problems associated with low-quality drinking 3.5.2.5 Dry Regions (Mexican Dry Subtropics and water could decrease the work force and income in rural areas, North Eastern Brazil)—Increasing Drought Stress leading to increased crime, social exclusion, and other problems Threatening Rural Livelihoods and Health related to rural-urban migration during drought events. Increasing There has been significant development progress in these regions water stress may also lead to further over-exploitation of aquifers in recent decades, which has lifted a number of communities in the Northern part of Mexico. This in turn would lead to the out of extreme poverty. The possibility of increasing droughts, release of groundwater minerals, affecting groundwater quality, however, threatens to force many of these populations back into and, in coastal aquifers, lead to sea water intrusion. In general, extreme poverty. hydropower and energy systems will be stressed across these dry The Central and Northern arid areas in Mexico and the semi- regions. Direct damages from droughts and secondary impacts arid areas in Mexico and Northeast Brazil are already under water on the agriculture sector and related labor markets may result in stress and are sensitive to inter-annual climate variability. For negative GDP growth rates in the agriculture sector. example, parts of northeast Brazil are situated within the so-called 98 3.6 Synthesis Table—Latin America and the Caribbean Table 3.15: Synthesis table of climate change impacts in LAC under different warming levels. OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s1) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Heat Highly Unusual Absent Around 10% of Up to 30% of land area 30–40% of land area Around 65% of land area Around 90% of land area Extremes Heat Extremes land area affected affected in DJF affected in DJF affected in DJF affected in DJF in DJF Unprecedented Absent Absent Around 5% of land area Around 15% of land area Around 40% of land area Around 70% of land area Heat Extremes affected in DJF affected in DJF affected in DJF affected in DJF Regional Warming (austral 0.8°C 1.5°C, warming limited 5.5°C, warming limited summer temperatures) along the Atlantic coast along the Atlantic coast of Brazil, Uruguay, and of Brazil, Uruguay, and Argentina with about Argentina with about 0.5–1.5°C. The central 2–4°C. The central South-American region South-American region of Paraguay, northern of Paraguay, northern Argentina, and southern Argentina, and southern Bolivia with more Bolivia with more pronounced warming, pronounced warming, up up to 2.5°C* to 6°C* Precipitation Relatively small changes Peru, Ecuador, and and disagreement among Colombia on the Pacific climate models. Peru, coast increase in annual Ecuador, and Colombia mean precipitation of about on the Pacific coast with 30%, most pronounced a small increase in annual during the summer. The mean precipitation of up Caribbean, Patagonia to 10%. Reduction in (southern Argentina winter precipitation over and Chile), Mexico, and southeastern Amazon central Brazil become drier rainforest* (10–40%). Central America becomes drier in winter (up to 60%). The annual mean precipitation in southeastern Amazon rainforest is projected to drop by 20% mostly because of a strong decrease in winter precipitation (–50%)* 99 100 Table 3.15: Continued. OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s1) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Extreme Precipitation Robust increase in Annual extreme daily Annual extreme daily Annual extreme daily intensity of extreme precipitation with 20–year precipitation with 20–year precipitation with 20–year precipitation events for return interval increases return interval increases return interval increases South America2 by 7% and the 20–year by 11% and the 20–year by 25% and the 20–year return value of maximum return value of maximum return value of maximum precipitation returns precipitation returns every precipitation returns every every 15 years*3 12 years*3 6.5 years. Important 5%, 7%, and 3% 9%, 7%, and 8% increase in areas are the Caribbean, increase in maximum maximum 5–day precipitation Meso-America, Southern 5–day precipitation in in the Amazon, Central Argentina, and Chile as well the Amazon, Central America, and Southern as parts of Brazil and the America, and Southern South America respectively*4 Pacific coastline of Ecuador, South America Peru, and Colombia*3 respectively*4 16%, 8%, and 12% increase in maximum 5-day precipitation in the Amazon, Central America, and Southern South America respectively*4 Drought Severe droughts in 2005 4-, 1- and 2-days 1%, 4% and 9% increase 8-, 2- and 2-days longer 17-, 10- and 8-days and 2010 in the Amazon5 longer droughts in the in days under drought droughts in the Amazon, longer droughts in the Increase in drought Amazon, Central America conditions in Caribbean, Central America and Amazon, Central America, conditions in Central and Caribbean, and Meso-America, Caribbean, and Southern and Caribbean and America6 Southern South America and South America South America respectively*4 Southern South America respectively*4 respectively*7 11.5%, 12%, and 12.5% respectively*4 increase in days under 22%, 25%, and 22% drought conditions in increase in days under Caribbean, Meso-America, drought conditions in and South America Caribbean, Meso-America, respectively*7 and South America Reduction by 5 to 9% in respectively*7 annual soil moisture content in Amazon and Central America*8 Increase in extreme droughts in the Amazon, Brazil, Central America, northern Mexico, and Southern Chile*8 OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s1) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Aridity 33% of land area hyper- 36% of land area hyper- 41% of land area hyper- arid, arid, or semi-arid arid, arid, or semi-arid arid, arid, or semi-arid (increase of about 10%)* (increase of about 25%)* Sea-level Rise Above Median estimate across Median estimate across the Present (1985–2005) the region 0.27–0.39 m, region 0.46–0.66 m, with with highest sea-level highest sea-level rise on the rise on the Atlantic Coast Atlantic Coast and lowest and lowest on the tip of on the tip of the American the American continent. continent. Maximum Maximum 0.65 m sea- 1.14 m sea-level-rise and level rise in Recife* 1.4 m in Rio de Janeiro and Barranquilla on the Atlantic Coast by 2100* El Niño Southern ENSO has never been as Doubling of frequency of Oscillation (ENSO) variable as during the last extreme El Niño events*10 few decades9 Tropical Cyclones Tropical cyclone Power Dissipation Power Dissipation Index frequency increase in the Index increasing by increasing by 125–275%*12 North Atlantic11 100–150%*12 Increase of 80% in the Increase of 40% in frequency of the strongest the frequency of the category 4 and 5 Atlantic strongest Atlantic Tropical Tropical Cyclones*14 Cyclones*13 Glaciers Southern Up to 22% loss of glacial 21–52% loss of glacial 27–59% loss of glacial 44–72% loss of glacier Andes volume15,16 volume*15 volume*15, 16, 20 volume*15, 20 Reduction in glacier length by 3.6–36% (Northern Patagonian Ice Field), 0.4–27% (Southern Patagonian Ice Field), and 2.5–38% (Cordillera Darwin Ice Field)17 31.7% loss of glacial area15 23–26.6 Gt/yr glacial mass loss rate over Patagonian Ice Fields18 1.88 Gt/yr annual calving loss in Northern Patagonian Ice Field19 101 102 Table 3.15: Continued. OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s1) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Tropical Up to 90% loss of glacial 78–94% loss of glacial 66–97% loss of glacial 91–100% loss of glacier Glaciers volume15, 16 volume*15 volume*15, 16, 20 volume*15, 20 79% loss of glacial area15, 87% in Andes of Venezuela over 1952–2003, 11% in Andes of Colombia over 1950–1990s, 57% in Chimborazo over 1962– 1997, 37% in Cotopaxi and 33% in Artinsana over 1979–2007, and 20–35% in Peruvian Andes over 1960–2000s21 6 Gt/yr glacial mass loss rates22 Water Central Up to 10% less runoff23 13% decrease in total 10–30% decrease of mean 24% decrease in total America & annual reservoir inflow in annual runoff*25 annual reservoir inflow in Caribbean Rio Lempa*24 5–20% decrease in river Rio Lempa*24 Around up to 15–45% runoff23 10% in groundwater reduction of annual 20% decrease in discharge recharge*27 discharge39 from Rio Grande*26 15–45% reduction of annual discharge39 Andes Discharge in Cordillera Decreasing mean Wet season discharge Wet season discharge in 21% streamflow decrease Blanca decreasing annual runoff in the Llanganuco the Llanganuco catchment In the Limarí basin and annually and during the for northeastern catchment increases increases from 10–26% increasing winter flows dry season28 Chile29 from 10–26% and and dry season discharge (28.8–108.4%), decreasing dry season discharge decreases from 11–23%30 summer flows (–16.5 to decreases from –57.8%), and earlier center 11–23%30 timing of mass of annual Reduced groundwater flows for different sub- recharge for the central basins of the Limarí basin32 Andes region31 Likely increase in flood frequency33 OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s1) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Water Amazon Decreasing mean annual Low-flows become more Total annual runoff decreases Low flows increase discharge and monthly pronounced over several in the southern half of the between 10–30% in minimum discharge Amazonian sub-basins35 Amazon River36 the western part of the for Tapajós in the Median high-flows Duration of inundation Amazon35 southeastern Amazon, increase by 5–25% in 0.5–1 month shorter in Low flows and high flows the Peruvian Amazon the western part of the eastern Amazonia*37 would increase each by Rivers, and the upstream Amazon basin35 Inundation area will increase 5% at Óbidos35 Madeira34 Median low-flows with a 2–3 month longer decrease significantly, by inundation time in the 20% for the Japura and western part of the Amazon Negro river and 55% at basin*37 the Río Branco35 Northeast Seasonality of river Strong decreases and Brazil discharge remains stable increases in mean but mean river discharge groundwater discharge decreases38 depending on the GCM27 No clear signal in relative change of annual discharge39 Rio de la Plata Increase in river runoff of Mean river flow from Increase in mean relative Increase in frequency and 10–30%40 –20% to +18% for the runoff for the Rio de la Plata duration of fluvial floods in Río Grande, a tributary of region of 20–50%23 the Uruguay and Paraná the Paraná41 basin42 Decrease in the 20th century 100–year return period for floods for the Parana43 Southernmost Decrease in mean relative Decrease in mean relative 15–45% reduction of South America runoff up to 10%23 runoff by 10–30%23, 39 annual discharge39 Crop Wheat Brazil: –23%44 Brazil: up to –50%44 Brazil: –41% to –52%44 Argentina: –16%48 yield Central America and Central America and Central America and Caribbean: –43%44 Caribbean: –56%44 Caribbean: –58% to –67%44 LAC: 6.5–12%45 and LAC: 0.9–12%45 and –5.5 0.3–2.3%46 to 4%46 Chile*: up to –10%45,47 Argentina: –11%48 103 104 Table 3.15: Continued. OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s1) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Crop Maize Panama: up to México: –29%44 Ecuador and Brazil up to Brazil: –30 to –45%50 Argentina: –24%48 yield –0.5%45,49 Panamá: 0.8%45,49 –64%44 Panama: 4.5%45,49 Ecuador: –54%48 México: up to –45%44 Panama: 1.5–2.4%45,49 LAC: –2.3 to +2.2%45 and –0.4 to –2.8%46 Brazil: –15 to –30%50 Soybean Brazil: –45%44 Brazil: up to –70%44 Brazil: –66% to –80%44 Argentina: –25%48 Brazilian Amazon: LAC: 18–19%45 and –2.5 Brazilian Amazon: –44%51 –1.8%45,51 to 4%46 LAC: 19.1–19.5%45 and Argentina: –14%48 –1.2 to 2.3%46 Rice Central America and Central America and Central America and Ecuador: 37%48 Caribbean: +3%44 Caribbean: –4%44 Caribbean: +1.5–4%44 LAC: –1.2 to +13%45 and LAC: 6.7–745 and –0.8 to –6.4 to 5%46 –1.8%46 Beans Brazil: –15 to –30%50 Brazil: –30 to –45%50 Ecuador: –9%48 Coffee Ecuador: –23%48 Cocoa Ecuador: –21%48 Bananas Ecuador: –41%48 Sugarcane Southern Brazil: 15%45,52 Southern Brazil: 59%45,52 Ecuador: –36%48 Livestock Livestock choice in Livestock choice in 7 to 16% decrease in 22 to 27% decrease in beef Argentina, Brazil, Argentina, Brazil, Chile, beef cattle numbers in cattle numbers in Paraguay48 Chile, Colombia, Colombia, Ecuador, Paraguay48 Ecuador, Uruguay, Uruguay, and Venezuela: Livestock choice in and Venezuela: Beef Cattle: –1.6 to 5% Argentina, Brazil, Chile, Beef Cattle: –12.5 Dairy cattle: –6.7 to 2.5% Colombia, Ecuador, to 5.7% Pigs: –0.8 to 0.0% Uruguay, and Venezuela: Dairy cattle: –6.6 Sheep: 0.0 to 7.0% Beef Cattle: –11.0 to to 1.2% Chicken: –1.0 to 1.3%53 0.3% Pigs: –1.6 to 0.2% Dairy cattle: –10 to 5% Sheep: –5 to Pigs: –0.9 to 0.1% 20.1% Sheep: 0.0 to 19% Chicken: –2.9 to Chicken: –1.5 to –0.3%53 1.4%53 OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s1) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Biodiversity Extinction rates of Extinction rates of 68% loss of suitable area for In most LAC ecoregions, species: 2–5% for species: 2–8% for cloud forest and extinction of amphibian species mammals, 2–4% for mammal, 3–5% for 9 of 37 vertebrate species in experience at least 30% birds, 1–7% for butterfly birds, 3–7% for butterfly Mexico63 turnover; in western South species in Mexico, and species in Mexico, and 78% reduction in geographic America and Central 38–66% for plant species 48–75% of plant species distribution of 110 Brazilian America at least 50%*57 in Cerrado54 in Cerrado54 Cerrado plant species64 Up to 21 out of 26 Changes in amphibian biogeographic ecoregions species ranges in in South America faces the Atlantic Forest severe ecosystem Biodiversity Hotspot55 change*58 Marsupial species ranges Loss of habitat between declining in Brazil56 11.6% to 98.7% and 85–95% of LAC change in species richness amphibian species face between –25% to –100% net loss in range size*57 for plants in Amazon61 1 out of 26 biogeographic ecoregions in South America faces severe ecosystem change*58 Climatically suitable areas for cloud forest reduced by 54–76%59 44 of 51 birds species lose distribution area in Brazilian Atlantic forest60 Loss of habitat between 8.2% to 81.5% and change in species richness between –4.1% to –89.8% for plants in Amazon61 Majority of 430 amphibian species would face range contractions accompanied by an overall species loss in the Atlantic Forest Biodiversity Hotspot62 105 Table 3.15: Continued. 106 OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s1) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Amazon Dieback Model agreement on Carbon loss (kg C/m2) by 10–80% forest cover above-ground live –1.8 to –0.6 in Eastern loss*68 biomass loss: 14.3% Amazonia, –1.2 to 0.6 in –35% to +40% change for climate change Northwestern Amazonia and of carbon without only but increasing to –3.3 to –2.6 in Southern deforestation and 43.1–58.6% with different Amazonia*66 –55% to –5% with 50% deforestation scenarios65 Carbon increase (kg C m2) by deforestation*69 5.5–6.4 in Eastern Amazonia, 10–80% forest cover loss*70 2.9–5.5 in Northwestern Carbon losses: 70 GtC Amazonia and 2.1–4.3 in (Vegetation carbon), Southern Amazonia*45,56 150 GtC (Soil carbon)*71 Decrease of LAI by 12.6% 69% reduction in rainforest and increase in land- extent*72 atmosphere carbon flux of Model agreement on about 27.2% due to fire*67 above-ground live biomass loss: 25.5% for climate change only but increasing to 48.1–65.9% with different deforestation scenarios65 Coral Reefs Strong bleaching event 20–40% and up to 60% <60% and >60% >60% probability of annual in 2005, less severe in probability of annual probability of annual bleaching events in all 2010, in the Caribbean bleaching events in bleaching events in regions*73 Sea, Guyana, Suriname, Caribbean Sea and Caribbean Sea and French Guiana, and north Guyana, Suriname, Guyana, Suriname, Pacific Ocean73 French Guiana, and and French Guiana north Pacific Ocean respectively73 respectively73 60–80% and up to Coral cover halved from 100% probability of initial state in Virgin annual bleaching events Islands and Eastern in Caribbean Sea and Caribbean74 Guyana, Suriname, Onset of bleaching and French Guiana events starts 2046*75 respectively*73 Coral cover less than 3–5% in Virgin Islands and Eastern Caribbean74 Onset of bleaching events starts 2040*75 OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s1) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Marine Fisheries Species shifts toward 35% decline in Large increase in catch higher latitudes76 phytoplankton, potential in the south zooplankton, and fish (up to 100%), strong density77 decrease in parts of the Caribbean Sea (up to 50%)76 Health 5–13% increase in Expansion of malarial 12–49 million people less 19–169 million people less relative risk of diarrheal areas mostly in Brazil80 exposed to risk of malaria exposed to risk of malaria diseases in South No net changes in for at least three months of for at least three months of America78 increased malaria length the year*82 the year*82 12–22% increase in of transmission season 1–16 million people more 5–42 million people less Dengue incidence in except in southernmost exposed to risk of malaria exposed to risk of malaria Mexico79 Brazil and Uruguay*81 for at least one month of the for at least one month of 31–33% increase in year*82 the year*82 Dengue incidence in Increased malaria length Increased malaria length Mexico79 of transmission season in of transmission season in southern Brazil, Uruguay, some highland areas of and parts of Mexico*81 southern Brazil, Uruguay, Decreased malaria length Argentina, Bolivia, Peru, of transmission season Ecuador, Colombia, and in parts of the Amazon Mexico*81 basin in Brazil, Bolivia, and Decreased malaria length Paraguay*81 of transmission season for 14–36% increase in relative tropical Latin America*81 risk of diarrheal diseases in South America*78 40% increase in Dengue fever incidence in Mexico*79 107 Table 3.15: Continued. 108 OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s1) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Energy 683,421 GWh/yr 688,452–861,214 GWh/ 715,173–838,587 GWh/ Decrease in firm power by maximum hydropower yr maximum hydropower yr maximum hydropower 1.58%*86 energy potential in La energy potential in La energy potential in La Plata Energy demand for 2,679 Plata River Basin83 Plata River Basin83 River Basin83 cooling degree days in Energy demand for 0,63TWH (or 0.05%) Decrease in hydropower South America84 1,802 cooling degree and 0,3TWH (or 0.03%) capacity for the two main days in South America 84 increase in electricity large reservoirs used for production in South hydroelectricity generation in America and in the El Salvador: Cerron Grande Caribbean respectively85 and 15 Setiembre*24 Decrease in hydropower Decrease in firm power by capacity for the two 3,15%*86 main large reservoirs used for hydroelectricity generation in El Salvador: Cerron Grande and 15 Setiembre*24 The impacts reported in several impact studies were classified into different warming levels (see Appendix for details) LATI N AME R I CA A ND THE CA RIBBEA N Endnotes 1 Years indicate the decade during which warming levels are exceeded with a 50 percent or greater change (generally at start of decade) in a business-as-usual scenario (RCP8.5 scenario) (and not in mitigation scenarios limiting warming to these levels, or below, since in that case the year of exceeding would always be 2100, or not at all). Exceedance with a likely chance (>66 percent) generally occurs in the second half of the decade cited. Impacts are given for warming levels irrespective of the timeframe (i.e., if a study gives impacts for 2°C warming in 2100 then the impact is given in the 2°C column). If a study refers to a warming level by the end of the century, this is marked with an asterisk (*). Impacts given in the observations column do not necessarily form the baseline for future impacts. Impacts for different warming levels may originate from different studies and therefore may be based on different underlying assumptions; this means that the impacts are not always fully comparable (e.g., crop yields may decrease more under a 3°C than a 4°C scenario because underlying the impact at 3° warming is a study that features very strong precipitation decreases). Moreover, this report did not systematically review observed impacts. It highlights important observed impacts for current warming but does not conduct any formal process to attribute impacts to climate change. 2 Skansi et al. (2013). 3 Kharin et al. (2013); 20–year return value of maximum precipitation refers to 1986–2005. 4 Sillmann et al. (2013b). 5 Marengo et al. (2011); Zeng et al. (2008). 6 Dai (2012). 7 Prudhomme et al, (2013) increase in days under drought conditions refers to 1976–2005. 8 Dai (2012). 9 Li et al. (2013). This is a tree-ring-based reconstruction of ENSO strength over the last 700 years, but attribution to climate change is uncertain. 10 Cai et al. (2014). 11 IPCC AR5 WGI (2013). Frequency increase in the North Atlantic over the past 20–30 years. 12 Villarini et al. (2013). The Power Dissipation Index is a combination of frequency and intensity. 13 Knutson et al. (2013). 14 Bender et al. (2010); Knutson et al. (2013). 15 Marzeion et al. (2012). Past period for glacial volume loss and area loss refers to 1901–2000. 16 Giesen and Oerlemans (2013). For past: 6.1 percent (southern) and 7.3 percent (tropical) loss of glacial volume over 1980–2011 compared to 1980. 17 Lopez et al. (2010). 18 Ivins et al. (2011); Jacob et al. (2012) refers to 2000s. 19 Schaefer et al. (2013) refers to 1990–2011. 20 Radic et al. (2013). 21 Rabatel et al. (2013). Andes of Venezuela over 1952–2003; Andes of Colombia over 1950–1990s; Chimborazo over 1962–1997; in Cotopaxi and in Artinsana over 1979–2007; and Peruvian Andes over 1960–2000s. 22 Jacob et al. (2012) for past refer to 2000s. 23 Milly et al. (2005). 24 Maurer et al. (2009). 25 Hidalgo et al. (2013). 26 Nakaegawa et al. (2013). 27 Portmann et al. (2013). 28 Baraer et al. (2012). 29 Arnell and Gosling (2013). 30 Juen et al. (2007). No differentiation possible for changes in warming for >1.5°C in 2050 and >2°C in 2080. 31 Döll (2009). 32 Vicuña et al. (2010). 33 Hirabayashi et al. 2013). 34 Espinoza Villar et al. (2009). 35 Guimberteau et al. (2013). 36 Nakaegawa et al. (2013). 37 Langerwisch et al.(2013). 38 Döll and Schmied (2012). 39 Schewe et al. (2013). 40 García and Vargas (1998); Jaime and Menéndez (2002); Menéndez and Berbery (2005); Milly et al. (2005). 41 Nóbrega et al. (2011). 42 Camilloni et al. (2013). 43 Hirabayashi et al. (2013). There was little consistency across the 11 GCMs used. 44 Fernandes et al. (2012). 45 With CO2 fertilization. 46 Nelson, Rosegrant, Koo et al. (2010). 47 Meza and Silva (2009). 48 ECLAC (2010). 49 Ruane et al. (2013). 50 Costa et al. (2009), including technological progress. 51 Lapola et al. (2011). 52 Marin et al. (2012), including technological progress. 53 Seo et al. (2010). 109 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL 54 Thomas et al. (2004). Mammal species (n=96), bird species (n=186), and butterfly species (n=41) in Mexico all with dispersal; plant species in Cerrado (n=163) without dispersal. The study was criticized by Harte et al. (2004) for overestimating potential extinction rates by using a common species-area exponent z for all species which may not be justified. 55 Loyola et al. (2013). 56 Loyola et al. (2012). 57 Lawler et al. (2009). 58 Gerten et al. (2013). 59 Rojas-Soto et al. (2012). 60 Souza et al. (2011). 61 Feeley et al. (2012). Large range stems from different assumption about deforestation, land-use, and adaptation and migration potentials. 62 Lemes et al. (2014). 63 Ponce-Reyes et al. (2012). 64 Simon et al. (2013). 65 Poulter et al. (2010). 66 Rammig et al. (2010). 67 Cook et al. (2012). 68 Zelazowski et al. (2011). 69 Gumpenberger et al. (2010). 70 Cox et al. (2004). 71 Betts et al. (2004). 72 Cook and Vizy (2008). 73 Meissner et al. (2012). 74 Buddemeier et al. (2011). 75 Van Hooidonk et al. (2013). 76 Cheung et al. (2010). 77 Blanchard et al. (2012). 78 Kolstad and Johansson (2011), compared to 1961–1990 levels. 79 Colon-Gonzalez et al. (2013), compared to 2000. 80 Beguin et al. (2011). 81 Caminade et al. (2014). 82 Van Lieshout et al. (2004). 83 Popescu et al. (2014). 84 Isaac and van Vuuren (2009). 85 Hamududu and Killingtveit (2013), compared to 2005 levels. 86 De Lucena et al. (2009), compared to 1971–2000. 110 Chapter 4 Middle East and North Africa The Middle East and North Africa region is one of the most diverse in the world in economic terms, with per-capita annual GDP ranging from $1,000 in Yemen to more than $20,000 in the Arab Gulf States. As a consequence, adaptive capacity and vulnerability to climate risks varies enormously within the region. The region will be severely affected at both 2°C and 4°C warming, particularly because of the large increase in projected heat extremes, the substantial reduction in water availability and expected consequences for regional food security. In some countries, crop yields could decrease by up to 30 percent at 1.5–2°C and by almost 60 percent at 3–4°C in parts of the region. Deteriorating rural livelihoods may contribute to internal and international migration, adding further stresses on particularly urban infrastructure with associated health risks for poor migrants. Migration and climate‐related pressure on resources might increase the risk of conflict. 4.1 Regional Summary the region’s agricultural production is currently rain-fed, which leaves the region highly vulnerable to temperature and precipitation The population in Middle East and North Africa is projected to changes, and the associated implications for food security, social double by 2050, which together with projected climate impacts, security, and rural livelihoods. This, in combination with social puts the region under enormous pressure for water and other changes and strong urbanization rates, indicates a very vulner- resources. The region is already highly dependent on food imports. able future for the Middle East and North Africa, particularly for Approximately 50 percent of regional wheat and barley consump- the urban and rural poor. All countries in the region face a severe tion, 40 percent of rice consumption, and nearly 70 percent of maize and fast growing resource squeeze, especially relating to severe consumption is met through imports. The region has coped with its water and land scarcity. The region is very diverse in terms of inherent water scarcity through a variety of means: abstraction of socio-economic and political conditions. Thus, adaptive capacity groundwater, desalinization, and local community coping strategies. and vulnerability to climate risks varies enormously, especially Despite its extreme water scarcity, the Gulf countries use more between the Arab Gulf States and the other countries. water per capita than the global average, with Arab residential water and energy markets among the most heavily subsidized in 4.1.1 Regional Patterns of Climate Change the world. The region is very diverse in terms of socio-economic and political conditions. Thus, adaptive capacity and vulnerability 4.1.1.1 Temperatures and Heat Extremes to climate risks varies enormously, especially between the Arab Warming of about 0.2° per decade has been observed in the Gulf States and the other countries in the region. region from 1961–1990, and at even faster rate since then, which Middle East and Northern Africa heavily relies on agriculture in line with an increase in frequency in temperature extremes. as a source of food and income, not only in the historically impor- Geographically, the strongest warming is projected to take place tant “fertile crescent” of the Euphrates and Tigris region, but also close to the Mediterranean coast. Here, but also in inland Algeria, at the Mediterranean coast and the Nile, while at the same time Libya and large parts of Egypt, warming by 3°C in a 2°C world being largely covered by drylands and deserts. Seventy percent of is projected by the end of the century. In a 4°C world, mean 113 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 4.1: Multi-model mean temperature anomaly for RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) for the months of June-July-August for the Middle East and North African region. Temperature anomalies in degrees Celsius are averaged over the time period 2071–2099 relative to 1951–1980. summer temperatures are expected to be up to 8°C warmer in monsoon systems), in particular to the southern Arabian Penin- parts of Algeria, Saudi Arabia and Iraq by the end of the century sula (Yemen, Oman). Consequently, projected annual mean pre- (see Figure 4.1). cipitation changes show a clear North-South dipole pattern, with By the end of the century, in a 2°C world, highly unusual38 regions north of 25°N becoming relatively drier and regions to the heat extremes will occur in about 30 percent of summer months south becoming wetter. The absolute increase in precipitation in almost everywhere in the MENA region. This implies that on the southern regions, however, will be very small, because these average one of the summer months each year will exceed tem- regions (with the exception of Yemen) are already very dry today. peratures warmer than three standard deviations beyond the Furthermore, the effect of an increase in precipitation on water baseline average. Unprecedented heat extremes, however, will availability should be counteracted by a simultaneous increase in remain largely absent in a 2°C world, except for in some isolated temperature, resulting in a higher rate of evaporation. Lastly, an coastal regions including the Mediterranean coasts of Egypt, and increase in precipitation in the southern part of the region may in Yemen, Djibouti and Oman. Here these events are projected to be associated with more intense and extreme precipitation events. be relatively rare in a 2°C world, but are nevertheless expected There is a close match between the pattern of change in to occur in 5–10 percent of summer months. the annual mean aridity index (AI) and projected precipitation Whereas the increase in frequency of heat extremes is expected changes. Changes in the aridity are primarily driven by changes to level off by mid-century in a 2°C world, in a 4°C world it will in precipitation, with wetter conditions south of 25°N and in continue increasing until the end of the century. in a 4°C world, most southern parts of the Arabian Peninsula causing a drop in 80 percent of summer months are projected to be hotter than 5-sigma aridity, and drier conditions north of 25°N causing aridity there to (unprecedented heat extremes) by 2100, and about 65 percent are increase. In the Mediterranean coastal region, the relative increase projected to be hotter than 5-sigma during the 2071–2099 period. in aridity is more pronounced than would be expected from the drop in precipitation, because there is a substantial increase in 4.1.1.2 Precipitation and Aridity evapotranspiration here due to enhanced warming. Future northward shifts of air moisture associated with a stronger North Atlantic Oscillation (NAO) anomaly are projected to reduce 4.1.2 Regional Sea-Level Rise rainfall in North Africa, Maghreb, and Mashrek. In a 4°C world, countries along the Mediterranean shore, notably Morocco, Algeria, In the Mediterranean area, tide gauges recorded below- and Egypt, are projected to receive substantially less rain. However, average sea-level rise during the 20th century, with average a projected northward shift of the Inter-Tropical Convergence Zone rise of 1.1–1.3 mm per year (slower than the global average of (ITCZ) is expected to increase moisture delivery to the southern 1.8 mm per year). There has been significant interdecadal vari- parts of the region (which are already under the influence of ability, however, with a slow gradual rise from 1960–1990, and rapid (above-average) rise after 1990. 38 In this report, highly unusual heat extremes refer to 3-sigma events and unprec- Analysis for the 21st century indicates slightly below-average rise edented heat extremes to 5-sigma events (see Appendix). in the Mediterranean basin mostly as a result of the gravitational 114 MI DDLE E AST A ND NOR TH A FRICA Figure 4.2: Multi-model mean of the percentage change in the aridity index in a 2°C world (left) and a 4°C world (right) for the Middle East and North Africa by 2071–2099 relative to 1951–1980. Hatched areas indicate uncertain results, with two or more out of five models disagreeing on the direction of change. Note that a negative change cor- responds to a shift to more arid conditions (see Appendix).39 influence of Greenland ice sheet. Tunis, on the Mediterranean on food imports. From the current situation of critical water and Sea coast, is projected to experience a median sea-level rise of arable land scarcity, both the 2°C and 4°C warming scenarios 0.56 m (with a maximum of 0.96 m) by the end of the century would put further pressure on water resources and agriculture. in a 4°C world. This is 8 cm less than in Muscat, on the Arabian • Cropland: Warmer and drier climate is projected to shift veg- Sea coast, where a median 0.64 m (low estimate: 0.44 m, high etation and agricultural zones northward (e.g., by 75 km for estimate: 1.04 m) sea-level rise is projected. On the Atlantic coast, 2090–2099 relative to 2000–2009 in a 4°C world). a 0.58 m sea-level rise is projected for Tangier (low estimate: 0.39 m, high estimate: 0.98 m). In a 1.5°C world, median sea • Length of growing period: Lower rainfalls and higher tem- level rises of 0.34 m, 0.35 m and 0.39 m are projected for Tunis, peratures will shorten growing periods for wheat in large parts Tangier, and Muscat. 39 of the region by about two weeks by mid-century (2031–2050). The wheat growing period in Tunisia is expected to be short- 4.1.3 Sector-based and Thematic Impacts ened by 10 days for 1.3°C warming, by 16 days for 2°C, by 20 days for 2.5°C and by 30 days for 4°C warming. 4.1.3.1 The Agriculture-Water-Food Security Nexus • Crop yields: Crop yields are expected to decline by 30 per- The Middle East and North Africa region is water scarce, with cent with 1.5–2°C warming and up to 60 percent with 3–4°C most of the land area receiving less than 300 mm of annual warming, with regional variation and without considering rainfall (200–300 mm represents the lower limit of rain-fed agri- adaptation. Reductions in crop productivity of 1.5–24 percent culture). Semi-arid belts along the coasts and mountains are the are expected for the western Maghreb and 4–30 percent in only water source areas and provide productive land for rain-fed parts of the Mashrek, by mid-century. Legumes and maize agriculture. The annual availability of renewable water resources crops are expected to be worst affected in both areas as they in most countries is below 1000 m3 per capita (except for Iraq, are grown during the summer period. Oman, Syria and Lebanon) and as low as 50 m3 per capita for • Livestock: Climate change will impact livestock production Kuwait. This water scarcity prevents countries from producing through various pathways, including changes in the quantity all required food domestically and makes the region dependent and quality of available feeds, changes in the length of the grazing season, additional heat stress, reduced drinking water, 39 and changes in livestock diseases and disease vectors. Some individual grid cells have noticeably different values than their direct neighbors (e.g., on Turkey’s Black Sea coast under RCP8.5). This is due to the fact Uncertainty in projections arises from different approaches, that the aridity index is defined as a fraction of total annual precipitation divided by different climate models, and the persistence of CO2-effects because potential evapotranspiration (see Appendix). It therefore behaves in a strongly non- linear fashion and year-to-year fluctuations can be large. As the results are averaged increasing atmospheric CO2 concentration can potentially increase over a relatively small number of model simulations, this can result in local jumps. plant water-use efficiency (and thus crop productivity). 115 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL As a result of regional warming and changes in precipitation cholera (which correlate with high temperatures and can follow patterns, water availability is projected to decrease in most parts extreme weather events that disrupt water supply) have in recent of the region throughout the 21st century. For example, in the years caused deaths in Iraq, the Islamic Republic of Iran and the eastern Anatolian mountains (headwaters of Euphrates and Tigris Republic of Yemen. rivers) a runoff decrease of 25 percent to 55 percent is projected The Middle East and North Africa region is already character- with 4°C warming. ized by very high summer temperatures, making the populations of Mountain areas in Morocco, Algeria, Lebanon, Syria, Iraq, Iran the region highly susceptible to further temperature increases. In a and Turkey play an important role in the water supply of the region, 2°C world, the annual number of hot days with exceptionally high as they store a fraction of precipitation as snow. With projected temperatures and high thermal discomfort is expected to increase reduction in snowfall and snow water storage, peak flows of melt in several capital cities, from 4 to 62 days in Amman (Jordan), from water will shift towards earlier months, with negative impacts 8 to 90 days in Baghdad (Iraq) and from 1 to 71 days in Damascus for downstream river systems and water availability in distant (Syria). The greatest increase is expected in Riyadh (Saudi Arabia) regions. For example, snowpack in the upper Nahr el Kalb basin where the number of hot days is projected to rise from 3 to 132 in Lebanon was projected to shrink by 40 percent with 2°C warm- days per year. In a 4°C world, the average number of hot days is ing, and 70 percent with 4°C warming. Hence, drought periods projected to exceed 115 days per year in all of these cities. would occur 15–20 days earlier under a 2°C warming scenario, and more than a month earlier under a 4°C warming scenario. 4.1.3.4 Migration and Security The literature review revealed a link between climate change 4.1.3.2 Desertification, Salinization, and Dust Storms and migration in the region. It is expected that migration options The importance of climate change for desertification varies depend- will be more limited in a warmer world. Internal migration will ing on local conditions, and interactions between drivers can be continue to be important, but traditional patterns of mobility multifaceted. An increase in temperatures and evapotranspiration, might be disrupted. Many people will be forced to move, while change of precipitation regime, and the intensification or change others trapped in poverty will be forced to stay. This indicates that in frequency of extreme events can directly trigger or enhance the climate-induced migration should be addressed not only within desertification processes. Being covered mostly by drylands, the the context of climate change but also within economic, cultural, region is frequently threatened by dust storms, causing damage technological, and political frameworks. and disruption to people, agriculture and the economy. While Climate change could act as a threat multiplier in the region there are no direct projection studies on dust storms in the region, by placing additional pressure on already scarce resources and wind as a driving factor can be projected from climate models. reinforcing preexisting threats as political instability, poverty, and However, there are no regional studies on changing wind patterns unemployment. This can create the conditions for social uprising and under climate change in the region as yet, and future trends have violent conflict. Establishing a direct link between climate change to be derived from global studies. and conflicts is challenging due to contradictory conclusions and An increase in salinization under climate change holds for all methods. The findings are in some cases based on a single extreme water resources in the region. The densely populated coastal areas event; others use rainfall or temperature variability as proxies for in the region are most affected by climate-change-induced saliniza- long-term changes; and some examine short-term warming. Further tion (seawater intrusion), which is accelerated by climate-induced research is needed to investigate and establish the link between sea level rise. River salinization, meanwhile, is documented in climate change and conflict and to relate long-term climate change, studies of the Euphrates and Tigris, the Jordan River, and the Nile. instead of single climatologic hazards, to migration and to conflicts. The salinization process is complex, however, and climate change is but one important factor among others (including irriga- 4.1.3.5 Coastal Infrastructure and Tourism tion, water uptake and land subsidence). Climate change and, in Middle East and in North Africa countries are vulnerable to the particular, projected drier conditions in the region, are expected to impacts of sea-level rise. The population at risk in coastal cit- compound these other drivers (e.g., as more irrigation is needed ies numbered approximately 60 million in 2010; that number is for agriculture). expected to rise to 100 million by 2030. Separating out the socio- economic drivers of vulnerability from the effects of sea-level 4.1.3.3 Human Health rise, a study of 136 coastal cities identified Alexandria, Benghazi, The region is currently experiencing a resurgence of several and Algiers as particularly vulnerable to a 0.2 m sea-level rise by vector-borne and viral diseases that had previously been in 2050. The study projected that, in the event of the failure of flood decline. Climate change may compound the challenge of managing defenses, the effects of sea-level rise would increase damages from these diseases, including such vector-board diseases as malaria, $16.5 billion to $50.5 billion in Alexandria, from $1.2 billion to lymphatic filariasis, and leishmaniasis. In addition, outbreaks of $2 billion in Benghazi, and from $0.3 billion to $0.4 billion in 116 MI DDLE E AST A ND NOR TH A FRICA Algiers. Annual losses would increase to $58 billion, $2.7 billion 4.2 Introduction and $0.6 billion with 0.4 m of sea-level rise for these three cities respectively. A sea-level rise of one meter could impact 10 percent In this report, the Middle East and North Africa (MENA) is com- of Egypt’s population, five percent of its urban area, and decrease posed of 20 countries from Morocco to Iran. For the projections on the country’s GDP by six percent. One study estimated that a changes in temperature, precipitation, aridity, heat extremes, and sea-level rise of 0.30 m (projected for 2025 in this study) would sea-level rise, the MENA region stretches from 2°W to 63°E and flood 30 percent of metropolitan Alexandria, forcing about 545,000 from 10°N to 42°N. The countries in the region can be divided people to abandon their homes and land, and leading to the loss into four groups, which share geographical, historical, and/or of 70,500 jobs. With a sea-level rise of 0.5 m, projected for 2050, economic similarities: the same study calculated that about 1.5 million people would • the Maghreb in the western part of North Africa: Morocco, be displaced and about 195,500 jobs lost. Algeria, Tunisia, and Libya. The impacts of climate change on tourism are unclear due to other non-climatic aspects of tourism, such as changes in travel • the Mashrek in the East: Egypt, Jordan, Lebanon, Iraq, and costs, demand, and options for tourism destinations. Syria. • the Arab Gulf States, defined here as member states of the 4.1.3.6 Energy Systems “Cooperation Council for the Arab States of the Gulf” (and not Three types of climate-change-related stressors could potentially as the countries bordering the Persian Gulf): Iraq, Kuwait, Bah- affect thermal power generation and hydropower generation: rain, Oman, Qatar, Saudi Arabia, and the United Arab Emirates. (1) Increased air temperatures could reduce thermal conversion efficiency; (2) changes in the water regime and water temperatures • the least developed countries (LDC) with the lowest indicators may decrease the available volume and decrease efficiency of of socioeconomic development following a definition from the water for cooling; and (3) extreme weather events could affect the United Nations: Yemen and Djibouti. production plants and the distribution systems. Regional studies The region also includes the Islamic Republic of Iran, Israel, published in English that quantify the impacts of climate change and the West Bank and Gaza. on thermoelectricity generation in the Middle East and in North MENA is one of the most diverse regions in the world from a Africa appear to be lacking. For North Africa, one study projects socioeconomic point of view as illustrated by the wide spectrum that hydropower production will decrease by almost 0.5 percent of per-capita annual GDP. It ranges from $1,000 in Yemen to more with 2°C warming compared to 2005 production levels due to than $20,000 in the Arab Gulf States. Qatar, Kuwait, the United Arab changes in river runoff. In the same study the production is pro- Emirates, Morocco, Egypt, and Yemen rank 4, 12, 27, 130, 132, jected to decrease by 1.4 percent in the Middle East. and 151 in GDP per capita on a list of 189 countries (Table 4.1). Accordingly, adaptive capacity and vulnerability to climate change 4.1.4 Overview of Regional Development and other risks also vary greatly among MENA countries. Narratives The main vulnerabilities of the MENA region related to climate change impacts are highlighted below. The development narratives build on the climate change impacts There is growing demand for water, food, and other agricul- analyzed in this report (Table 4.10) and are presented in more detail tural products, driven by a growing population but with strong in Section 4.5. The Middle East and North Africa region is one of sub-regional variances. While Qatar, Oman, the United Arab the world’s most climate vulnerable regions. With its high and Emirates, Kuwait, and Bahrain rank number 1, 2, 3, 4, and 10 in growing import dependency, the region is particularly vulnerable to population growth among 205 countries, with annual growth rates worldwide and domestic agricultural impacts and related spikes in ranging between 3 and 9 percent, Morocco, Iran, Tunisia, Libya, food prices. While never mono-causal, such climate-related market and Lebanon rank only 99, 104, 115, 130, and 132 on that same list, signals may fuel the potential for social unrest and migration and with annual growth rates between 1.0 and 1.3 percent (World Bank have a lasting effect on poverty in the region. Both the rural poor 2013h). Population is projected to double by 2050 (Verner 2012). and the urban poor would be hard hit by agricultural impacts, as poor farmers in rural areas are particularly vulnerable to hunger and Low-Economic Diversification malnutrition and the urban poor are hit hard by rising food prices. Agriculture employs more than 35 percent of the MENA population While biophysical impacts vary only slightly across the region, and contributes 13 percent to the region’s GDP (the global aver- there is also a clear division in vulnerabilities and socioeconomic age is 3.2 percent) (Verner 2012); if the agricultural value chain impacts between the (oil-) rich Arab Gulf States and other countries is accounted for, then this percentage more than doubles (Valdés in the region. The former have the financial means to afford adap- and Foster 2010). Agriculture generally depends more strongly on tation options, such as desalination technology and food imports. natural resources and climate than other sectors. Accordingly, the 117 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Syrian Arab Rep. Tunisia Lebanon I.R. of Iran Morocco West Bank and Gaza Iraq Kuwait Jordan Bahrain Qatar Algeria Arab Libya Saudi Rep. of Arabia Egypt United Arab Emirates Oman IBRD 41282 OCTOBER 2014 Rep. Of This map was produced by the Map Design Unit of The World Bank. The boundaries, colors, denominations and any other information Yemen shown on this map do not imply, on the part of The World Bank Group, any judgment on the legal status of any territory, or any Djibouti endorsement or acceptance of such boundaries. region’s economy also strongly depends on water, as well as on regions. On the other hand, food imports reduce the region’s direct climate conducive to food crop growth. As climatic conditions are vulnerability to local climate risks. The Nomura Food Vulnerabil- projected to become increasingly unfavorable, natural resources— ity Index40 (Nomura Global Economics and Strategy 2010) ranks in particular land and water—will become overexploited and Morocco, Algeria, Lebanon, and Egypt as the 2nd, 3rd, 5th, and degraded. The fraction of total employment made up by agriculture 6th most vulnerable among 80 countries listed. ranges from 1.5 percent in Bahrain and three percent in Jordan to 32 percent in Egypt, 41 percent in Morocco, and 54 percent High Electricity and Water Consumption in Yemen (2013b). Another aspect of the region’s low economic Demand-side measures do not feature prominently, as illus- diversification is its strong dependence on gas and oil revenues, trated by considerable over-consumption of various goods and which make up 38 percent of total regional GDP (Arab Monetary services—in particular among the Arab Gulf States. The Gulf Fund 2010). With decreasing availability of fossil fuels, export States hold world records in energy-intensity, per capita electric- revenues and adaptive capacity may suffer. ity consumption, greenhouse gas emissions, and domestic water demand. Despite its extreme water scarcity, the Arab region uses Dependence on Import more water per capita than the global average, due in consider- The region strongly depends on food and associated virtual water able part to very low resource use efficiencies of Arab residential imports (water embedded in the trade of agricultural commodities— water and energy markets as they are among the most heavily see Section 4.4.1, The Agriculture-Water-Food Security Nexus). subsidized in the world. Electricity consumption per capita is About 50 percent of regional wheat and barley consumption, twice as high as or higher than the world average in Bahrain, 40 percent of rice consumption, and nearly 70 percent of maize Israel, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab consumption is met through imports (Verner 2012), and the world’s Emirates (World Bank, 2013e), and water withdrawal per capita top nine wheat importers are MENA countries. High levels of per is nearly twice as high as the world average in Bahrain, Iraq, capita net food imports are due to severe domestic water (and and Kuwait (FAO 2013). land) constraints (i.e. low resource availability in combination with often low resource productivity). Import dependency results in a high vulnerability to fluctuating global food prices, which to 40 The index is composed of the indicators net food trade, household spending on some extent are also driven by climate impacts in food-exporting food, and per capita GDP. 118 MI DDLE E AST A ND NOR TH A FRICA Table 4.1: Basic socioeconomic indicators of MENA countries. URBAN LIFE URBAN POPULATION AGRICULTURE, EXPECTANCY INDICATOR POPULATION POPULATION GROWTH GDP PER CAPITA VALUE ADDED1 AT BIRTH² UNIT MILLION % OF POPULATION ANNUAL % CURRENT $1000 % OF GDP YEARS YEAR 2011 2012 2012 2012 2009–2010 2011 ID SP.POP.TOTL SP.URB.TOTL.IN.ZS SP.URB.GROW NY.GDP.PCAP.CD NV.AGR.TOTL.ZS SP.DYN.LE00.IN Arab Gulf States Bahrain 1 89 2.0 22 - 76 Kuwait 3 98 4.0 51 - 74 Oman 3 74 9.5 23 - 76 Qatar 2 99 7.2 90 - 78 Saudi Arabia 28 82 2.1 25 2.4 75 United Arab Emirates 9 85 3.4 39 0.9 77 Least Developed Countries Djibouti 1 77 1.6 - - 61 Yemen, Rep. 23 33 4.1 1 7.7 63 Maghreb Algeria 38 74 3.0 5 6.9 71 Libya 6 78 1.1 103 75 Morocco 32 57 2.1 3 15.4 70 Tunisia 11 67 1.3 4 8.0 75 Mashrek Egypt 79 44 2.0 3 14.0 71 Iraq 32 66 2.5 6 - 69 Jordan 6 83 2.5 5 3.4 74 Lebanon 4 87 1.1 10 5.6 80 Syrian Arab Republic 22 56 2.7 3 23.0 75 Other MENA Countries Iran, Islamic Rep. 75 69 1.5 7 - 73 Israel 8 92 1.9 33 - 82 West Bank and Gaza 4 75 3.3 - - 73 World 6,966 53 2.1 10 3.2 71 1 Agriculture corresponds to ISIC divisions 1–5 and includes forestry, hunting, and fishing, as well as cultivation of crops and livestock production. Value added is the net output of a sector after adding up all outputs and subtracting intermediate inputs. It is calculated without making deductions for depreciation of fabricated assets or depletion and degradation of natural resources. ²Life expectancy at birth indicates the number of years a newborn infant would live if prevailing patterns of mortality at the time of its birth were to stay the same throughout its life. 3 in 2009. Source: World Bank (2013b). 119 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Governance Underemployment Challenges and uncertainties associated with climate change, in High fertility rates in rural areas and technological changes are combination with other pressures, require diverse, adaptable, driving the movement of labor from the agricultural sector, leading and participatory governance structures (Folke 2003). Judging to rural migration and urban population growth. Forty percent of from commonly used indicators of democracy, corruption, and the region’s unemployed are young job seekers (UNESCWA 2007). gender equity, the region may not be well prepared in this respect. According to the Economist’s democracy index41 of 167 countries, Undernourishment and Malnutrition the MENA countries rank very poorly at positions 90 (Tunisia), The number of undernourished people in the region45 in 2011–2013 149 (United Arab Emirates), 150 (Bahrain), 158 (Iran), 163 (Saudi reached 21 million in Western Asia (about 10 percent of the popu- Arabia), and 164 (Syria) (The Economist 2012). UNDP’s gender lation) and 4 million in Northern Africa (about 3 percent of the inequality index42 (UNDP 2012), which lists 148 countries, ranks population), higher than in 2008–2010 (FAO 2013); on the other Qatar at position 117, Syria at 118, Iraq at 120, Egypt at 126, Saudi hand, an estimated one quarter of the population is obese.46 Arabia at 145, and Yemen at 148. The representation of women in All MENA countries face severe and fast-growing resource Arab governments is nine percent, or just half of the global aver- squeezes, in particular relating to extreme water and land scarcity; age (Verner 2012). According to the corruption perception index43 this leaves most of these countries unable to produce enough food (Transparency International 2012), which lists 174 countries, for their population. The main focus of this chapter is therefore Syria, Yemen, Libya, and Iraq are ranked at positions 144, 156, on agriculture and water resources. 160, and 169 respectively. Regarding the economic governance The most recent IPCC Assessment report found that a reduction of the region, large-scale and centralized supply-side measures in precipitation by the end of the 21st century is likely over Northern are seen as solutions to the growing demand-supply gap, such Africa under the A1B and A2 scenarios, and that land temperatures as large-scale water transfers in Tunisia and Libya. There is less over Africa are likely rising faster than the global average. There is emphasis on diverse, distributed, decentralized, and small-scale high confidence that climate change will amplify existing stresses solutions, which would increase the region’s flexibility, diversity, on water availability across the whole of Africa; acting together and resilience to climate change (Folke 2003; Sowers et al. 2011). with non-climate drivers, this will exacerbate the vulnerability of agricultural systems, particularly in semi-arid regions (Niang et al. Low Investment in Research and Development 2014). Water scarcity is expected to be a major challenge in West Innovation in supply-and-demand-side management and adaptation Asia, including the Arabian Peninsula and other MENA countries, to climate change and other risks requires research and develop- due to increased water demand and lack of good management ment. While globally countries spend on average 1.7 percent of (medium confidence), although the impacts of climate change on GDP on research and development, the Arab region44 ranks lowest food production and food security in West Asia will vary by region. of all the world’s regions at 0.2 percent in 2007 (UNESCO 2010). Studies on the regional impacts of climate change are still inadequate, Private investment in the region as a share of total investment including for the Arab Gulf States and parts of the Mashrek (e.g. is around 40–45 percent (lower than Africa, Latin America, and there is only limited information and knowledge gaps on the impacts the Caribbean, where the share of total investment is around on crops and farmland remain) (Hijioka et al. 2014). 75–80 percent) (Verner 2012). 4.3 Regional Patterns of Climate Change 41 The index, on a 0 to 10 scale, is based on the ratings for 60 indicators grouped in five categories: electoral process and pluralism; civil liberties; the functioning of 4.3.1 Projected Temperature Changes government; political participation; and political culture. Each category has a rating on a 0 to 10 scale, and the overall index of democracy is the simple average of the The projected increase in temperatures during the boreal summer five category indexes. 42 (June, July, August or JJA) over the Middle East and North African A composite measure reflecting inequality in achievements between women and men in three dimensions: reproductive health, empowerment, and the labor market. Data land area is shown in Figure 4.3 for both the 2°C and 4°C world. are those available to the Human Development Report Office as of October 15, 2012. 43 The index ranks countries and territories based on how corrupt their public sec- 45 tor is perceived to be. A country or territory’s score indicates the perceived level of The FAO region Near East and North Africa: Algeria, Egypt, the Islamic Republic public sector corruption on a scale of 0- to 100, where 0 indicates that a country of Iran, Iraq, Jordan, Kuwait, Lebanon, Libya, Mauritania, Morocco, Saudi Arabia, is perceived as highly corrupt and 100 indicates it is perceived as not corrupt. A Sudan, Syria, Tunisia, United Arab Emirates, and Yemen. 46 country’s rank indicates its position relative to the other countries and territories According to WHO. Data for adults aged 15 years and older from 16 countries in the included in the index. region show the highest levels of overweight and obesity in Egypt, Bahrain, Jordan, 44 The Arab region in the UNESCO report is composed of Algeria, Djibouti, Egypt, Kuwait, Saudi Arabia, and United Arab Emirates. The prevalence of overweight and Libyan Arab Jamahiriya, Mauritania, Morocco, Sudan, Tunisia, Bahrain, Iraq, Jordan, obesity in these countries ranges from 74–86 percent in women and 69–77 percent Kuwait, Lebanon, Oman, West Bank and Gaza, Qatar, Saudi Arabia, Syria, United in men. A person with a BMI of 25 or more is considered by WHO to be overweight, Arab Emirates, and Yemen. while obesity is defined as having a BMI of 30 or more. 120 MI DDLE E AST A ND NOR TH A FRICA Figure 4.3: Temperature projections for the Middle East and The multi-model mean warming by 2100 is about 2.5°C in a 2°C North African land area compared to the baseline (1951–1980). world and about 7.5°C in a 4°C world, which is substantially more than the global mean land warming (see Figure 2.5 in World Bank 2013). Under the low-emissions scenario (i.e., a 2°C world), sum- mer temperatures in MENA peak by 2040 at about 2.5°C above the 1951–1980 baseline and remain at this level until the end of the century. In a 4°C world, warming continues almost linearly beyond 2040, reaching about 7.5°C above the 1951–1980 baseline by 2100 (Figure 4.3). Geographically, the strongest warming is projected to take place close to the Mediterranean coast (see Figure 4.4). Here, and also inland in Algeria, Libya, and large parts of Egypt, regions warm by 3°C in a 2°C world. In a 4°C world, mean summer temperatures in 2071–2099 are expected to be up to 8°C warmer in parts of Algeria. Warming over the Sahel region (i.e., below about 20°N in Figure 4.4) is more moderate (2°C in a 2°C world Temperature projections for the Middle East and North African land area and 5°C in a 4°C world), which is likely related to an increase in compared to the baseline (1951–1980) for the multi-model mean (thick line) precipitation (see Figure 4.7). and individual models (thin lines) under RCP2.6 (2°C world) and RCP8.5 (4°C world) scenarios for the months of JJA. The multi-model mean has The lower panels of Figure 4.4 show the normalized warm- been smoothed to give the climatological trend. ing (i.e., the warming expressed in terms of the local year-to-year natural variability—see Section 6.1, Methods for Temperature, Precipitation, Heat Wave, and Aridity Projections) over the Middle Figure 4.4: Multi-model mean temperature anomaly for RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) for the months of JJA for the Middle East and North African region. Temperature anomalies in degrees Celsius (top row) are averaged over the time period 2071–2099 relative to 1951–1980, and normalized by the local standard deviation (bottom row). 121 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL East and North African land area. The normalized warming is a over North Africa, the Eastern Mediterranean, and the Middle useful diagnostic tool as it indicates how unusual the warming is East (Donat et al. 2013; Donat, Peterson et al. 2013; Hoerling et compared to fluctuations experienced in the past in a particular al. 2012; Kuglitsch et al. 2010). region (Coumou and Robinson 2013; Hansen et al. 2012; Mora and As expected from the large shifts in normalized warming, the Frazier et al. 2013). In a 4°C world, the probability density function number of threshold-exceeding extremes is expected to strongly of monthly temperatures (associated with the year-to-year vari- increase across the Middle East and North African region under ability of monthly temperatures) shifts by six standard deviations projected future warming (Figure 4.5 and Figure 4.6). In a 2°C world, toward warmer conditions across all regions, from the Sahara to by the end of the century about 30 percent of summer months will the Arabian Peninsula to the eastern Mediterranean coast. Such a be hotter than 3-sigma (see Section 6.1, Methods for Temperature, large shift implies that summer temperatures here will move to a Precipitation, Heat Wave, and Aridity Projections) almost every- new climatic regime by the end of the 21st century. Such a dramatic where in this region. This implies that every year on average one change would be avoided in a 2°C world; even then, however, a of the summer months (June, July, or August) is expected to exceed substantial shift is expected (i.e., by about 2–3 standard deviations). temperatures by more than three standard deviations beyond the 1951–1980 mean. This value is substantially higher than the global 4.3.2 Heat Extremes mean projections (20 percent of summer months) (see Figure 2.7 in World Bank 2013). The Mediterranean and Middle East are particu- An increase in the number of extremely high temperatures since the larly prone to an increase in extreme temperatures and heat waves, 1960s has been detected, as would be expected for an increase in because temperature extremes in these regions are expected to be global mean temperatures over the same period (Seneviratne et al. amplified by a reduction in soil moisture, resulting from decreasing 2012). An increase in the warm spell duration index (WSDI47) for precipitation (Lorenz et al. 2010; Orlowsky and Seneviratne 2011; heat waves has also been observed and is particularly pronounced Vautard et al. 2013). Summer months warmer than 5-sigma, however, will remain largely absent, except for some isolated coastal regions 47 The Warm spell duration index (WSDI) is one of the 27 indices developed and (including the Mediterranean coasts of Egypt and over Yemen, recommended by the WMO CCl/CLIVAR/JCOMM Expert Team on Climate Change Djibouti, and Oman). Here 5-sigma events will still be relatively Detection and Indices (Zhang et al. 2011b). It is defined as the longest annual spell rare in a 2°C world but are nevertheless detectable and expected of at least six consecutive days with maximum temperatures exceeding the local 90th percentile relative to a reference period (in days). to occur between 5–10 percent of summer months. Figure 4.5: Multi-model mean of the percentage of boreal summer months in the time period 2071–2099, with temperatures greater than 3-sigma (top row) and 5-sigma (bottom row) for scenarios RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) over the Middle East and North Africa. 122 MI DDLE E AST A ND NOR TH A FRICA Figure 4.6: Multi-model mean (thick line) and individual Table 4.2: Mean WSDI (Warm Spell Duration Index) for capital models (thin lines) of the percentage of Middle East and North cities in the MENA region for different levels of global warming African land area warmer than 3-sigma (top) and 5-sigma based on regional climate model projections by Lelieveld et al. (bottom) during boreal summer months (JJA) for scenarios (2013). RCP2.6 (2°C world) and RCP8.5 (4°C world). WARMING LEVEL [°C] PERIOD OF OBSERVATION OBSERVED 1.5 2.0 3.0 4.0 Abadan 1951—2000 6 43 82 99 134 Amman 1959–2004 4 31 62 84 115 Ankara 1926–2003 7 44 67 111 128 Athens 1951–2001 1 40 61 121 166 Baghdad 1950–2000 8 47 90 113 162 Beirut – – 47 93 126 187 Belgrade 1951–2010 9 39 39 76 113 Cairo – – 32 53 80 94 Damascus 1965–1993 1 36 71 98 129 Istanbul 1960–2010 0 26 41 78 113 Jerusalem 1964–2004 7 26 46 73 102 Kuwait – – 45 87 123 167 Nicosia 1975–2001 6 25 58 81 162 Riyadh 1970–2004 3 81 132 157 202 Sofia 1960–2010 1 40 49 88 136 Tehran 1956–1999 5 48 92 122 159 Tirana 1951–2000 6 49 71 125 168 Tripoli 1956–1999 3 13 22 33 59 The observational record is taken from Klok and Tank (2009). The annual WSDI is averaged over the observational period as well as over 20-year periods from the A1B scenario over the 21st century that correspond to a mean warming of the given temperature level. The 4.0°C warming level is averaged over the 2079–2099 interval from the A2 scenario. The increase in frequency of heat extremes is projected to plateau by mid-century in a 2°C world. In a 4°C world, however, in some isolated coastal areas (The Egyptian Red Sea coast and heat extremes are projected to become ever more frequent up to the Mediterranean coasts of Egypt, Libya, and Tunisia) is the and beyond the end of the century (see Figure 4.5). This is simi- frequency higher than the regional mean during the 2071–2099 lar to the timing of the mean summer warming (see Figure 4.3). period (i.e., close to 80 percent of summer months). The large The multi-model mean projection indicates that in a 4°C world shift in normalized temperatures to warmer conditions (i.e., by 80 percent of summer months will be hotter than 5-sigma by 2100 about six standard deviations in a 4°C world, see Figure 4.4) (Figure 4.6), with about 65 percent of summer months reaching will cause almost all summer months to be warmer than 3-sigma this level during the 2071–2099 period (Figure 4.5). Figure 4.6 also (i.e., more than 90 percent of summer months for the 2071–2099 clearly shows that there is substantial inter-model spread, with period in a 4°C world). the area of land experiencing 5-sigma events by 2100 ranging from These projections in heat extremes are consistent with published 30 percent to almost 100 percent in different models. The inter- analyses based upon the full CMIP5 dataset of climate projections. model spread for projected mean summer temperatures is much These studies indicate that over the Mediterranean, the Sahara, more limited (see Figure 4.3) compared to that of the projected and the Arabian Peninsula, the minimum nighttime temperature frequency of 3- and 5-sigma events (see Figure 4.6). It is possible and the maximum daytime temperature increase by about 2oC that the large spread in the projected frequency of heat extremes under RCP2.6 and by 6oC under RCP8.5 by 2081–2100 compared is primarily due to the difference in simulated inter-annual vari- to 1981–2000 (Sillmann et al. 2013a; b); this is very similar to the ability in surface temperatures in the models for this particular seasonal changes shown in Figure 4.3. In addition, the increase in region. Further research would be needed to confirm this. Only the percentage of 3-sigma and 5-sigma monthly heat extremes as 123 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL shown in Figure 4.5 is consistent with the results of Coumou and months in a 2°C world. The increase in future heat wave risk as Robinson (2013), who analyzed all climate models in the CMIP5 indicated by the increase in the WSDI represents a serious risk for dataset. Although an above-global average increase in Mediterra- a variety of sectors, including health and tourism. nean summer temperatures is robust over a wide range of regional and global climate models, it might be partly overestimated due 4.3.3 Projected Precipitation Changes to systematic warming bias (Boberg and Christensen 2012). Using a high resolution regional climate model, Lelieveld et al. Annual precipitation is projected to decrease north of ~25°N (2013) projected a substantial increase in the warm-spell duration and to increase south of that latitude. It is important to note that index (WSDI) for several capital cities in the region (Table 4.2). the projections shown in Figure 4.7 are given as relative changes The mean WSDI over the observational period lies between zero (percentage changes compared to the 1951–1980 average); thus, and about one week; it is projected to exceed four months for most especially over already dry regions, large relative changes do capital cities in the region in a 4°C world and even six months for not necessarily reflect large absolute changes. The Sahel and Beirut and Riyadh. The mean WSDI would be limited to between the southern part of the Arabian Peninsula bordering the Indian 1–2 months for most capital cities and would rarely exceed three Ocean (Yemen, Oman) are projected to become wetter in both a Figure 4.7: Multi-model mean of the percentage change in winter (DJF, top), summer (JJA, middle) and annual (bottom) precipitation for RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) for the Middle East and North Africa by 2071–2099 relative to 1951–1980. XXX XXXXXXX XXX XXX XXXXXXX XXX XXX XXX XXX XXX XX X XXXXXXX XXXXXX XXXXXX XXXX XXXXX Hatched areas indicate uncertain results, with two or more out of five models disagreeing on the direction of change. Note that projections are given as relative changes; thus, over dry regions like the Sahel/Sahara, large relative changes reflect only small absolute changes. 124 MI DDLE E AST A ND NOR TH A FRICA 2°C and 4°C world, primarily during the summer months (JJA). that is associated in the observational record with extreme heavy The IPCC AR5 reports similar seasonal precipitation changes for precipitation events in the Middle East (de Vries et al. 2013). the Sahel, but the wetting region is substantially smaller (limited While the projected changes in extreme precipitation events are to the southeast only) and model disagreement is larger. Thus, below the global average for Saudi-Arabia and Iran, the southern tip the projected JJA wetting of the Sahel as shown here should be of the Arabian Peninsula (notably Oman and Yemen) is projected treated with caution due to the limited number of models used. to experience a substantial intensification of extreme precipitation During the winter months (DJF), the eastern part of the Sahel events over the 21st century under the RCP8.5 scenario (Kharin et region in fact becomes drier. Wetting of the southern Arabian al. 2013; Sillmann et al. 2013b). This is consistent with the robust Peninsula is part of a larger and well-documented pattern of wetter projection for increasing annual precipitation over the Horn of Africa. conditions over the Horn of Africa expected under future climate The North African countries (and notably Morocco, Algeria, change (Giannini et al. 2008; World Bank 2013). Rainfall in this and Tunisia), as well as the countries of the Middle East, are con- region is associated with tropical convection, which is projected sistently projected to become global hotspots for drought by the to strengthen under future warming. Future wetting in this region end of the 21st century under the RCP8.5 (Dai 2012; Orlowsky and is also supported by paleo-climate data, and it is likely to intensify Seneviratne 2013; Prudhomme et al. 2013; Sillmann et al. 2013b). inter-annual rainfall variability (Wolff et al. 2011). Dai (2012) projects severe drought conditions for Morocco and In a 2°C world, the relative wetting of regions south of 25°N is the Middle East under the RCP4.5 scenario, which is found to pronounced (~50 percent more rain), whereas the relative drying strongly intensify under RCP8.5 (Prudhomme et al. 2013). Based to the North is small (with the Mediterranean coasts receiving about on the comprehensive ISI-MIP modeling framework, Prudhomme 10–20 percent less rain annually). In fact, the climate models used et al. (2013) report an increase of more than 50 percent in the here disagree on the direction of change over substantial areas, number of drought days around the Mediterranean by the end of notably between 25°N and 30°N. This is true for both seasons. 21st century (2070–2099) under the RCP8.5 scenario relative to the However, the larger set of CMIP5 models used for the IPCC AR5 1976–2005 period. For the same region, Orlowsky and Seneviratne project robust drying of the Mediterranean coastal regions. (2013) project an average of more than six months per year with at In a 4°C world, the magnitude of change in the models used least moderate drought conditions in the 2080–2100 period under becomes much more pronounced in the region north of 25°N and the the RCP8.5 scenario, compared to less than one month per year models converge toward drier conditions. Only over Saudi Arabia, under RCP2.6. Although projections of future droughts not only the Islamic Republic of Iran, and parts of Libya do the projected suffer from large model uncertainties—and also largely depend relative changes in annual-mean precipitation remain small, and on the methodology and baseline periods chosen (Trenberth et al. models show disagreement on the direction of change. Clearly, in 2014)—the projections for an increase in extreme drought condi- a 4°C world, countries along the Mediterranean shore (notably tions around the Mediterranean, Northern Africa, and the Middle Morocco, Algeria, Egypt, and Turkey) will receive substantially less East are consistent across a variety of studies (IPCC 2012). Drought rain annually (up to 50 percent less precipitation). In this region projections for the Mediterranean are strongly dependent on the relative changes in seasonal precipitation are similar to changes in emissions scenario used, and are much less pronounced under the annual mean, indicating that the drying happens year-round. RCP2.6 (Orlowsky and Seneviratne 2013; Prudhomme et al. 2013). In the Islamic Republic of Iran, a weaker intensification of 4.3.4 Extreme Precipitation and Droughts future droughts is expected compared to projections for Northern Africa and the Middle East (Dai 2012; Orlowsky and Seneviratne The observational record for North Africa, the Middle East, and 2013; Prudhomme et al. 2013); uncertainty in this region, however, the Arabian Peninsula indicates an overall reduction in extreme is large (Hemming et al. 2010) and these projections might not precipitation events since the 1960s, despite a local positive trend be statistically robust (Sillmann et al. 2013b). For the Arabian over the Atlas mountains since the 1980s (Donat and Peterson et al. Peninsula, a possible reduction in future droughts, or at least no 2013). At the same time, an increase in meteorological drought has further intensification in the already-extreme drought conditions, been reported since the 1960s, consistent with an overall regional is projected by Dai (2012). drying trend (Donat and Peterson et al. 2013; Sousa et al. 2011). Despite a global trend toward more extreme precipitation events 4.3.5 Aridity over the 21st century, Kharin et al. (2013) conducted an analysis using CMIP5 models under the RCP8.5 scenario and reported no The availability of water for both people and ecosystems is a significant change or even a slight decrease in heavy precipitation function of supply and demand. The long-term balance between across most of North Africa and the Middle East. It is not clear, supply and demand fundamentally determines whether ecosystems however, to what extent these models are able to reproduce rare and agricultural systems are able to thrive in a certain area. This climatological phenomena like the Active Red Sea Trough (ARST) section assesses projected changes in the aridity index (AI), an 125 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 4.8: Multi-model mean of the percentage change in the annual-mean (ANN) of monthly potential evapotranspiration for RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) for the Middle East and North African region by 2071–99 relative to 1951–80. Hatched areas indicate uncertain results, with two or more out of five models disagreeing on the direction of change. indicator designed to identify regions with a structural precipita- in regions directly surrounding the Mediterranean Sea. In a 4°C tion deficit (Zomer et al. 2008). AI is calculated as the total annual world, a more substantial increase is observed, again especially precipitation divided by the annual potential evapotranspiration—a in countries bordering the Mediterranean Sea. standardized measure of water demand (see Section 6.1.3, Aridity There is a close match between the pattern of change in Index and Potential Evaporation). Although several meteorological the annual mean AI value (Figure 4.9) and projected precipita- variables affect potential evapotranspiration, including radiation tion changes (Figure 4.7). This indicates that changes in AI are and near-surface wind speed, it is to a large extent governed by primarily driven by changes in precipitation causing an increase changes in temperature. A smaller AI value indicates a larger water in AI (wetter conditions) south of 25°N (i.e., the Sahel and the deficit (i.e. more arid conditions), with areas classified as hyper- most southern part of the Arabian Peninsula) and a decrease in arid, arid, semi-arid, and sub-humid as specified in Table 6.1). AI (drier conditions) north of 25°N. The relative increase in AI Annual-mean monthly potential evapotranspiration is projected values in the southern region is similar to the relative increase in to increase throughout the region with robust model agreement. annual mean precipitation (about 50 percent wetter conditions), The magnitude of the relative changes forms rather uniform geo- as the change in potential evapotranspiration is small. Note that graphical patterns (Figure 4.8). There is a small increase under this relative increase in AI south of 25°N is imposed on an already RCP2.6, typically of only 10 percent although somewhat more very low AI value, which results in AI values still classified as arid. Figure 4.9: Multi-model mean of the percentage change in the aridity index under RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) for the Middle East and North Africa by 2071–2099 relative to 1951–1980. XXX XX XXXXXX X XXXXX Hatched 48 areas indicate uncertain results, with two or more out of five models disagreeing on the direction of change. Note that a negative change cor- responds to a shift to more arid conditions.48 48 Some individual grid cells have noticeably different values than their direct neighbors (e.g. on Turkey’s Black Sea coast under RCP8.5). This is due to the fact that the aridity index is defined as a fraction of total annual precipitation divided by potential evapotranspiration (see Appendix). It therefore behaves in a strongly non-linear fashion and year-to-year fluctuations can be large. As the results are averaged over a relatively small number of model simulations, this can result in local jumps. 126 MI DDLE E AST A ND NOR TH A FRICA In the Mediterranean coastal region, the percentage decrease in 2008). For that reason, this report notes that the steric-dynamical49 AI is more pronounced than that for precipitation because there is section of the analysis of the new CMIP5 ensemble is explorative. a substantial increase in evapotranspiration here due to enhanced In the Mediterranean area, tide gauges recorded below-average warming. In other words, the opposing trends of more evapotrans- sea-level rise during the 20th century, with an average rise of piration and less annual-mean precipitation mean that this region 1.1–1.3 mm per year (Tsimplis and Baker 2000); this is lower than is expected to see a shift to much more arid conditions. In fact, this the global average of 1.8 mm per year (Meehl et al. 2007). There is linked to a feedback between precipitation and evaporation via has been significant inter-decadal variability, with reduced rise temperature. In regions where the soil dries out because of a decline between 1960 and 1990 and rapid (above-average) rise after 1990. in precipitation, less heat can be converted into latent heat and thus In the 2009/2010 and 2010/2011 winters, sea levels rose 10 cm above more heat is present to warm surface temperatures (Coumou and the seasonal average. Atmospheric influence is thought to be the Rahmstorf 2012; Schär et al. 2004). Higher surface temperatures then primary driver, where pressure and wind variations associated with lead to enhanced potential evapotranspiration. Thus, a decline in the North Atlantic Oscillation control water flow through the Gibral- precipitation can make local conditions more arid in two ways: directly tar Strait (Gomis et al. 2006; Landerer and Volkov 2013; Tsimplis et via a reduction in the supply of water, and indirectly via an increase al. 2013). Compared to the well-studied Mediterranean basin, tide in surface temperatures which further enhances evapotranspiration gauge records in the Red Sea and Arabian Sea are much sparser— (Trenberth 2011). This results in a reduction of up to 30 percent in limited to a few, non-continuous records.50 Available evidence from AI (increased aridity) over the African and eastern shores of the the neighboring Northern Indian Ocean, also including satellite Mediterranean in a 2°C world; in a 4°C world, essentially all coastal altimetry and modeling, suggests past rates of rise consistent with regions will see a reduction in AI of about 50 percent. Trends in AI the global mean (Han et al. 2010; Unnikrishnan and Shankar 2007). over the Arabian Peninsula are weaker and models disagree on the This report’s analysis of the 21st century projections indicates direction of change due to the uncertainty in precipitation changes. slightly below-average rise in the Mediterranean basin (Figure 4.10 The shift in AI in Figure 4.9 translates into a shift of categoriza- and Figure 4.11 top panel), mostly as a result of the gravitational tion of areas into specific aridity classes. The total hyper-arid area is influence of the Greenland ice sheet (Figure 4.11 bottom panel). projected to grow by about five percent in a 4°C world, something In our GCM ensemble, this is compensated by the above-average which would mostly be avoided in a 2°C world (see Table 4.3). steric-dynamic contribution in this area (but lower than in the Less than half of this increase in hyper-aridity can be explained by neighboring Atlantic Ocean) (Figure 4.11 middle panel). The com- a reduction in arid, semi-arid, and sub-humid areas, indicating that bined gravity- and steric-dynamic pattern induces a stronger rise humid regions (i.e., AI > 0.65) have shifted into the arid classification. in the Arabian Sea compared to the Mediterranean (see Figure 4.10 and Figure 4.11). Consistent with this finding, Tunis is projected to 4.3.6 Regional Sea-level Rise experience 0.56 m (0.38–0.96 m) sea-level rise by the end of the century (2081–2100) compared to 1986–2005 in a 4°C world (see Sea-level rise projections for MENA present a particular challenge Figure 4.12 and Table 4.4). This is eight cm less than in Muscat, due to the semi-enclosed nature of both the Mediterranean and Red where 0.64 m is projected (0.44–1.04 m). On the Atlantic coast, Sea basins. They are connected to the broader Atlantic and Indian 0.58 m (0.39–0.98 m) sea-level rise is projected for Tangier. Across Oceans, respectively, through relatively narrow straits which are not all the locations present in the figures, the projected high-end well represented in the coarse resolution of GCMs (Marcos and Tsimplis rates of sea-level rise range from 6.4 mm per year (Alexandria) to 7.8 mm per year (Tangier) in a 1.5°C world and from 20 mm per year (Tunis) to 21.4 mm per year (Alexandria) in a 4°C world Table 4.3: Multi-model mean of the percentage of land area in (Table 4.5). This is comparable with global mean projections (up the Middle East and North African region which is classified as to 7.2 mm per year in a 1.5°C world and 21.9 mm per year in a hyper arid, arid, semi-arid and sub-humid for 1951–1980 and 4°C world by the end of the century). 2071–2099 for both the low (2°C world, RCP2.6) and high (4°C Note that this report excludes the steric-dynamic component world, RCP8.5) emissions scenarios. of three models (MIROC-ESM, HadGEM-ES, CSIRO-Mk3-6-0) with 2071–2099 2071–2099 obvious deficiencies in this area (e.g., projecting more than one meter 1951–1980 (RCP2.6) (RCP8.5) deviation from the global mean or from the neighboring oceans, Hyper Arid 68.9 70.7 74.2 or simulating a reversed sea-level gradient across Gibraltar). The remaining 5 models simulate a present-day sea-level difference of Arid 13.2 13.0 12.4 Semi-Arid 7.1 7.0 6.3 Sub-Humid 1.8 1.7 1.4 49 Related to changes in ocean density and circulation. 50 Tide gauge records can be found at http://www.psmsl.org/data/obtaining/map.html. 127 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 4.10: Patterns of regional sea-level rise (m). Median (left column) and upper range (right column) of projected regional sea-level rise for the RCP2.6 scenario (1.5°C world, top row) and the RCP8.5 scenario (4°C world, bottom row) for the period 2081–2100 relative to the reference period 1986–2005. Associated global mean rise is indicated in the panel titles. Figure 4.11: Regional sea-level rise anomaly pattern and its Table 4.4: Sea-level rise between 1986–2005 and 2080–2099 contributions to the median RCP8.5 scenario (4°C world). in selected MENA locations (m). RCP2.6 (1.5°C WORLD) RCP8.5 (4°C WORLD) Tangier 0.35 (0.21, 0.57) 0.58 (0.39, 0.98) Tunis 0.34 (0.20, 0.57) 0.56 (0.38, 0.96) Alexandria 0.35 (0.21, 0.57) 0.58 (0.40, 1.00) Muscat 0.39 (0.22, 0.64) 0.64 (0.44, 1.04) Numbers in parentheses indicate low and high bounds (see Section 6.2, Sea-Level Rise Projections for an explanation of the 1.5° world). Table 4.5: Rate of sea-level rise in MENA between 2080–2100 (mm per year). RCP2.6 (1.5°C WORLD) RCP8.5 (4°C WORLD) Tangier 3.3 (1.8, 7.8) 10.2 (6.8, 21.0) Tunis 3.5 (1.2, 6.6) 10.1 (6.4, 20.0) Alexandria 3.4 (1.2, 6.4) 10.9 (6.9, 21.4) Muscat 4.2 (1.5, 6.9) 12.0 (8.9, 20.6) Numbers in parentheses indicate low and high bounds (see Section 6.2, Sea-Level Rise Projections for an explanation of the 1.5° world). Total sea-level rise (top), steric-dynamic (middle), and land-ice (bottom) contributions to sea-level rise are shown as anomalies with respect to the global mean sea-level rise. Global mean contributions to be added on top of the spatial anomalies are indicated in the panel titles. 128 MI DDLE E AST A ND NOR TH A FRICA Figure 4.12: Sea-level projections for Tangier, Tunis, and Alexandria. Time series for sea-level rise for the two scenarios RCP2.6 (1.5°C world, blue) and RCP8.5 (4°C world, green). Median estimates are given as full thick lines and the lower and upper bound given as shading. Full thin lines are global median sea-level rise with dashed lines as lower and upper bounds. Vertical and horizontal black lines indicate the reference period and reference (zero) level. 11–30 cm between the neighboring Atlantic Ocean and Mediterra- basin than along the neighboring Atlantic coasts, except for a nean Sea; this is consistent with observations (Rio et al. 2014). The few hotspots (such as in the Gulf of Gabes) due to the presence difference between the Mediterranean and nearby Atlantic sea-level of large tides (Marcos et al. 2009). Available evidence from past rise depends on processes at the Strait of Gibraltar which are poorly observations and modeling indicate that sea-level extremes will represented in GCMs (Marcos and Tsimplis 2008) and which are just change in line with mean sea levels (Marcos et al. 2009). beginning to be investigated with regional climate models (Artale et As mentioned in the introduction, this analysis only considers al. 2010; Gualdi et al. 2013; Somot et al. 2008). Any addition of mass absolute sea-level changes; it omits vertical land movements despite in the Atlantic, such as from melting ice sheets, would be immedi- their relevance for local planning and adaptation. Alexandria is a ately transferred through the Strait of Gibraltar; this may not be the well-known example where sediment compaction in the Nile delta case for steric-dynamic changes (including atmospheric influences). provokes land subsidence (Syvitski et al. 2009), thereby enhancing On top of these projections, decadal and shorter-term variability the effect of climate-induced sea-level rise. Alexandria is ranked will continue to occur (Calafat et al. 2012; Landerer and Volkov top among the cities with projected increased loss of local GDP 2013). Sea-level extremes are generally lower in the Mediterranean as a result of damages from sea-level rise by 2050 (Hallegatte et 129 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 4.13: Major farming systems in the MENA region. FAO Disclaimer The designations employed and the presentation of the material in the maps do not imply the expression of any opinion whatsoever on the part of FAO concerning the legal Notes: or constitutional status of any country, territory or sea area, Projection = Geographic (Lat/Long) or concerning the delimitation of frontiers. The major farming systems have been identified based on the available natural resource base (water, land, grazing areas, slope, farm size, tenure, and organization) and the dominant pattern of farm activities and household livelihoods, broadly grouped and mapped. Note that besides the systems displayed there are small but important irrigated areas which might not be visible in the figure. Source: Dixon et al. (2001b). al. 2013). Post-glacial rebound and tectonic movements may also exceed the critical threshold of 40 percent in all MENA countries provoke land subsidence or uplift in the Mediterranean area. There except Lebanon; they exceed 100 percent in Jordan, Yemen, Libya, is a sustained effort to detect these processes with a combination and most of the Arab peninsula countries (FAO-AQUASTAT 2012), of tide gauges, satellite altimetry measurement, GPS stations, and leading to groundwater resource depletion. modeling (Lambeck and Purcell 2005; Ostanciaux et al. 2012; This water scarcity prevents MENA countries from producing Wöppelmann and Marcos 2012), as well as via archaeological all required food domestically and makes the region depending information (Anzidei et al. 2011). on food imports. From the current situation of critical water and arable land scarcity, 2°C and 4°C warming scenarios will result 4.4 Regional Impacts in further increased pressures on water resources and agriculture. Egypt, Syria, and Iraq depend strongly for their water supply 4.4.1 The Agriculture-Water-Food Security Nexus and food security on precipitation in upstream countries situated in different climatic zones. Egypt depends primarily on Ethiopia (via the 4.4.1.1 Water Scarcity and Climatic Limitations Nile), and Syria and Iraq depend to a large extent on Turkey (via the to Agricultural Production Euphrates and Tigris rivers). These two major river basins support Most of the land area of the MENA region receives less than 300 mm most of the irrigated farmland in the MENA region (Figure 4.13). annual rainfall (and 200–300 mm per year roughly represents Despite conditions of extreme water scarcity, MENA uses more the lower limit for rain-fed agriculture). Semi-arid belts along the water per capita than the global average (see Figure 4.14) due in coasts and mountains are the only water sources and provide large part to very low resource-use efficiencies, with Arab resi- productive land for rain-fed agriculture (Immerzeel et al. 2011). dential water and energy markets being among the most heavily The availability of renewable water resources is generally below subsidized in the world. 1000 m3 per capita per year (except for Iraq, the Islamic Republic The growing water scarcity prevents MENA countries from of Iran, and Lebanon), and as low as 50 m3 per capita for most producing all their required food domestically. The region as a countries on the Arab peninsula (Selvaraju 2013; Sowers et al. whole exceeded its capacity for domestic food production in the 2011; Verner 2012). Accordingly, withdrawal-to-availability ratios 1970s (Allan 2001). The growing deficit has had to be met by food 130 MI DDLE E AST A ND NOR TH A FRICA Figure 4.14: Water footprints in m3 per capita and year. Figure 4.15: Average cereal yields (kilograms per hectare) from 1961–2010 for Northern Africa and Western Asia as compared to the world average. Source: Mekonnen and Hoekstra (2011) as cited in Saab (2012). Source: FAOSTAT data in Selvaraju (2013). imports, with associated imports of virtual water (i.e., water that has been used in the production of food abroad). Across much of MENA, the conditions for rain-fed agriculture Africa, the MENA region is the only region where the number of are marginal in terms of absolute water availability and the fact undernourished has increased since the 1990s (UNDP RBAS 2009). that the months of highest temperatures coincide with lowest pre- While there are opportunities for intensification, agricultural cipitation. Despite this, around 70 percent of cropland area is rain- production in the MENA region will remain severely water-limited. fed, with even higher percentages in most Maghreb and Mashrek This is due to three key reasons: (1) that rain-fed agriculture countries. Wheat, barley, and some rice and sorghum are the main often takes place under marginal precipitation conditions, so crops (Selvaraju 2013). Between 60–90 percent of all water used in that the projected reduction in precipitation, in combination MENA countries goes into agriculture (Selvaraju 2013). The Arab with increasing temperatures, will cause conditions to pass the Gulf States depend exclusively on groundwater for irrigation, while threshold whereby irrigation would be required to maintain crop- the rest of the region (with the exception of Egypt) depends almost ping systems; (2) that blue water resources are already fully or equally on surface water and groundwater (Siebert et al. 2010). even overexploited (with competition from other sectors rapidly Irrigated agriculture is generally more productive per hectare growing) such that, in the future, water allocations for irrigation than rain-fed agriculture and, as a result, makes up more than cannot be increased or may even need to be reduced; and (3) that 50 percent of total agricultural production in the region. The pro- increasing precipitation variability and extremes will reduce the ductivity of rain-fed agriculture depends on stable and continuous reliability of the crop water supply. rainfall. Rainfall is more variable in the MENA region than almost anywhere else (Bucknall 2007); in combination with often inefficient 4.4.1.2 Climate Risks to the Water Sector systems of production, this causes cereal yields to consistently According to the IPCC WGII report, there is high agreement that remain below world averages (see Figure 4.15) (Selvaraju 2013). climate change will reduce renewable surface water and ground- As a result of low agricultural productivity and other factors water resources considerably in most dry subtropical regions, (e.g., supply chain losses, lack of access to food, and low income aggravating competition for water within and among sectors. levels), about 25 million people in the MENA region are currently Reductions in water availability will generally be greater than the undernourished—four million in Northern Africa (2.7 percent of underlying reductions in precipitation, due to increases in potential the total population)51 and 21 million in Western Asia (10 percent evapotranspiration from warmer temperatures and non-linearity of the total population)52 (FAO 2012a, 2013). Besides sub-Saharan in the hydrological system (e.g. as precipitation is transformed into river runoff or groundwater recharge). 51 With highest shares in Morocco (5.5 percent). The region includes Algeria, Egypt, This analysis distinguishes between green water (plant-available Libya, Morocco, and Tunisia. 52 water in soils, directly resulting from precipitation) and blue With highest shares in Iraq (26 percent) and Yemen (32 percent). The region includes Iraq, Jordan, Kuwait, Lebanon, Saudi Arabia, Syria, Turkey, United Arab water (water in rivers and lakes, groundwater, and other water Emirates, and Yemen. bodies), because these two types of water have different uses and 131 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 4.16: Relative change in annual water discharge in the Middle East and North Africa region in a 4°C world. Relative to the 1986–2006 period based on an ISI-MIP model intercomparison using climate projections by CMIP5 GCMs as an input for global hydrology models (GHMs)21 (based on Schewe et al. 2013). Colors indicate the multi-model mean change, whereas the saturation indicates the agreement in sign of change over the ensemble of GCM—GHM combinations. opportunity costs and respond differently to climate change. Blue In a Lebanon-based study, Shaban (2008) found that over the water is used for irrigation and direct human consumption; green past four decades there has been a significant decrease (in the order water supports rain-fed agriculture as well as other land-based of 25 percent) in those water resources, including most rivers and ecosystems. Reductions in green water availability closely follow groundwater reservoirs, that are subject to climate change and reductions in precipitation. Reductions in blue water availability other direct human pressures. The decrease was somewhat smaller are generally greater than the underlying reduction in precipitation. (about 15 percent) for those water resources only subject to climatic As a result of global warming, especially changes in precipi- trends (e.g., snow cover and precipitation). Lebanon’s total water tation patterns, water availability will decrease in most parts of resources are projected to decrease by 6–8 percent for an increase the MENA region throughout the 21st century (García-Ruiz et al. in average annual temperature of 1°C, and by 12–16 percent for 2011) with decreases possibly exceeding 15 percent in a 2°C world an increase of 2°C, using the HadCm3 and PRECIS global and and 45 percent in a 4°C world in parts of the region (Schewe et regional climate models (Republic of Lebanon 2011). al. 2013). The exception is the southernmost areas Figure 4.16 shows changes in discharge (i.e., blue water availability) in the Surface Water 21st century as a result of a multi-model inter-comparison with In a study of the tributaries of the Jordan River, Samuels et al. (2010) five global climate models and nine global hydrological models projected a reduction in mean daily runoff of 17 percent for the under the RCP8.5 scenario (Schewe et al. 2013). period 2036–2060 (relative to 1980–2004), using the A1B scenario The trend toward decreasing average water availability will and the ECHAM 5 global/RegCM regional climate model and the be compounded by higher variability and more extremes, such as HYMKE watershed model. For the Zarqa basin in Jordan, Abdulla et droughts and flooding, leading to a loss of reliability and increas- al. (2008) projected changes in surface runoff ranging from –23.6 to ing uncertainty in water management (García-Ruiz et al. 2011; +36.6 percent, and changes in groundwater recharge from –57.5 to Törnros and Menzel 2014). Examples of previous severe flooding +89.8 percent, in response to stylized combinations of temperature events include those in Algiers in 2001, Morocco in 2002, and increases of 0–3.5°C and precipitation changes of between –20 and Tunis in 2003. +20 percent, using the BASINS-HSPF model. It is important to note While decreasing annual average precipitation in combination that the positive changes for precipitation (and subsequently runoff with higher temperatures is likely to cause a reduction in surface and recharge) are very hypothetical, given the strong agreement of runoff and groundwater recharge (aggravating current trends global climate models projecting a decrease in precipitation. in groundwater overexploitation and depletion), the effects of For the eastern Anatolian mountains (the headwaters of the increasing extremes (with higher rainfall per event) are not so Euphrates and Tigris rivers), snow water storage is expected to straightforward. Liu (2011) found that in some arid regions the decrease (see Box 4.1: Snow Water Storage) and, accordingly, a higher intensity of extreme precipitation events may increase the runoff decrease of 25–55 percent is projected between 1961–1990 fraction of precipitation that enters the soil and becomes plant- and 2071–2099, using different SRES scenarios (A1FI, A2 and B1), available soil moisture and contributes to groundwater recharge. different GCMs (ECHAM5, CCSM3 and HadCM3), and the RegCM3 In other regions, the fraction of runoff and loss of water to unpro- regional climate model (Bozkurt and Sen 2013). ductive evaporation, in contrast to productive plant transpiration, For Morocco, the 2nd National Communications (Kingdom may increase. of Morocco 2010) projected a decrease in river discharge of 132 MI DDLE E AST A ND NOR TH A FRICA For Saudi Arabia, increases in reference evapotranspiration are Box 4.1: Snow Water Storage expected to reduce groundwater recharge by 2–12 percent of total annual recharge in 2070–2100 relative to 1960–1990 using the regional Mountainous areas notably in Morocco, Algeria, Lebanon, Syria, climate model PRECIS and SRES A2 (Kingdom of Saudi Arabia 2011). Iraq, the Islamic Republic of Iran, and Turkey, play an important role in the water supply of the MENA region. Under climate change, 4.4.1.3 Climate Risks to the Agricultural Sector however, mountainous areas are expected not only to experience a and Food Security reduction in total precipitation but also a reduction in the fraction of The potential impacts of climate change on agriculture are vari- precipitation falling as snow, thereby affecting snow cover and snow ous. In water-limited regions, the length of the growing period water storage. is generally reduced if precipitation decreases while temperature Changes in melt-water regimes are expected to result in a and evaporative demand simultaneously increase. Higher rainfall shift in peak river flows toward the earlier months of the year, with variability with more intense and more frequent droughts, and negative impacts for downstream riparian systems in terms of water possibly also floods, can cause more frequent crop failures. With availability and seasonal shortages during the hot and dry summer months. In the 2nd National Communications of the Republic of increasing temperatures, several crops will exceed their tempera- Lebanon (2011), the upper Nahr el Kalb basin was analyzed under a ture tolerance levels. All of these effects contribute to an overall stylized warming of 2°C and 4°C. Snowpack volume was projected reduction in crop yields. to shrink from a total of 1,200 million m3 to 700 million m3 under 4°C Increasing atmospheric CO2 concentration, on the other hand, warming and 350 million m3 under 2°C warming. Drought periods can decrease crop water demand and increase crop productivity. It would thereby be expected to occur 15–20 days earlier (for 2°C is important to note that the following studies do not account for warming) and more than a month earlier (for 4°C warming), accord- adaptation, such as crop breeding, improved agricultural manage- ing to the report. For the Euphrates and Tigris basins, snow water ment practices, and additional irrigation. Adaptation measures have equivalent (the amount of water stored in the highland snowpack) great potential given the very low level of agricultural productivity has been projected to decrease by 55 percent in the B1 scenario across the MENA region. and 87 percent in the A1F1 scenario by 2071–2099, relative to 1961–1990 using the CCSM3 global climate model (Bozkurt and Length of the Growing Period and Crop Yields Sen 2013). Lower rainfalls and higher temperatures are expected to shorten the growing season for wheat in the MENA region by about two 15–55 percent by 2080 (relative to 1961–1990) for different sce- weeks by mid-century (Ferrise et al. 2013), as projected under the narios and river basins, with an average decrease under the B2 A1B scenario to 2031–2050, using the AORCM model. For Tunisia, scenario of 21 percent, and a decrease of 34 percent under the A2 Mougou et al. (2010) projected a shortening of the wheat growing scenario, using the WatBal model. period by 10 days for 1.3°C, by 16 days for 2°C, by 20 days for 2.5°C and by 30 days for 4°C warming, and a reduction in yields Ground Water of 10 percent for a 10 percent decrease in precipitation, 30 percent Groundwater in many water-scarce regions serves as a buffer for a temperature increase of 1.5°C, and 50 percent for a combina- for variable surface water availability. The buffering function of tion of both scenarios, using the DSSAT crop model. groundwater can only be maintained in the long term if there is a Most agricultural activities in the MENA region take place balance between recharge and withdrawal. In most MENA coun- in the semi-arid climate zone, either close to the coast or in the tries, however, groundwater is severely over-extracted (beyond highlands (Figure 4.13), where rainfall and green water availability the recharge rate). With falling groundwater levels, extraction are predicted to decline most strongly (see Section 4.3.3). The becomes ever more energy intensive and expensive, eventually resulting increases in irrigation water demand will be difficult forcing the abandonment of wells. Climate change is expected to to meet due to a simultaneously increasing blue water scarcity. accelerate the loss of groundwater buffer capacity as decreasing Due to increasing water scarcity, in combination with higher tem- rainfall decreases groundwater recharge in a non-linear way. peratures which are expected to deviate more and more from the For the West Bank and Gaza, Mizyed (2008) projected a decrease temperature optima of several crops (and possibly even exceed in recharge between 7.1 and 50.9 percent for stylized warming their heat tolerance levels), agricultural productivity is expected scenarios from 2°C to 6°C in combination, with a decrease in to drop in the MENA region. precipitation of 0–16 percent, using a GIS-based spatial analysis. In a study by Drine (2011), North African crop yields were pro- For Algeria, the 2nd National Communication report (Repub- jected to decrease by 0.8–12.8 percent under conditions of 20 percent lique Algerienne Democratique et Populaire 2010) projected a less precipitation, or by 1.6–26.6 percent under conditions of 2°C decrease in groundwater resources of 10–15 percent by 2050 relative warming, using a logarithmic regression model. For Algeria and to 1961–1990 under an IS92a scenario and using the GR2M model. Morocco, yield reductions of 36 percent and 39 percent (26 percent 133 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL and 30 percent including the CO2 fertilization) respectively are Table 4.6: Summary of crop yield responses to climate projected for 2080 (relative to 2003) for a B2 scenario (Schilling change, adaptation measures, and CO2 fertilization. et al. 2012). For the southwestern Mediterranean (Tunisia, Alge- ria, and Morocco), Giannakopoulos et al. (2009) projected yield SLOPE r2 t-STAT p-VALUE reductions for a variety of crops of between 1.5 and 23.9 percent; Full dataset –0.08 0.16 –5.5 <0.001*** for the southeastern Mediterranean (Jordan, Egypt, and Libya), Full dataset (below 2°C) –0.06 0.028 –1.86 0.065 yield changes of between +3.7 and –30.1 percent were projected CO2 fertilization 0 0 –0.04 0.96 for the period 2030–2060 relative to 1961–1990. This study used Adaptation measures 0.11 0.33 1.22 0.3 the A2 and B2 scenarios, the HadCM3 climate model, and the CROPSYST crop model. Without adaptation –0.08 0.16 –3.54 0.007** measures or the effects of In a study of Jordan’s Yarmuk basin, Al-Bakri et al. (2011) CO2 fertilization projected changes in wheat yields by mid-century ranging from –0.5 to +1.5 tons per hectare and changes in barley yields ranging Results of a general linear model applied to all studies with reported values for changes in yield and changes in temperature, to studies considering the from about –0.05 to –0.5 tons per hectare, starting from current effect of CO2 fertilization, to studies not considering the effect of CO2 fertil- values of about 1 ton per hectare. These ranges were calculated for ization, and to studies not considering the effects of adaptation measures. different stylized temperature (+1°C to +4°C) and precipitation Significance levels: *P<0.05, **P<0.01, ***P<0.001. (–20 to +20 percent) scenarios, using the DSSAT crop model. The scenarios that include precipitation increases are very hypothetical, given the strong agreement of global climate models regarding a the influence of adaptation measures and CO2 fertilization on crop projected decrease in precipitation. yields could only be assessed below a temperature increase of 2°C. For Syria, a study by Verner and Breisinger (2013) used the Overall, there exists a significant correlation between crop DSSAT crop model to project reductions in rain-fed wheat yields yield decreases and temperature increases (see Figure 4.17 and of between 23 and 57 percent for 2041–60, relative to 1991–2010, Table 4.6) regardless of crop type or whether the effects of CO2 under the A1B scenario. fertilization or adaptation measures are taken into account. If Egypt’s 2nd National Communications report (EEAA 2010) only studies with temperature increases below 2°C are included, projected a reduction in productivity under climate change of then the correlation is no longer significant (see Table 4.6). This 11–20 percent by 2050 for wheat, rice, maize, and barley, and an suggests that below a 2°C threshold the effects of adaptation increase in irrigation water demand by 2050 of about five percent measures and CO2 fertilization may compensate for the adverse across all SRES scenarios. effects of climate change. For Tunisia, a study by Lhomme et al. (2009) found increases If the effects of CO2 fertilization are considered, no significant as well as decreases in yields for 2071–2100 relative to 1961–1990, relationship between crop yield change and temperature increase is depending on the location within the country and the sowing rules revealed (Figure 4.17 and Table 4.6). The beneficial effects of CO2 applied, using the ARPEGE climate model for the A1B scenario. fertilization are highly uncertain and there is considerable doubt that the full benefits can be obtained (Ainsworth et al. 2008) (see Box 2.4). The relationship between the change in temperature Meta-Analysis of the Impacts of Climate Change on Crop Yields and crop yield response, in a scenario under which the effects of A meta-analysis of the impacts of climate change on crop yields for adaptation measures are taken into account, is positive but not the MENA region was conducted based on one single dataset that significant (Figure 4.18 and Table 4.6). consists of data from 16 different studies analyzed (see Section 6.3, Previous meta-analyses (Easterling et al. 2007; World Bank 2013) Meta-analysis of Crop Yield Changes with Climate Change, for details showed that above this temperature threshold the negative effects on data processing and methodology). The aim of the meta-analysis of climate change on crop yields are dominant, with crop yields was to summarize the range of projected crop yield changes in the considerably declining regardless of the positive effects of adaptation literature and to assess consensus for the MENA region. The dataset measures and CO2 fertilization. This could not be tested here due was used to address three main questions: (1) what are the likely to lack of data. The relationship between the change in crop yield impacts of incremental degrees of warming on yields?; (2) what and temperature increase is significantly negative in scenarios that is the impact when also considering adaptation measures and the do not consider adaptation measures or CO2 fertilization. effects of CO2 fertilization (see Box 2.4: The CO2 Fertilization Effect) Cropland on change in crop yields?; and (3) to what extent can adaptation According to Evans (2008), a warmer and drier climate is projected measures and/or the effects of CO2 fertilization counteract the nega- to shift vegetation and agricultural zones northward (e.g. by 75 km) tive effects of increased temperature? Due to the lack of data, both for 2090–2099 relative to 2000–2009 for the A2 scenario at the 0.9 134 MI DDLE E AST A ND NOR TH A FRICA Figure 4.17: Meta-analysis of the impact of temperature increase on crop yields. Best-fit line over the full dataset for the MENA region (blue line) and 95 percent confidence interval of regressions consistent with the data based on 500 bootstrap samples (blue shade). The influence of temperature increase on crop yield change is significant. Figure 4.18: Meta-analysis of the impact of temperature increases on crop yields excluding adaptation and CO2 fertilization. Best-fit lines for MENA studies considering neither the effect of adaptation measures nor of CO2 fertilization (blue line) and 95 percent confidence intervals of regressions consistent with the data based on 500 bootstrap samples (blue shade), for studies considering the effects of adaptation measures (green line) and studies considering the effects of CO2 fertilization (orange line). The solid line depicts a significant relationship and dotted lines non-significant relationships. 135 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL significance level. In Lebanon, shifts in agro-climatological zones the region. There is a close relationship between climate risks may push special crops, including cherry and olive plantations, to for the water sector and the agricultural sector, with feedbacks higher elevations (Republic of Lebanon 2011). Ferrise et al. (2013) from agricultural water use (e.g., increasing water demand, more also projected a shift northward for olive plantations. irrigation, and pollution of rivers) potentially compromising other Large fractions of currently marginal rain-fed cropland are water uses (the water-agriculture nexus). expected to be abandoned or transformed into grazing land; cur- The MENA region is very likely to experience a decrease in crop rent grazing land, meanwhile, may become unsuitable for any productivity unless effective adaptation measures are taken and agricultural activity (Evans 2008). Increasing aridity reduces the global CO2 emissions are reduced. The impacts of climate change available soil moisture, exacerbating the effects of ongoing land on agricultural production and food security, which are expected degradation. Evans (2008) projected the area viable for rain-fed to grow throughout the 21st century, will be aggravated by water agriculture in the Middle East to decrease by more than 8,500 km² and land resource degradation and also by indirect climate change by mid-century, and by more than 170,000 km2 by late-century (at effects (occurring in major food exporting regions) through higher the 0.9 significance level), when using the 200 mm isohyet as the food prices on world markets. If groundwater buffers fail due to limit for rain-fed agriculture, below which only seasonal grazing prolonged droughts leading to long-term decreases in recharge, is practiced. The study used the A2 scenario and an ensemble of as recently observed in northeastern Syria, agricultural land use 18 GCMs which project a temperature increase of 1.4°C in the first and livelihoods will have to be abandoned. half, and an additional 2.5°C warming in the second half of the MENA countries have at their disposal a wide range of adap- century, as well as a decrease in precipitation of 8.4 percent in the tation options for reducing their overall vulnerability. Demand first half and another 17 percent in the second half of the century. management and increased resource use efficiencies can provide many opportunities for adapting to increased water scarcity. Diver- Livestock sification of the MENA countries’ economies also holds a large Livestock is an important component of rural livelihoods and food potential for increasing resilience to climate and other risks. The security in the MENA region. Climate change will impact livestock MENA countries are blessed with rich renewable energy resources, production through various pathways, including changes in the in particular solar and wind, providing them with climate change quantity and quality of available feeds and the length of grazing adaptation and mitigation options (and combinations of them, season, additional heat stress, loss of drinking water availability, for example, via solar desalination). Knowledge and technology and changes in livestock diseases and disease vectors (Thornton transfer and investments across the region from more to less et al. 2009). These climate-related pressures may be amplified advanced countries can help to overcome the resource predica- by, or may amplify the effects of other pressures (e.g. rangeland ment and contribute to prosperity and stability. degradation, fragmentation of grazing land, changes in land tenure, While biophysical impacts vary only slightly across the region, and lack of market access). vulnerability and socioeconomic impacts show a clear division The vulnerability of livestock production systems to droughts between the (oil) rich Arab Gulf countries and the rest of the region. was recently on display in northeastern Syria, where herders While the former have the financial means for adaptation (e.g., lost almost 85 percent of their livestock as a result of recurring desalination technology, increasing food imports), the latter are droughts from 2005–2010 (Selvaraju 2013). The precarious situa- much more vulnerable to climate risks. Vulnerability also varies a tion of the rural population in this region caused large waves of lot within countries, with poor and rural populations at particular migration to the cities. risk (e.g., due to their dependency on agriculture). 4.4.1.4 Synthesis 4.4.2 Desertification, Salinization, The MENA region, also known as the “cradle of agriculture” or and Dust Storms “cradle of civilization,” is one of the world’s most vulnerable regions to climate change in combination with other pressures. 4.4.2.1 Desertification From the current situation of critical water and land scarcity, both Large parts of the MNA region are covered with drylands, either the 2°C and 4°C warming scenarios will result in further increased dry-subhumid, semi-arid, arid or hyper-arid. These harsh envi- pressures on water resources and agriculture, so that by the end ronments support, in addition to their limited water resources, of the century climate risks are projected to be much more severe. fragile ecosystems. The concept of desertification is vague, and Other pressures, including increasing demand for food and water many definitions exist to describe the phenomenon (Verstraete and further resource degradation, if following current trends, will et al. 2009). Most reports and studies on this topic refer to the multiply the impacts of climate change. UN Convention to Combat Desertification (UNCCD 1994), which Impacts on water resources will have severe consequences defines desertification as land degradation affecting drylands for the agricultural sector, which is the largest user of water in (Adeel et al. 2005; Boko et al. 2007; Reynolds and Stafford Smith 136 MI DDLE E AST A ND NOR TH A FRICA Figure 4.19: The far-reaching impacts and downward spiral of established over generations, is destabilized. A new equilibrium desertification. state is established in the degraded land in a process that is all but irreversible (Safriel and Adeel 2005; Seely et al. 2006). EXTREME POVERTY FOOD INSECURITY INCREASED RISK Desertification and Climate Change AND HUNGER OF DROUGHT AND Degradation of the WATER STRESS The role of climate change in desertification can be multifaceted, resource base of Degradation of the and it varies depending on local conditions and the interactions the poor resource base of Resilience impeded the poor among drivers. An increase in temperatures and evapotranspiration rates, a change in the precipitation regime, and an intensification or change in frequencies of extreme events can all directly trigger Feed me to feed you or enhance desertification processes (Aguirre-Salado et al. 2012; INCREASED H BIODIVERSITY LOSS witAND Belaroui et al. 2013; Scheiter and Higgins 2009; Schilling et al. 2012; h c LE EMISSION OF are CARBON Species extinction Verstraete et al. 2009). Furthermore, these climatic alterations can and loss of soil biodiversity indirectly push the dryland ecosystem towards desertification—for Reduced adaptive DLDD & Linkages to capacity other Global Issues example, via a shift in biomes, an increase in bush fires, or the lowering of groundwater tables (Geist and Lambin 2004). No study has attributed recent climate change as the single DEFORESTATION INSTABILITY ENVIRONMENTAL driver of desertification. As a result, it is difficult to quantify the AND AND CRISES INDUCED role of climate change relative to other drivers (Evans and Geerken ECOSYSTEM LOSS MIGRATION Threats to peace Changing migration 2004; Herrmann and Hutchinson 2005; Wessels et al. 2007). It is Degradation of and security patterns due to clear, however, that climate change can modify the natural con- agricultural land is the greater competition main driver of ditions of biomes in many ways and that it generally makes the over natural resources deforestation ecosystems more vulnerable to desertification processes (Evans and Geerken 2004; Verstraete et al. 2009; Xu et al. 2011). Source: (UNCCD Secretariat 2013). The desertification process can also affect the climate. At the regional scale, the decline of plant cover reduces evapotranspira- tion and increases mean temperatures. This leads to a decrease in 2002; Reynolds et al. 2007; Safriel 2009). This process includes a humidity within the regional climate system, which in turn may change in soil properties, vegetation, or/and climate (D’Odorico reduce the mean annual precipitation and/or cause greater vari- et al. 2013). ability in precipitation patterns (Abiodun et al. 2007; D’Odorico Desertification transforms a dryland ecosystem into a new and Bhattachan 2012; Foley et al. 2003; Mahmood et al. 2013). ecosystem with a lower level of service provision (Safriel and At the global scale, desertification is a driver of climate change Adeel 2005). This causes various stresses. Loss of vegetation through the loss of carbon sequestration capacity in dryland eco- cover, soil erosion, dust storms, salinization, and a decrease in systems and an increase in land-surface albedo (Adeel et al. 2005; soil productivity are some common threats in a downward spiral Aguirre-Salado et al. 2012). Recent studies show, however, that the that lead to a decrease in agricultural yields, loss of biodiversity, effect of CO2 fertilization on drylands (see Box 4.2) can potentially poverty, reduced human wellbeing, and migration (Bayram and reverse the feedback loop and transform desert into grassland. Öztürk 2014; Boko et al. 2007; D’Odorico et al. 2013; Safriel and There are attribution studies on desertification or national Adeel 2005, 2008) (see Figure 4.19). reports explaining the causes of desertification for all the MENA The processes and drivers which trigger and control this eco- countries. All of these studies attribute desertification at least in system change are often labeled as either the “desert paradigm” part to malpractices in land management. or the “desert syndrome.” The Millennium Ecosystem Assessment Some studies identify climate change as an additional driver of (2005) described this phenomenon as the “long-term failure to desertification. In most cases, the observed driver was a change in balance demand for and supply of ecosystem services in drylands.” rainfall pattern, with either a negative trend or higher variability. Population growth, poverty, marginalization, and environmen- For example, Conca et al. (2010) analyzed changes in humidity tal degradation are coupled in socioecological feedback loops in the United Arab Emirates, and concluded that a reduction in that drive the system into a downward spiral of desertification humidity linked to climate change during the last decade could (D’Odorico et al. 2013; Easdale and Domptail 2014; Reynolds and be responsible for an increase in desertification in the hyper-arid Stafford Smith 2002; Safriel and Adeel 2005; Stafford Smith 2008; areas. For many of the studies, however, it is not clear whether Verstraete et al. 2009). Finally, the fragile balance between the the increase in aridity is within the normal climatic variation or dryland ecosystem and the people living in it, which has been connected to recent climate change. 137 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Scheiter and Higgins (2009) projected a reduction in desert Box 4.2: The CO2 Fertilization Effect ecosystem area over Africa because of the CO2 fertilization effect. on Drylands They used the adaptive dynamic global vegetation model aDGVM, which can also account for the influence of bush fires on vegeta- Recent studies show a major influence of the CO2 fertilization effect tion. The model was driven using projected atmospheric CO2 on vegetation cover/growth in dryland ecosystems. Field studies concentrations under the SRES scenario A1B, and temperature found that the CO2 effect already influences African dryland eco- and rainfall projections were derived from the climate model systems, notably savannahs (Donohue et al. 2013; Kgope et al. ECHAM5. At the baseline year of 2008, 32.8 percent of Africa´s 2009). The increasing atmospheric CO2 concentration affects the land surface was covered with deserts. By 2100, desert cover is photosynthetic pathway of C3 plants (trees) and C4 plants (grasses) projected to decline to 27.1 percent including bush fires, and to by improving the plants’ water use efficiency. This forces transitions 26.9 percent without taking bush fires into account. The authors in ecosystems characterized by higher biomass and/or woody-plant explained that the projected reduction in desert ecosystem area is dominance. For grassland ecosystems, like savannahs, the CO2 fertilization due to the CO2 fertilization effect and an increase in the potential effect leads to an increase in woody species which profit more from growing season for grasses and trees under elevated temperatures. the additional CO2 in the atmosphere than C4 plants (Bond and In a similar study of Africa, Higgins and Scheiter (2012) once Midgley 2012). However, C4 plants also profit from the CO2 fertiliza- again used the aDGVM and ECHAM5 models under the SRES tion effect, and Higgins and Scheiter (2012) project a substantial A1B scenario. They projected a reduction in desert area from re-greening of deserts on the African continent even under decreas- 28 percent in 1850 to 23 percent in 2100, whereby desert biomes ing precipitation. shift to grassland mainly due to the positive effects of increased CO2 concentrations on plant growth and biomass. 4.4.2.2 Salinization While saline and sodic soils can be found naturally in many parts Projected Desertification of the MENA region, this report refers to the soil degradation Although human influences related to land management have processes of salinization and sodification. Salinization is a form been identified as direct drivers of desertification, climate change of soil degradation whereby water-soluble salts accumulate in is expected to alter the underlying natural conditions under which the soil (Jones et al. 2013; JRC 2009). A special form of salinity is desertification may occur. Projections need to incorporate the natural sodicity, whereby sodium is accumulated in the soil (D’Odorico and human aspects of desertification (Verstraete et al. 2009). To et al. 2013). Salinization is typically connected to the desertifica- date, however, there is no regional study of future land degrada- tion process. High levels of salinity in soils affect plant growth tion that assesses both climate change and social dynamics as through an increase in osmotic pressure (making it more difficult drivers of desertification. As land degradation of drylands is likely for plants to draw water from the soil) as well as through the to occur under certain climate change conditions, projections of toxic effects of salts. These processes make salinization a threat these climate parameters and their potential impact on ecosystems to both agriculture and natural ecosystems (Sowers et al. 2011; may be used as indicators to assess impending desertification. Vengosh 2014). Gao and Giorgi (2008) projected the climate in the Mediter- The most common causes of salinization are non-groundwater- ranean region (including North Africa and the Western Mashrek) associated salinity (whereby dissolved salts from rocks are for the period 2071–2100 with the regional climate model RegCM, introduced to the soil via rain), groundwater-associated salinity driven by the global model HadAM3H under the SRES scenarios (groundwater infusing the salt into the soils), and irrigation- A2 and B2. From the results, they derived three different measures associated salinity (D’Odorico et al. 2013). Climatic conditions of aridity from precipitation and temperatures. As a first measure in particular influence the latter two mechanisms. A decrease they used the changes in climatic conditions according to the in precipitation and an increase in temperatures lead to a higher Köppen-Geiger classification under both emissions scenarios demand for irrigation, which causes salinization when drainage is (Köppen 1936) and found that subtropical summer-dry climates not adequate. The intensification of irrigation may also lead to an turn into dry arid or dry semi-arid climates. The Budyko-dryness increase in freshwater withdrawal from the groundwater, trigger- index (Budyko 1958) and the UNEP dryness index (UNEP 1992) ing salinization of aquifers and also aggravating soil salinization support this change and they show a strong increase of aridity in (D’Odorico et al. 2013; Vengosh 2014). the whole region. The authors concluded that North Africa and The increase in salinization with further intrusion due to sea- western Mashrek are especially at risk of both increased water level rise under climate change holds not only for groundwater stress on natural ecosystems and desertification. but also for other water resources in the region. (Sowers et al. 138 MI DDLE E AST A ND NOR TH A FRICA 2011). River runoff and aquifer replenishment rates typically and organic matter (Akbari 2011; Elasha 2010; Notaro et al. 2013). decrease under drier conditions, causing an increase in the salt- In addition to economic losses, there are severe health impacts concentration of the remaining water. The higher salinity in the for people from breathing dust particles and from the airborne river also affects other water resources (e.g., lakes and groundwater microorganisms that are transported in dust clouds (Goudie 2009; aquifers), leading to a general increase in salinity, including soil Griffin 2007; Kanatani and Ito 2010; de Longueville et al. 2013). salinity (Vengosh 2014). In the MENA region, river salinization Fine atmospheric dust, as an aerosol, can also affect the local can be observed in the Euphrates and the Tigris rivers (Odemis climate, as it influences cloud formation and precipitation even et al. 2010), in the Jordan River (Farber et al. 2004), and in the in remote regions (Creamean et al. 2013; Jung et al. 2013). Dust Nile (Elewa and El-Nahry 2008; El-Nahry and Doluschitz 2010). storms not only influence local climate, but the frequency and In coastal areas, the intensive extraction of groundwater leads magnitude of dust storms is also affected by climate change and to seawater intrusion into the aquifers, causing severe salinization. desertification in different ways. Decreasing vegetation cover not This process is accelerated by climate-change-induced sea-level only increases the sources of dust, due to an increase in bare rise (Carneiro et al. 2009; Niang et al. 2010). The Nile Delta (see soils (Bayram and Öztürk 2014; Mulitza et al. 2010; Pierre et al. Box 4.3), an area that is home to more than 35 million people 2012), but may also affect wind speeds by decreasing the surface and that provides 63 percent of the agricultural production of roughness (Cowie et al. 2013; Pierre et al. 2012). Egypt, is especially vulnerable to salinization under changing There are no studies linking projected climate change and climate conditions (Bohannon 2010; El-Nahry and Doluschitz 2010; changes in dust storm occurrence in the MENA region (Goudie Hereher 2010). According to Mabrouk et al. (2013), salinization is 2009). The frequency and strength of dust storms depend on both very likely to rapidly worsen under climate change; the authors the availability of dust (and therefore regional vegetation cover) call for an integrated three-dimensional groundwater modeling and wind patterns. The vegetation coverage is directly linked to of the Nile Delta in order to fully understand the implications of desertification, which is a highly uncertain process under future sea-level rise and salinization. climate conditions and depends on both natural and human fac- tors. Wind is a climatic factor and changes in wind patterns can 4.4.2.3 Dust Storms be projected from climate models. However, there are currently no Dust storms are a typical phenomenon in arid regions without full regional studies on changing wind patterns under climate change vegetation cover (Squires 2002). Covered mostly by drylands, the in the MENA region; as a result, future trends have to be derived MENA region is frequently threatened by dust storms that cause from global studies. widespread damage to people, to agriculture, and to the overall economy (Akbari 2011; Kumar 2013; UN 2013; UNCCD Secretariat 4.4.2.4 Synthesis 2013; Verner 2012). Desertification, salinization, and dust storms are closely related The extent of dust storms varies widely, depending on the in the MENA region. In an arid environment in which irrigation meteorological conditions, the land surface, and the size of the accelerates the salinization of soils and results in a degraded desert particles transported. In the MENA region, dust storms reach large environment, dust storms carry away the soils, putting pressure scales in terms of mass lifting, spatial extent (i.e., 100–1000 km), on agriculture and further driving the desertification process. and duration (i.e., from several hours to several days) (Ghoneim This narrative is referred to in the literature as the desertification 2009; Kocha et al. 2012; Miller et al. 2008; Pey et al. 2013). The paradigm; it is oversimplified, but the core message is clear—none wall of dust and sand can reach concentrations of 6000 μg particles of these processes stands alone. In many aspects, salinization is per m3 (Goudie 2009), causing traffic accidents, disruptions of an important driver of desertification (JRC 2009), but land use flight traffic, destruction of telecommunications and mechanical and land cover changes (e.g., land clearing and replacing natural systems, and damage to crops. In addition, dust storms influence vegetation with annual crops) can also reinforce or even trigger the performance of solar photovoltaic power plants, directly by the salinization process (Vengosh 2014). depositing dust on the panels and indirectly by reducing radiation. In general, desertification, salinization, and dust storms are well In Egypt, for example, Elminir et al. (2006) found a 12–52 percent understood and described in the literature. There are still large gaps, reduction in transmittance, depending on the amount of dust and however, when it comes to the multifaceted role of climate change tilt angle, and a 17 percent per month reduction in output power. (Verstraete et al. 2008, 2009) and the role of CO2 fertilization (Donohue Mani and Pillai (2010) reviewed several impact studies and reported et al. 2013; Higgins and Scheiter 2012). This may be one reason for reductions in photovoltaic performance by 17–32 percent in Saudi the lack of quantitative projections on desertification that account for Arabia and Kuwait. biophysical (including climate change), socioeconomic, and anthro- Agricultural productivity is also affected in the long term by soil pogenic drivers (including population growth and land use change). loss, as dust storms in particular remove light, nutrient-rich particles Numerous studies on the observed impacts of desertification in the 139 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL MENA region do however demonstrate a strong need for integrated for at least one month of the year by the 2080s is 20 million under modeling approaches to assess desertification under global climate the B2 scenario, 34 million under B1, 62 million under A2, and change. As the MENA region is already prone to desertification, and 39 million under the A1F1 scenario (van Lieshout et al. 2004). with climate change bringing increasingly arid conditions, sensitivity Lymphatic filariasis, commonly known as elephantiasis, is a to desertification is likely to increase. A reduction in precipitation, disease of the lymphatic system caused by parasitic worms trans- an increase in extreme events (e.g., drought and flash floods), and mitted by mosquitoes. Like malaria and other mosquito-borne an increase in temperature and evapotranspiration will reinforce diseases, the prevalence of lymphatic filariasis could be affected desertification in the region (WMO 2007). by climate change (via the effects of changes in temperature, rainfall, and humidity on mosquito breeding and survival rates). 4.4.3 Human Health While it is currently present in North Africa only in Egypt’s Nile Delta, Slater and Michael (2012) project with a 75–100 percent People in the MENA region face a variety of health risks, many of probability that the range of the disease could extend to include which are exacerbated by the hot and arid conditions and relative much of coastal North Africa by 2050 under both the SRES B2 water scarcity that generally characterize the region. For example, and A2 scenarios. trachoma (an infectious disease affecting the eyelids) tends to Leishmaniasis, a skin disease carried by sandflies endemic to occur in dry areas with poor sanitation; it is endemic in Morocco, the MENA region, is also affected by climatic factors (Ben-Ahmed Algeria, Libya, Egypt, Iraq, the Islamic Republic of Iran, Oman, and et al. 2009; Toumi et al. 2012); it is also affected by environmental Yemen (Smith et al. 2013). The MENA region is also experiencing factors such as dam construction (Riyad et al. 2013). The disease, a resurgence of several vector-borne and viral diseases that had which occurs in several forms, can result in morbidity, disfigurement, previously been in decline (Lelieveld et al. 2012). Climate change and mortality (McDowell et al. 2011)—and is considered a major may compound the challenge of managing these diseases. public health problem in the region (Postigo 2010). Outbreaks are reportedly becoming more frequent in Tunisia, Algeria, and Morocco, 4.4.3.1 Vector-Borne Diseases where the range of the disease has expanded due to a variety of Malaria is rarely endemic in North Africa, and the measures in environmental (e.g., agricultural projects) and human factors (e.g., place to control the disease are considered effective. The disease migration of nonimmune populations) (Riyad et al. 2013). is also present in the Middle East, with cases reported in the A serious and potentially fatal form of the disease is visceral Islamic Republic of Iran, Iraq, and Saudi Arabia, but in most leishmaniasis. A study conducted in the northwest part of the countries endemicity is relatively low and localized. Malaria is, Islamic Republic of Iran reported a strong association between however, prevalent in Djibouti and Yemen (WHO 2013a); (WHO rising temperatures and growing populations of carrier sandflies and EMRO 2009). (Oshaghi et al. 2009). A study from Tunisia showed that large While the relationship between climate change and malaria numbers of Mediterranean visceral leishmaniasis (MVL) cases remains somewhat in dispute, evidence indicates that changes tend to be preceded two years earlier by a particularly rainy season in climatic factors can affect the incidence of the disease in the (Ben-Ahmed et al. 2009). Increased rates of MVL transmission in MENA region. Malaria distribution and seasonal occurrence have the canine reservoir following the rainy season could explain the been linked to temperature, elevation, humidity, and low rainfall delayed increase in the incidence of human cases. in the Islamic Republic of Iran (Salehi et al. 2008). Environmental Schistosomiasis, or bilharzia, is transmitted by snails and is changes indirectly associated with climate change could also favor found in North Africa, the Islamic Republic of Iran, Iraq, Saudi the disease vectors of malaria such as the Anopheles sergentii Arabia, and Yemen. Schistosomiasis is expected to be affected by mosquito. This mosquito is known as the “oasis vector” due to temperature changes, although in a non-linear way (Mangal et al. its prevalence in oases across the Sahara and its ability to cope 2008). Mangal et al (2008) found that the burden of infection within with extreme climatic conditions. Lotfy (2013) suggested that infected people increases by more than tenfold as mean ambient the underground water reservoirs in the Western Desert being temperatures rise to 30°C; mean temperatures above 30°C could constructed by the Egyptian government to mitigate water stress cause the mortality of snail hosts, however, potentially limiting could aid the emergence of Anopheles sergentii in new areas. In further transmission. This means that increased temperatures could the World Health Organization Eastern Mediterranean region B result in increased morbidity and mortality for infected people (EMR-B),53 the additional population exposed to risk of malaria rather than an increased prevalence of infection. 53 4.4.3.2 Food and Water-Borne Diseases This region includes Bahrain, the Islamic Republic of Iran, Jordan, Kuwait, Lebanon, Libya, Oman, Qatar, Saudi Arabia, Syria, Tunisia, and the United Arab Emirates. Note The prevalence of food-borne diseases such as salmonella and that it excludes Djibouti, Egypt, Iraq, Morocco, and Yemen. Malaria endemicity in Escherichia coli, and water-borne diseases such as cholera, dys- Djibouti, Iraq, and Yemen is currently relatively high. entery, and typhoid fever, is expected to be affected by changing 140 MI DDLE E AST A ND NOR TH A FRICA temperature and rainfall patterns. Outbreaks of cholera, for increased hospital admissions in the Israeli cities of Tel Aviv and example, have followed seasonal patterns in the last three decades, Haifa (Novikov et al. 2012; Portnov et al. 2011). with the effect being stronger at latitudes further away from the Increased rates of heat stress under climate change can affect Equator (Emch et al. 2008). Cholera outbreaks correlate with high labor productivity. Kjellstrom et al. (2009) noted that, in densely temperatures and can follow extreme weather events that disrupt populated cities of the world that already experience very high water supplies (e.g., drought and flooding). In recent years, chol- maximum temperatures and are projected to experience the era has caused deaths in Iraq, the Islamic Republic of Iran, and greatest increases under climate change (including some densely Yemen (WHO 2013b). populated cites in the MENA region), heavy outdoor work (e.g., The incidence of diarrheal disease among children is high in in agriculture and construction) will become more challenging parts of the MENA region where warm weather, inadequate access and may take a heavier toll on the health of workers. to drinking water, poor sanitation, and poverty collide (Kolahi et According to the Climate Vulnerability Monitor produced al. 2010). Kolstad and Johansson (2011) projected increased rates by the Spanish nonprofit organization DARA, the proportion of of diarrheal diseases as a result of climatic changes in the A1B the workforce expected to be particularly affected between 2010 scenario. The relative risk of diarrheal disease (compared to a and 2030 by reduced productivity under the A2 scenario ranges 1961–1990 baseline) is expected to increase in North Africa by between 10–20 percent in North Africa, as well as in Israel, Jordan, 6–14 percent for the period 2010–39, and by 16–38 percent for the Lebanon, Syria, the Islamic Republic of Iran, and Iraq. For the period 2070–99, and by 6–15 percent in the Middle East for the countries of the Arabian Peninsula, the proportion is projected to period 2010–2039, and 17–41 percent for the period 2070–2099. be higher—ranging from 15–40 percent (DARA 2012). 4.4.3.3 Impacts of Extreme Heat Events 4.4.3.4 Synthesis The MENA region is already characterized by very high summer Several studies exist that investigate how various climatic changes temperatures, making the populations of the region highly susceptible affect transmission of vector-borne disease. These studies tend to to further temperature increases (Habib et al. 2010; Lelieveld et al. be location-specific, however, which might prevent generalization 2012). Giannakopoulos et al. (2013) assessed changes in levels of at a broader scale. Emch et al. (2008) show a relatively strong thermal discomfort as indicated by the humidex, an index used to observed correlation between cholera outbreaks, elevated tem- express the temperature perceived by people. The number of days peratures, and contaminated water supply; this appears to point characterized by high thermal discomfort (humidex > 38°C) in to a substantial risk to the MENA region under climate change. the base period of 1961–1990 is approximately 100 days in North A strong correlation is also apparent between extreme heat and Africa and the eastern and southern parts of the Arabian Penin- increased mortality rates, although further research on future sula. For 2040–2069 under the SRES A1B warming scenario, this patterns of heat-related illness, as well as on the indirect effects is projected to increase by approximately 35 days in North Africa of extreme heat, is needed. Likewise, there is evidence in the lit- and by 70 days in the Arabian Peninsula. erature to suggest that the region will face greater health burdens High temperatures can cause several medical conditions, associated with air pollution; this, however, refers to the effect of including heat stress, heat exhaustion, and heat stroke, with the anthropogenic emissions on atmospheric composition and does not elderly, young children, and people with existing medical condi- include the effect of climatic change on air pollution. As pointed tions most vulnerable to heat-related mortality. Novikov et al. out by Lieleveld et al. (2013), further work linking projections (2012) reported that the number of emergency hospital admissions of atmospheric composition to climate simulations is required. in Tel Aviv increases by 1.47 percent per 1°C increase in ambient temperature. Regression modeling has further found extreme high 4.4.4 Migration and Security temperatures to be linked to increased mortality rates in Tel Aviv (Leone et al. 2013; Peretz et al. 2012), Tunis (Leone et al. 2013), Mobility as an integral part of people’s livelihoods allows for and Beirut (El-Zein et al. 2004). Peretz et al. (2012) reported an diversification and the securing of income (Gemenne 2011). increase in mortality of 3.72 percent for every one unit increase in Migration, in the form of pastoral nomadism, has long been a the human thermal discomfort index (that involves temperature part of traditional lifestyles in MENA and the Sahelian adjacent and relative humidity as both additive and multiplicative factors) territories in the south (Brücker et al. 2012; Fritz 2010). Nomads above a discomfort threshold of 29.3. and their livestock have for thousands of years been cyclically There are also indications that extreme heat events may have migrating to places where they could find sufficient fodder and an indirect effect on health by worsening air pollution, which can water (Nijeri Njiru 2012). Moreover, migration has always been aggravate respiratory illnesses (Markandya and Chiabai 2009). a human response to climatic hazards. Migration, in the context Together with increased temperatures, elevated concentrations of of climate change, happens as soon as the physical, economic, sulfur dioxide and particulate matter have been associated with social, or political security of a population decreases and no other 141 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 4.20: Push factors as interrelated drivers for migration and determinants for decision making. PERSONAL/HOUSEHOLD SOCIAL DRIVERS CHARACTERISTICS SPHERE OF Education POLITICAL DRIVERS Age, sex, education, wealth, PUSH FACTORS Poverty Discrimination/persecution marital status, family size, Marginalization/social Marginalization/social number of dependents, exclusion ENVIRONMENTAL/PHYSICAL DRIVERS exclusion preferences, personality, Physical exposure to climate change- Political instability mobility, ethnicity, religion, induced hazards Weak governance and language, culture Indirect exposure to hazards institutions Destruction of belongings Corruption Loss of livelihood Interrelated and Conflict/insecurity Failure/change of ecosystem services overlapping Policy incentives MIGRATE Resource depletion drivers at Direct coercion location of origin DECISION ECONOMIC DRIVERS DEMOGRAPHIC DRIVERS STAY INFLUENCE OF Loss of livelihood Naturally growing population CLIMATIC AND Low income Population pressure from ENVIRONMENTAL Poverty immigration CHANGE ON Low producer prices Population structure INTERVENING OBSTACLES DRIVERS High consumer prices Disease prevalence AND FACILITATORS Political/legal framework, cost of migration, distance, social networks, diasporic links, recruitment agencies, technology Source: Adapted from Foresight (2011), p. 33. resources can be mobilized to adapt to the new conditions. Some The research challenges include a general lack of disaggre- scholars therefore consider migration as a last resort (Laczko gated data on drivers of migration and conflict as well as very and Piguet 2014; Warner et al. 2010, 2008), while others debate divergent projections for the future. Even though the occurrence whether migration should be considered a successful adaptation of migration and conflict spatially overlaps with vulnerability to strategy or a failure to adapt (Bardsley and Hugo 2010; Fritz 2010; climate change, this does not necessarily indicate a topical correla- Gemenne 2013; Luecke 2011; Tacoli 2009). tion. Moreover, environmental migrants, environmental refugees, The goal of a migratory movement is to reach a higher level of climate migrants, and climate refugees (as they are referred to in physical, economic, social, or political security, which can bring new the literature) are not officially recognized as refugees55 by the opportunities and resilience (Black, Bennett et al. 2011; Scheffran et al. United Nations High Commissioner for Refugees (UNHCR). This 2011). An improvement in living conditions is not always achieved, is a possible explanation for why there is little available data on however, due to new physical, economic, and social vulnerabilities climate-change-related migration. encountered en-route or at the point of destination (Warner et al. 2010). Conceptual studies on migration as a response to climate change 4.4.4.1 Climate Change and Migration and the relationship of climate and violent conflict are abundant. in the MENA Region Practical studies, especially those providing numbers, are however Due to the interrelation of several factors and their common con- scarce (Gemenne 2011). The EACH-FOR studies, a series of local tribution to decisions for and against migration, it is difficult to say qualitative household surveys investigating the motivation to whether migrants are driven by climate changes or are economically, migrate in Egypt, Morocco, and the Western Sahara (Afifi 2009; socially, or politically motivated (Brücker et al. 2012). Tacoli (2011) Gila et al. 2009; Hamza et al. 2009) appear to be among the only identified demography as a further dimension (see Figure 4.20). comprehensive studies on migration and climate change. Wodon It is also difficult to extrapolate future migration patterns from et al. (2013) also conducted household interviews in five Arab current patterns, because these may change with increasing tem- countries on migration, climate change, and related topics. Other perature levels and changing socioeconomic conditions, and are also studies are mostly based on documentary research54 and do not dependent on future population growth. Projecting the nature and provide original findings. magnitude of migration, especially at the local and regional levels, is also fraught with uncertainty. Gemenne (2011) predicted that 54 The use of sources and documents from the personal, private or public domain, 55 such as personal papers, commercial records, state archives, legislation or speeches, Conditions anchored in the 1951 Geneva Convention text exclusively comprise in order to categorize, investigate and interpret information relating to a topical field. persecution for reasons of race, religion, nationality, membership of a particular This technique of data collection is commonly used in social sciences. social group or political opinion (UNHCR 2010). 142 MI DDLE E AST A ND NOR TH A FRICA migration options will be limited in a warmer world, that internal For 2030, the annual water availability per capita is estimated to migration will prevail, and that traditional patterns of mobility (e.g., be only 468 m³, whereas in the 1990s it was 1000 m³ per capita nomadism, temporary or circular labor migration) might be disrupted. (Wodon et al. 2014). Other major migratory patterns that have Many people will be forced to move, but others will be forced to stay been identified in Egypt include migration from south to north, because they lack the financial resources or social networks facilitat- and migration from throughout the country to the Suez Canal ing mobility. This indicates that climate-induced migration should be zone (Wodon et al. 2014). addressed not only within the framework of climate change, but also There are several studies that examine the effect of sea-level within other economic, cultural, technological or political conditions rise in Egypt and the potential impact on migration. Kumetat that might foster or limit migration (Gemenne 2011). (2012) found that a one meter sea-level rise could drive up to six Migrants from the MENA countries move internally within their million migrants from the Nile delta region. The statement for this countries or across borders to neighboring countries. The internal estimate is limited, however, because a reference year is not given. migration dynamic within the MENA region is rather short-distance For Morocco, internal migration exceeds international migration; (Afifi 2009; Wodon et al. 2013). From past observations it is well similar to Egypt and Algeria, rural areas are being abandoned in known that migrants in Tunisia, Algeria, and Morocco who were left favor of cities. The principal destinations of Moroccan migrants homeless after sudden climatic hazards have a high propensity to are cities on the Atlantic coast, which grew rapidly between 2000 return to their homes after a disaster (Gubert and Nordman 2010). and 2010 (Wodon et al. 2014). Rural livelihoods in Morocco are Slow-onset hazards that may drive migration include increasing more prone to climate change, especially due to a high share of water scarcity, drought, desertification, and soil degradation (see employment in agriculture. This means that, in the future, the Section 4.4.2, Desertification, Salinization, and Dust Storms). urbanization trend will likely continue. Water stress is already Icduygu and Sert (2011) found that rural populations in particular a severe problem in the country, but it will become worse due migrate to nearby or larger cities once food self-sufficiency and to high population growth (Hamza et al. 2009); this will further livelihoods are threatened. Jónsson (2010), however, found that influence migratory movements. drought does not necessarily lead to migration. Water availability is also a serious problem in Syria, especially Contrary to the situation during the 1950s and 1960s, when labor for the rural population. The country experienced a drought in migration from the Maghreb to Europe was prevalent (Wodon et al. 2007–2008 which affected about 1.3 million Syrians (Wodon et al. 2014), today the MENA countries are a region of internal migration. 2014). As a result of the drought, 800,000 people, among them Urbanization is currently the predominant form of migration (Brücker small-scale farmers and herders, are believed to have lost their et al. 2012), and it adds additional pressure to urban infrastructure. livelihoods. As a consequence, about 40,000–60,000 households In Algeria, for example, migrants are shown to move from rural migrated to urban areas (DREF 2009), putting enormous pressures areas to mid-sized towns rather than to large urban areas (Gubert on local urban infrastructure (Wodon et al. 2014). and Nordman 2010). This movement is partly due to slow-onset Yemen has experienced strong internal migration flows to its environmental degradation, including water scarcity, soil erosion, fast-growing urban centers (Wodon et al. 2014). Water shortages and desertification which threatens agricultural livelihoods in rural as a result of decreasing rainfall and resource depletion are prob- Algeria (Wodon et al. 2014). However, unlike for many of its neigh- lematic in Yemen, and average per capita water availability is low. bors, climate change is not predicted to drastically increase migra- This has serious implications, including for an agricultural sector tion in Algeria (Wodon et al. 2014) since the country’s agricultural that contributes 15 percent to the country’s GDP and employs over sector makes up only a small fraction of GDP (and relatively few half of the population (Wodon et al. 2014; World Bank 2013g). people are employed in this sector). As Algeria imports 45 percent Because Yemen’s population is already exposed and vulnerable to of its food, however, it is exposed to increasing global food prices, water shortages, climate change could have a large future impact which may eventually lead to migration if demand can no longer on migration (Joseph and Wodon 2013). be satisfied (Brown and Crawford 2009; Wodon et al. 2014). According to Grant et al. (2014) in a study on the causes of As of 2009, a large share of the Nile Delta and Nile valley migration, migrants in Morocco and Egypt were motivated more populations had migrated from the fertile rural areas to Cairo. frequently by socioeconomic opportunities and freedom, whereas This movement was mostly driven by unemployment and poverty, for example Algerians, and Yemenis were motivated to migrate which was in turn caused by land degradation and water scarcity as a result of inhospitable climatic conditions and frequent crop (Afifi, 2009). Due to high population growth, the yearly water failures. Besides agriculture, climate change will probably have quota from the Nile is reported to be insufficient for the increas- impacts on other sectors and economic branches, including energy ing population that is still using outdated and inefficient irrigation production, coastal infrastructure, manufacturing, and tourism methods (Afifi, 2009).The steady reduction in water availability (Gubert and Nordman 2010). These impacts will also contribute per capita in Egypt will further drive migration in the country. to loss of livelihoods and constitute push factors for migration. 143 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 4.21: The “Arc of Tension.” According to Werz and Conley (2012), p.18. Migratory routes from Nigeria and Niger to the densely populated coastal areas of Morocco and Algeria are areas that are prone to climate-change-induced hazards and suffer from weak governance and frequent conflicts. As identified by Wodon et al. (2013) and by Werz and Conley There is common consensus among the research community (2012), the MENA countries are also a popular destination and that climate change functions as a “threat multiplier” (Center transit region for migrants coming from Sahelian and Sub-Saharan for Naval Analysis 2007), amplifying such preexisting threats as African countries. This Sub-Saharan in-migration and transit migra- political instability, high levels of poverty, unemployment, and tion constitutes an additional population burden for the Maghreb other factors. This implies that conflict-proneness and sensitivity countries, which are themselves weakened by climate change and to climate change through adaptation failure derive from insti- weak governance (see Figure 4.21). tutional fragility and poor governance (Smith and Vivekananda 2012). Gemenne et al. (2014) suggest extending the focus in this 4.4.4.2 Climate Change, Conflict, research field, so that not only risks and threats, but also factors and the Security of Nations of peace and cooperation, capabilities, and the power of institu- The identification of climate change as a direct cause of conflict tions, are taken into account. and insecurity is challenging. To date, research has been able to Sterzel et al. (2013) showed the relationship between the present some evidence of a statistical correlation; the underlying vulnerability profiles of small dryland farms (composed of envi- causal relationship between climatic change, the occurrence of ronmental scarcities, resource overuse, and poverty-related factors) conflicts, and insecurity has not however as yet been explicitly and the occurrence of violent conflicts. They stated that it is the explained (Gemenne et al. 2014). For this reason, this report extent of resource overuse that determines conflict proneness focuses on the studies that establish relationships between climatic under conditions of intermediate resource availability; they found changes and conflicts. that poverty is the crucial factor when resource availability is A framework by Scheffran et al. (2011) summarizes the either very poor or rather good (Sterzel et al. 2013). Competition climate-society interaction. Climate change puts pressure on over scarce resources, such as land and water, may constitute a natural resources, which in return can have adverse impacts considerable threat to security, and may even foster conflicts, for human security. Elements which make up human security especially when freeriding56 occurs (Brown and Crawford 2009; include access to water, food, energy, transportation, health, and livelihoods, as well as education, lifestyle, and community. Where these aspects of human security are no longer guaranteed, 56 Taking advantage of a commonly shared resource by intensifying individual use possible consequences include societal instability (both violent beyond one’s own fair share, while other resource users respect a fair allocation. and non-violent conflicts). This may lead to overexploitation of the collective resource. 144 MI DDLE E AST A ND NOR TH A FRICA Osman-Elasha 2010). In Yemen, water scarcity has led to conflicts In brief, the scientific discourse shows that this emerging field over water wells (Hoffman and Werz 2013). of scientific inquiry has not yet succeeded in establishing a con- Water scarcity in the MENA region will increasingly consti- sensus on primary causes, mechanisms, links, and interventions tute a problem for the population—and, in particular, for those between climate change and conflicts and insecurity (Gemenne et engaged in agriculture. The problem will be exacerbated in the al. 2014). Adger et al. (2014) concluded that it is not yet possible Maghreb region by natural population growth in combination to make confident statements about the present or for the future with in-migration, and, in the Mashrek region (e.g. in Iraq), by as to how changes in climate may affect armed conflict. This is weak institutions together with poor resource management (Maas due to the absence of commonly supported theories and a lack and Fritzsche 2012; Sowers et al. 2011; Wodon et al. 2014). These of evidence about causality. In the following section, the Arab findings are consistent with a study on Sub-Saharan Africa by Spring serves as case study to illustrate the possible implications Burke et al. (2009), who found the variation in precipitation and of climate change on national security in the MENA region. temperature affecting the agricultural performance to be the major mechanism linking global warming and conflict in Africa; they 4.4.4.3 A Possible Link between Climate Change and the Arab Spring? concluded that global warming increases the risk of civil war. Finding a possible link between climate change and the Arab Studies from the peace research community disagree. Drought, spring, with its series of violent events, seems far-fetched. Indeed, in particular, is thought to be unlikely to cause civil war (Theisen a more global view has to be taken in order to make the causal et al. 2011), even though the authors admit that climate events links between climate patterns and the Arab Spring visible. Original may cause poverty (which in turn can lead to conflict). Addition- empirical evidence is provided by Lagi et al. (2011) and Sternberg ally, a multivariate regression analysis revealed that countries that (2011, 2012), who attribute the outbreak of violence in Egypt not experience climatic disasters are less likely to experience civil war only to political instability, unemployment, and poverty but also (Slettebak 2012). to a food crisis induced by extreme climatic conditions and market Studies linking climate change to conflict are somewhat mechanisms. Lagi et al. (2011) found this link by analyzing the contradictory. From an econometric perspective, and based on timing of protests and global food price peaks (see Figure 4.22). historical data, Bergholt and Lujala (2012) argue that while certain Sternberg (2012) quantified the drought in China’s eastern wheat types of climate-related disasters may be serious threats for the belt in late 2010 and early 2011, and displayed China’s and Egypt’s economy of the affected countries, this still does not necessar- reaction to this drought and to the world price for wheat. Both Lagi ily increase the risk of armed conflict. By reanalyzing the data et al. (2011) and Sternberg (2012) argue that these rather-indirect of Burke et al. (2009), Buhaug (2010) attempted to show that causes of the Arab Spring have received only little attention. Both weather or climate patterns do not increase the risk of civil war provide studies with original empirical evidence, in contrast to in Sub-Saharan Africa. This study was criticized by Hsiang and research produced by several policy-related institutions, which is Meng (2014), however, for incorrect and insufficient statistical often based on documentary review (Femia and Werrell 2013; e.g. testing. Another recent study by Hsiang et al. (2013) extended Johnstone and Mazo 2011, 2013). the analysis of temperature influence to types of violence (e.g. Egypt has high population growth rates (World Bank 2013h). violent crime, domestic crime, and murder). The authors found Being a largely arid country, Egypt is not able to meet domestic that even small changes from the mean climate toward warmer food demand without importing wheat on a large scale. This temperatures or more extreme rainfall increases the frequency dependency on wheat imports exposes the country to commodity of interpersonal violence by four percent and the frequency of price fluctuations and variability in commodity supplies on the intergroup conflict by 14 percent in median estimates. This study world market. Generally, food expenditures in Egypt are at about was criticized among the scientific community for not being com- 38 percent of income (Sternberg 2012). It is important to note that prehensive enough and ignoring important controversial findings Egypt’s unemployment and poverty rates are relatively high, with from previous studies (e.g., Buhaug 2014). Most recently, Selby unemployment at nine percent as of 2010 (World Bank 2013i) and and Hoffmann (2014) reversed the mainstream thought on water 15.5 percent of the population living on less than $2 per day (as scarcity, state failures, and under-development and introduced of 2008—World Bank 2013j). Bread subsidies help to stabilize the a model of environment-conflict relations focusing on resource social order in Egypt, with the government spending three percent abundance, globally embedded processes of state-building, and of the country’s GDP on wheat subsidies (Sternberg 2013). development. They suggested that violent conflict can also emerge Drought conditions in eastern China in late 2010 and early 2011 from water abundance accompanied by inefficient management lead to a reduction in the winter wheat harvest (Lagi et al. 2011). and state-directed processes of economic development as well as Sternberg (2012) estimates the shortfall at 10 million tons of wheat, from internal colonization. However, this model is only applied or approximately 10 percent of annual wheat production. Earlier to investigate links between water issues and conflict and does in the same year, Russia and Ukraine also experienced drought not take into account other kinds of climatic issues. 145 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 4.22: Food prices and conflict. Time dependence of the FAO Food Price Index (Jan. 2004–May 2011). Red dashed lines mark the beginning dates of “food riots” and protests in the MENA region. Numbers in () show overall death tolls. The blue dashed line marks the day (12/13/2010) when Lagi et al. (2011) warned the U.S. government of the link between food prices, social unrest, and political instability. Inset: FAO Food Price Index 1990–2011. Source: Lagi et al. (2011), p. 3. conditions; meanwhile, heavy rainfall in Canada and Australia led Mazo 2013), and as food became unaffordable for the population to poor wheat harvests (Sternberg 2012) and decreases in wheat again in 2011, social conflicts developed (Sternberg 2013). While production.57 In 2011, this drop in supply lead to an increase in the partly the result of economic and political instability, conflicts world market price for wheat, a commodity for which 18 percent were also due to discontent with the food supply situation as, at of the total global harvest is slated for export (Sternberg 2012). the time, 50 percent of the population relied on ration cards and China, a country with a high population growth rate, saw the country’s bread subsidy system was fraught with corruption the need to secure its domestic wheat demand. While China’s (Johnstone and Mazo 2013). Johnstone and Mazo (2013) report wheat production was reduced by only 0.5 percent, the country’s similar linkages between food prices and social unrest in Algeria, wheat consumption increased by 1.68 percent (Sternberg 2013). with a correlation between unemployment and high food prices Consequently, China made a large-scale wheat purchase from the for sugar, oil, and other staples. world market. Russia levied an export ban on wheat due to its In high income countries such as Israel or the United Arab domestic shortage, decreasing imports to Egypt by five percent Emirates, by contrast, a relatively low percentage of income is (Sternberg 2012). Johnstone and Mazo (2013) report that Egypt spent on food and early stabilization of domestic food prices was received only 1.6 million tons of wheat from Russia in the second been successful. These countries were spared from protests and half of 2010 as compared to 2.8 million tons in the same period social conflicts during the same time period (Sternberg 2013). It in the previous year. These reductions in global supply lead to must be noted, however, that some Gulf monarchies purchased or further increases in commodity prices. leased agricultural land from highly volatile nations and regions, The climatic events (droughts and floods), together with such as Ethiopia, Sudan, and countries in South Asia to meet their global market forces, were contributing factors to high wheat domestic demand (ILC et al. 2014; Spiess, 2012). This is in line prices in Egypt and affected the price of bread. As Egypt had not with the conclusions of Gemenne et al. (2014) and Adger et al. yet fully recovered from its 2008 food price crisis (Johnstone and (2014) that a government response to price and scarcity signals is necessary to manage access to resources and markets. Lagi et al. (2011) suggest that the food price is crucial for nations’ social 57 Wheat production in 2010 was down 32.7 percent in Russia, 19.3 percent in stability, and that persistently high food prices might result in a Ukraine, 13.7 percent in Canada, and 8.7 percent in Australia (Sternberg, 2011, 2013). global increase in social disruptions. 146 MI DDLE E AST A ND NOR TH A FRICA Figure 4.23: Aggregated FAO Food Price Index and its sub-indices. The Food Price Index score in January 2011, when the Egyptian crisis broke out (black dashed line), was 238. This is particularly due to the high Sugar Price and Oils Price Indices, together weighted with 0.207, which is about one-fifth of the Food Price Index. The low Meat Price Index pulled it downward, constituting one third of the Food Price Index. Data taken from FAO (2014c). Weight data from FAO (2014a). Even though the link between the drought in China, harsh and institutions weakened by crisis or war have limited capacity to climatic conditions in other countries, and social conflicts in Egypt build resilience and, as a consequence, have less capacity to cope is plausible, it is not directly verifiable (Sternberg 2012). These find- with and adapt to climate change. The case of the Arab Spring ings should therefore be treated with caution. In addition, there is indicates that climate-related extremes and market mechanisms no evidence so far that the late 2010 and early 2011 drought events together may have serious implications for food security. Findings in China and Eurasia can be attributed to climate change. Attention on the link between climate change, human security, and conflict should be given to the methodologies of the two major studies are in agreement with recent IPCC analyses on human security scrutinized in this chapter. Whereas Sternberg (2011, 2012 and (Adger et al. 2014). 2013) used the world market price for wheat in his analysis, Lagi et al. (2011) used the FAO Food Price Index, which is a composite 4.4.5 Coastal Infrastructure and Tourism index of five differently weighted sub-indices (FAO 2014a). Since wheat was identified as the critical commodity in the Egyptian Globally, human populations, as well as agricultural, industrial, food crisis (Sternberg 2012), results based on the aggregated Food and other economic activity, tend to be concentrated in coastal Price Index by Lagi et al. (2011) seem somewhat biased, as wheat zones. Historically this has been amplified in MENA countries, is only represented in the Cereals Price Index along with rice and where coastal cities and agricultural areas have been particularly maize, while the Meat Price Index is the sub-index with the great- important due to the aridity of inland regions (Verner 2012). The est influence on the Food Price Index (Figure 4.23). This suggests population of MENA’s coastal cities was approximately 60 mil- that the results by Lagi et al. (2011) are less confident than those lion in 2010, with the number expected to reach 100 million by of Sternberg (2012)—and that the conclusion on the relationship 2030 (World Bank 2011b). In Morocco, for example, more than between FAO Price Index and the Egyptian food crisis, by Lagi et 60 percent of the population and over 90 percent of industry is al. 2011, should be treated carefully. located in key coastal cities (Snoussi et al. 2009). In summary, it can be said that countries with well-established As coastal populations and assets in coastal areas continue institutions have the capacity to build high levels of resilience and to grow, exposure to impacts associated with sea-level rise is also to respond to climatic changes, extreme events, and other shocks increasing. This is particularly true as populations expand into such as climate-related increases in food prices over time. Countries low-lying areas, and as wetlands and other ecosystem protections 147 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL against floods are removed (World Bank 2011b). One study has aggravated by the effects of sea-level rise (Kouzana et al. 2009; estimated that annual global flood losses are expected to increase Zarhloule et al. 2009). from $5 billion in 2005 to $52 billion by 2050 due to socioeconomic changes alone (Hallegatte et al. 2013). Climate change and coastal 4.4.5.2 Impacts of Sea-Level Rise subsidence will exacerbate this growing vulnerability. and Storms on Coastal Areas Several global studies have noted that while regions such as Asia 4.4.5.1 Types of Physical Impacts and Sub-Saharan Africa tend to be more vulnerable in terms of the The key impacts of climate change in coastal zones, and in rapidly total population exposed to the impacts of sea-level rise, MENA growing urban areas in particular, are expected to include inun- countries tend to be more vulnerable in terms of the percentage dation resulting from slow onset sea-level rise, floods, damage of population at risk (Dasgupta et al. 2011; Dasgupta et al. 2009; caused by extreme events (including storms and storm surges), Ghoneim 2009; Nicholls et al. 2011). and increased erosion (Brecht et al. 2012; Hunt and Watkiss 2011). In terms of the percentage of population exposed to a 1-meter Several MENA countries are highly exposed to sea-level- sea-level rise, Dasgupta et al. (2009) identified three MENA countries rise-related inundation due to their low-lying topographies. One among the ten most vulnerable in the world: Egypt (9.28 percent), study found that with one meter of sea-level rise, Qatar could lose Tunisia (4.89 percent), and the United Arab Emirates (4.59 percent). 2.7 percent of its total land area and Egypt would lose 13.1 per- One study of Africa suggested that, in the absence of adaptation, cent of its agricultural area (Dasgupta et al. 2009). In terms of 1.97 million people in Egypt could be affected by a sea-level rise the percentage of urban area lost to one meter of sea-level rise, of 0.54 m and 1.82 million people in Morocco could be affected by Egypt (5.52 percent), Libya (5.39 percent), the United Arab Emir- a sea-level rise of 0.44 m with 2.6°C global warming compared to ates (4.8 percent), and Tunisia (4.5 percent) were all found to be 1990 levels (Brown et al. 2011) (Table 4.7). Another study identified highly vulnerable (Dasgupta et al. 2009). Egypt, Tunisia, Morocco, and Libya as among the most vulnerable Global comparative studies have identified Egypt’s Nile Delta African countries in terms of total population affected by sea-level coastline as one of the most vulnerable areas to sea-level rise rise under scenarios of 0.42–1.26 meters in sea-level rise by 2100, (Dasgupta et al. 2009; Syvitski et al. 2009). Sea-level rise impacts assuming no adaptation (Hinkel et al. 2012). in the Nile Delta will be exacerbated by land subsidence, especially In terms of the GDP impact of a 1-meter sea-level rise, Egypt in the eastern part of the delta, and by extensive landscape modi- (6.44 percent) and Tunisia (2.93 percent) were also in the top fications resulting from both coastal modification and changes in ten countries globally (Dasgupta et al. 2009). In the absence of the Nile’s hydrogeology (Frihy and El-Sayed 2013; El Sayed et al. adaptation, the annual economic impact of sea-level rise in 2100 2010; Wöppelmann et al. 2013). under the A1B scenario with 2.6°C global warming compared to A study in Abu Dhabi found that, in the absence of adapta- 1990 has been calculated at $6.55 billion for Algeria, $6.52 billion tion measures, a 2-meter sea-level rise would inundate 15.9 per- for Egypt, $5.52 billion for Morocco, $3.46 billion for Tunisia, cent of Abu Dhabi’s urban area, increasing to 39.6 percent for and $1.76 billion for Libya (Brown et al. 2011). Table 4.7 shows a 3-meter sea-level rise (Ksiksi et al. 2012). The study estimates the corresponding sea-level rise and the economic impacts with that 20 percent of Abu Dhabi’s total land area would be inundated different global warming scenarios. by a 2-meter sea-level rise, and 30 percent by a 3-meter sea-level Another study considered a 1.26-meter sea-level rise in the rise (Ksiksi et al. 2012). absence of adaptation and calculated Egypt’s annual losses at Apart from direct inundation, sea-level rise is expected to have $5 billion per year, while the annual losses for Tunisia, Morocco, serious impacts in terms of saltwater intrusion, the salinization and Libya were estimated at less than $500 million per year (Hinkel of groundwater, rising water tables, and impeded soil drainage et al. 2012). When the effects of adaptation were included in the (Hunt and Watkiss 2011; Werner and Simmons 2009). These projections, Tunisia, Morocco, and Libya were no longer among effects can reach inland for several kilometers following relatively the most vulnerable countries and Egypt’s ranking had fallen small increases in sea level (Werner and Simmons 2009) (see Sec- from first to third (Hinkel et al. 2012). This indicates that for most tion 4.4.2, Salinization). North African countries adaptation could be effective at reducing Saltwater intrusion into coastal aquifers has been documented vulnerability to sea-level rise. across MENA, including in Tunisia, Egypt, and Israel (Kashef Another expected consequence of climate change is a greater 1983; Kerrou et al. 2010; Kouzana et al. 2009; Yechieli et al. 2010). intensity of storms (Knutson et al. 2010). Coupled with higher sea The principal causes of observed saltwater intrusion have been levels, more intense storms are likely to result in more powerful identified as the over-extraction of water for supplementary irri- storm surges. gation and reductions in the recharge of aquifers. These factors One study has calculated that under a 1-meter sea-level rise are likely to remain immediate challenges, although they will be and a 10 percent increase in storm intensity, 50 percent or more 148 MI DDLE E AST A ND NOR TH A FRICA Table 4.7: Damage and people affected by sea-level rise. TOTAL RELATIVE SEA COSTS OF LAND LOSS NET LAND SEA-LEVEL FLOOD RESIDUAL DUE TO LOSS DUE PEOPLE CHANGE COSTS DAMAGE SUBMERGENCE TO EROSION FLOODED SINCE 1995 (MILLIONS (MILLIONS $/YR) (KM²/YR) (KM²/YR) (THOUSANDS/YR) (M) $/YR) B1 Algeria 328.1 0 0 35.2 0.17 328.1 Djibouti 174.5 0 0 9.5 0.15 174.5 Egypt 2,134.5 38.6 0.79 927.6 0.25 1,482.9 Libya 186.7 3.2 0.31 3.7 0.21 167.1 Morocco 1,195.7 0 0.07 47.1 0.16 1,178.0 Tunisia 722.8 0 0.36 17.0 0.20 710.3 A1B Algeria 6,546.6 7.1 0 435.4 0.46 916.7 Djibouti 232.0 0.3 0 85.5 0.42 213.6 Egypt 6,518.5 19.4 1.4 1970.3 0.54 3,482.5 Libya 1,756.8 16.1 0.8 39.4 0.50 477.6 Morocco 5,524.3 14.5 0.4 1820.2 0.44 3,388.3 Tunisia 3,459.7 73.3 0.8 263.9 0.49 1,798.1 A1F1 Algeria 5,238.0 21.7 0 708.1 1.06 1,454.8 Djibouti 295.4 2.0 0 92.4 0.99 237.4 Egypt 12,476.3 43.6 2.9 3600.7 1.14 5,362.4 Libya 1,325.0 12.8 1.9 131.2 1.11 745.2 Morocco 5,886.5 4.8 1.1 2078.9 1.04 4,001.3 Tunisia 4,541.0 17.8 1.7 802.8 1.09 2,375.0 Numbers are for 2100 without adaptation in six MENA countries under SRES B1, A1B and A1F1 with global warming of 1°C, 2.6°C and 6.1°C relative to 1990, respectively. Source: extracted from (Brown et al. 2011). Box 4.3: The Nile Delta For Egypt, and assuming no adaptation, annual damages have been projected in the range of $5 billion by 2100 for a 1.26-meter sea-level rise (Hinkel et al. 2012) and $14.8 billion by 2100 under the A1B scenario (Brown et al. 2009). Dasgupta et al. (2009), meanwhile, projected losses of 25 percent of the Nile Delta’s land area, affecting 10.5 percent of the population and 6.4 percent of GDP with a 1-meter sea-level rise (Dasgupta, Laplante, Meisner et al. 2009). Another study estimated the value of assets exposed to a 0.5-meter sea-level rise by 2070 at $563 billion in the city of Alexandria alone (Hanson et al. 2011). The Nile Delta, however, has an extensive history of coastal defense that dates back to the construction of the Mohamed Ali Sea Wall in 1830. Modern defensive works and other infrastructure, including the International Coastal Road, provide further protection against flooding (Frihy and El-Sayed 2013). Studies by Egypt’s Coastal Research Institute concluded that, under the A1F1 scenario, 14.4 percent of the Nile Delta would be at risk from inundation by 2100; this fell to three percent when taking into account the value of the Mohamed Ali Sea Wall (EEAA 2010). Another study found that, although 22.5–29.2 percent of the area was vulnerable to inundation depending on the sea-level rise scenario and assuming no adaptation, much of that impact would fall on undeveloped lands and wetlands, with built-up land accounting for less than five percent of the impact (Table 4.8) (Hassaan and Abdrabo 2013). 149 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Table 4.8: Results for three scenarios of sea-level rise in the Nile Delta assuming no adaptation. INUNDATED DELTA AREA DISTRIBUTION OF INUNDATION IMPACT UNDEVELOPED PROJECTED BUILT CULTIVATED LAND AND SCENARIO RELATIVE SLR KM2 % LAND LAND WETLANDS A1F1 1.04 m 4,006 22.5 4.4% 42% 53.6% Rahmstorf 1.85 m 7,336 42.2 3.8% 60.7% 35.5% Pfeffer 2.45 m 8,769 49.2 3.7% 64% 32.3% Source: Extracted from Hassaan and Abdrabo (2013). of the coastal population in Kuwait, Djibouti, the United Arab with assumptions for expected coastal urbanization by 2030 and Emirates, and Yemen will be at greater risk from storm surges increases in both storm frequency and the extent of the coastline (Dasgupta et al. 2011). The effects of climate change would leave at high risk of erosion, could lead to adaptation costs of $12 mil- 2.7 million more people in Alexandria and 1.2 million more people lion per year (Ennesser et al. 2010). in Aden exposed to storm surges (Dasgupta et al. 2011). The same In Morocco, Snoussi et al. (2009) showed that significant study calculated that, for the MENA region, storm intensification potential land and infrastructure losses would result from a relative would cause additional annual losses to GDP of $12.7 billion by 0.86-meter sea-level rise by 2100 for the city of Tangiers. These 2100, with exposure particularly high in Kuwait, the United Arab potential losses included 36 percent of roads, 79.2 percent of Emirates, Morocco, and Yemen (Dasgupta et al. 2011). railways, 99.9 percent of port infrastructure, 90.8 percent of water Dasgupta et al. (2011) considered only the local effects of diversion canals, 63.4 percent of the city’s industrial zone, and sea-level rise in developing their projections. But the MENA 34.8 percent of the city’s high density urban area, largely due to region benefits from being bordered by semi-enclosed bodies of the increased area at risk from storm surges (Snoussi et al. 2009). water,58 and only those MENA nations exposed directly to the Indian Ocean (e.g., Yemen, Oman, and Djibouti) are regularly 4.4.5.3 Impacts on Coastal Tourism exposed to tropical storms, while the west coast of Morocco is Egypt, Morocco, Tunisia, Jordan, Lebanon, and Syria have developed exposed to Atlantic storms (Becker et al. 2013; Knapp et al. 2010). tourism services aimed at the international market (Safa and Hilmi This could mean that their projections for MENA countries other 2012). In 2006, international visitor spending contributed more than Djibouti and Yemen are less conservative than they might than 10 percent of GDP to the economies of Morocco, Bahrain, otherwise appear. Lebanon, and Jordan, and more than five percent to the economies Urban growth in coastal areas means that, even in the absence of Egypt, Tunisia, and Syria (Gössling et al. 2009). In addition of sea-level rise, increasing losses and damages can be expected to beach tourism, cultural tourism has been highly important in due to higher concentrations of people and assets exposed to flood- Jordan and Egypt in particular, and religious tourism has played ing. By separating socioeconomic drivers of vulnerability from the a key role in Saudi Arabia (Khattabi 2009; Safa and Hilmi 2012). effects of sea-level rise, a study of 136 coastal cities59 identified Many MENA countries also have well-developed domestic beach Alexandria, Benghazi, and Algiers as particularly vulnerable to tourism, with domestic tourists visiting the relatively cool coastal 0.20.4 m sea-level rise by 2050 (Hallegatte et al. 2013). The study areas during the hot summer months. estimates that, in the event of a failure of flood defenses, the effects Tourism infrastructure in coastal areas (e.g., hotels and mari- of 0.2 m sea-level rise would increase damages from $16.5 billion nas) and tourism assets (e.g., protected areas) are vulnerable to $50.5 billion in Alexandria, from $1.2 billion to $2 billion in to sea-level rise, although few specific assessments have been Benghazi, and from $0.3 billion to $0.4 billion in Algiers. Annual conducted for the MENA region (see Box 4.4 for the example of losses would increase to $58 billion, $2.7 billion, and $0.6 bil- Morocco). Comparison of slow-onset sea-level rise and shocks lion with 0.4 m sea-level rise for these three cities respectively. A to tourism demand have, however, indicated that short-term study for the city of Tunis found that a 0.2-meter sea-level rise, shocks resulting from fluctuations in tourist demand60 have a 58 The Mediterranean Sea, Red Sea, and Arabian Gulf are all semi-enclosed seas. 59 60 Including Alexandria, Beirut, Benghazi, Tel Aviv, Kuwait City, Dubai, Jeddah, Tourist demand relates to demand in tourism destination choices and tourist and Rabat. demand for recreational activities, hotels, restaurants, and related expenditures. 150 MI DDLE E AST A ND NOR TH A FRICA Box 4.4: Impacts of Climate Change on Tourism in Morocco In Morocco, different studies have explored the complex interlinkages between coastal development, climate vulnerabilities, and tourism. During the peak summer months, the population of Morocco’s Mediterranean area doubles from eight to 16 percent of the national population; this has driven rapid coastal urbanization (Anfuso et al. 2011). Urbanization has contributed to the increased erosion of Morocco’s sandy beaches, thereby increasing vulnerability to coastal flooding (Ahizoun et al. 2009; Anfuso et al. 2011). Ambitious plans to expand coastal tourism in Morocco could exacerbate coastal vulnerabilities by increasing the amount of exposed assets and as coastal modifications weaken the coast’s natural defenses (Anfuso et al. 2011). Snoussi et al. (2009) found that 99.9 percent of sandy beaches and 84.5 percent of tourism infrastructure in Tangiers would be lost to an 0.86-meter sea-level rise by 2100. Snoussi et al. (2008) projected that for the area between Saidia and Ras el Ma, 50 percent of sandy beaches would be eroded by 2050 under a 0.39-meter sea-level rise and 70 percent by 2100 for a scenario of a 0.86-meter sea-level rise, with consequent impacts on tourism. Other studies in Morocco have shown that climate change impacts on water supply and demand are likely to increase competition for water between tourism and other uses, which may ultimately affect the economic performance of specific tourism services (Tekken et al. 2009, 2013; Tekken and Kropp 2012). This is a particular issue in the context of growing luxury tourism, which requires high water demand for golf courses, swimming pools, and other leisure facilities (Tekken et al. 2013; Zarhloule et al. 2009). Together, these factors could contribute to a structural water deficit, requiring policy, planning, and investment to manage (Tekken and Kropp 2012). larger economic impact than slow onset changes, particularly in countries under conditions of climate change (Gössling et al. developing countries (Bigano et al. 2008). 2009). Developing countries that invest in tourism infrastructure Beach tourism services are also vulnerable to other aspects are dependent on future tourist numbers, whilst tourists are of climate change, including temperature increases and changes able to make short-term decisions each season in response to in water and energy supplies (Gössling et al. 2012; Moreno and warmer conditions. However, a pleasant climate is just one draw Amelung 2009; Perch-Nielsen 2010; Scott et al. 2011). A multi- for tourists, who can also be drawn to a location by its culture, criteria analysis of the vulnerability of beach tourism in 51 countries proximity, safety, standards, level of service, and the cost of food found that, of the MENA destinations, Egypt and Saudi Arabia are and accommodations (Gössling et al. 2012). Overall, the effects of moderate to highly vulnerable to climate change and that Morocco climate change on tourism are likely to be small in comparison and Israel are low to moderately vulnerable (Perch-Nielsen 2010). to changes induced by population and economic growth and/ Some studies modeling the impact of climate change on inter- or conflict and political stability (Gössling et al. 2012; Hamilton national tourism have found overall reductions in international et al. 2005; Safa and Hilmi 2012). tourism, with tourist preferences shifting to higher latitudes and altitudes. One study found that a 1°C temperature rise by 2025 4.4.6 Energy Systems would result in declines of 10–25 percent in international arrivals across the MENA region (Hamilton et al. 2005). Another study Energy access is a key requirement for development. Many economic projected a reduction in tourism demand of eight percent under activities depend on ample and reliable electricity access (Akpan et the A1B scenario by 2050 (Bigano et al. 2008). However, these al. 2013); similarly, at the individual and household level, electric- models do not necessarily capture the seasonal dynamics of tour- ity access enables income-generating activities, increases safety, ism markets, and reductions in summer months may be matched and contributes to human development (Deichmann et al. 2011). by increases in spring and autumn arrivals (Amengual et al. 2014; In the Middle East and North Africa, almost all of the population Giannakopoulos et al. 2013). has access to electricity in both rural and urban areas. However, in Coastal tourism in the MENA region may also experience several countries (notably Yemen, Morocco, Tunisia, and Algeria) indirect impacts of climate change through pathways such as electricity consumption per capita remains low compared to other biodiversity loss and ocean acidification (Rodrigues et al. 2013). Of nations in the region (see Table 4.9) (World Bank 2013l). particular concern are the potential impacts of ocean acidification Climate change impacts are projected to affect electricity and warming on coral reefs, which provide ecosystem services production and distribution both globally and within the region valuable to tourism in MENA countries, especially Egypt (Hoegh- (Sieber 2013), and MENA countries will have to increase or at least Guldberg et al. 2007). Perceptions of poor ecological health can maintain electricity production at current levels in order to support create additional vulnerabilities for tourism operators and reduce economic development and population growth. In general, three the economic value of ecosystems to tourism (Marshall et al. 2011). types of climate-change-related stressors could potentially affect These assessments have raised concerns regarding the thermal power and hydropower generation: (1) increased air tem- viability of tourism as a central economic strategy in developing peratures that reduce thermal conversion efficiency; (2) decreased 151 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL water availability (and increased temperatures) for cooling; and electricity. A much smaller fraction of the electricity consumed in (3) extreme weather events (Han et al. 2009; Sieber 2013). Indeed, the MENA region is generated by hydroelectric plants, which are extreme weather events and climate change affect not only power also vulnerable to the impacts of climate change as river runoff plants but also the grid systems and overall reliability of electric- is projected to be modified, with a more pronounced seasonality ity production; this may contribute to an increase in the price of and an overall decrease in water volume expected (Evans 2008, electricity and could even lead to power outages (Ward 2013). In 2009). Table 4.9 displays the percentage of electricity production this context, thermal electricity and hydroelectricity are projected by source and the total electricity power consumption per capita to be the most vulnerable electricity sources. in the different MENA countries. With the exception of Yemen, close to 100 percent of the population of the MENA region has access to electricity. In Yemen, 4.4.6.1 Energy Mix only about 40 percent of the population had access in 2011 (World The MENA countries are heavily dependent on thermal electric Bank 2013k), which is particularly low compared to the other sources for electricity production. The main characteristics of countries of the region. Electricity production capacity in Yemen these thermoelectric sources, including natural gas, coal, oil, could progressively increase according to the increasing electric- and nuclear, is the use of a cooling system to ensure the safe ity demand in the country. Taking into account demographic and transformation of heat produced by the combustion of fuels into socioeconomic development trends, it is projected that demand for Table 4.9: Electricity production from hydroelectric and thermoelectric sources, including natural gas, oil, coal, and nuclear in 2011. ELECTRICITY POWER ELECTRICITY PRODUCTION ELECTRICITY PRODUCTION FROM CONSUMPTION FROM HYDROELECTRIC THERMOELECTRIC SOURCES COUNTRY NAME (kWh PER CAPITA) SOURCES (% OF TOTAL) (% OF TOTAL) United Arab Emirates 9,389 0 100 Bahrain 10,018 0 100 Djibouti – – – Algeria 1,090 1.0 99.0 Egypt, Arab Rep. 1,743 8.3 91.8 Iran, Islamic Rep. 2,649 5.0 95.0 Iraq 1,343 7.6 92.4 Israel 6,926 0 100 Jordan 2,289 0.4 99.5 Kuwait 16,122 0 100 Lebanon 3,499 4.9 95.1 Libya 3,926 0 100 Morocco 826 7.5 92.5 Oman 6,292 0 100 Qatar 15,755 0 100 Saudi Arabia 8,161 0 100 Syrian Arab Republic 1,715 8.0 92.0 Tunisia 1,297 0.3 997 West Bank and Gaza – – – Yemen, Rep. 193 0 100 Data sources: World Bank (2013e, f; g; h; i; j). 152 MI DDLE E AST A ND NOR TH A FRICA electricity in the Middle East is going to increase from 822 TWh to As a substantial share of the electricity produced in the MENA 2010 TWh in 2030 (Granit and Löfgren 2010). In order to meet this region originates from thermal electric sources, the findings of increased demand for electricity, it is estimated that the countries the study by van Vliet et al (2012) give relevant insights into the of the region may have to invest approximately $4 trillion (Granit potential exposure of the sector. According to studies such as those and Löfgren 2010). from Bozkurt and Sen (2013) and Kaniewski et al. (2012), MENA MENA countries depend on both hydroelectricity and thermal could similarly be exposed to reduced river runoff, increased sea- electricity for their power needs. With a projected change in water sonality, and increased water temperatures, which are the main availability, notably induced by a projected overall decrease in biophysical drivers impacting electricity generation in the regions precipitation and river runoff and/or increased seasonality due studied by van Vliet et al. (2012). to climate change impacts, thermal electricity plant cooling sys- As climate change impacts thermal electric production, the tems may become less efficient and electricity production could vulnerability of power generation in all the MENA countries is therefore be affected (Mika 2013; Sieber 2013). A similar threat high. In addition, it is projected that economic development and a can be observed for hydroelectric power generation, for which a growing population in the region will increase energy demand (EIA change in water availability is also projected to affect electricity 2013). In addition to increased energy demand induced by socio- generation (Hamududu and Killingtveit 2012). economic development, climate change is projected to decrease the production of thermal electric power plants in other regions—the 4.4.6.2 Climate Change and Energy Production countries in the MENA region could also be put under increasing There appears to be a lack of studies specifically quantifying, pressure, contributing to a rise in electricity prices and a risk of at the regional level, the impacts of climate change on thermo- electricity shortages in the absence of any adaptation measures. electricity generation in the MENA region. However, drawing upon the conclusions of studies assessing the impacts of climate Hydropower change on energy systems at the global level or in other regions Hydropower plays a minor role in electricity production in the provides qualitative and quantitative indications regarding the MENA region. Some countries, such as Iraq (7.6 percent of the potential effects of climate change on electricity generation in total electricity produced), Egypt (8.3 percent), Morocco (7.5 per- the MENA region. cent), and Syria (8.0 percent), generate more than five percent of their electricity from hydroelectric sources (see Table 4.9 for Thermal Electric Generation more details). In this regard, it is therefore also prudent to exam- Assessment of the impacts of climate change on thermal electric ine hydroelectric production in the MENA region in the context generation in the Middle East and North Africa are lacking. However, of climate change. these impacts have been studied for European countries and the The core resource for hydroelectricity is river runoff, which United States; they provide qualitative and quantitative indica- needs to be stable (both inter- and intra-annually) in order for tions for potential climate change impacts. A study by van Vliet et hydropower plants to operate most efficiently (Hamududu and al. (2012) took into account the effects of changes in river water Killingtveit 2012; Mukheibir 2013). According to the projections temperature and river flows on thermal electricity production. The for river runoff and water availability in the MENA region under study evaluated these impacts under the IPCC SRES A2 and B1 climate change (see Section 4.4.1 The Agriculture-Water-Food scenarios for the 2040s (2031–2060) and 2080s (2071–2100).61 They Security Nexus), river runoff is projected to decrease and rainfall found that the capacity of nuclear and fossil-fueled power plants seasonality to increase across all scenarios in the coming decades could decrease by 6–19 percent in Europe and by 4–16 percent in (Bozkurt and Sen 2013; Kaniewski et al. 2012). the United States during the period 2031–2060 (2040s) compared A global study by Hamududu and Killingtveit (2012) assembled to the production levels observed from 1971–2000. Furthermore, 12 different climate models62 to project future rates of river runoff due to the increased incidence of droughts and extremely low river and, therewith, to project the electricity produced by hydroelectric flows, the mean number of days during which electricity produc- plants both at the national level and aggregated at the regional tion will be reduced by more than 90 percent was projected to be level. Applying the scenario IPCC SRES A1B, which corresponds multiplied by three compared to present level, from 0.5 day per to a 2.3°C increase by the middle of the 21st century (compared year to 1.5 days per year in Europe, and from 0.8 days to 1.3 days to pre-industrial levels), the authors estimated changes in hydro- per year in the US, both from 2031–2060 in the A2 scenario. power generation across the region. For North Africa, they found that production would decrease by 0.08TWh (or –0.48 percent) 61 The SRES A2 scenario corresponds to a temperature increase of 3.2°C and 1.5°C above pre-industrial levels in 2080 and 2040, respectively. The SRES B1 scenario corresponds to temperature increases of 2.3°C and 1.4°C above pre-industrial levels 62 in 2080 and 2040, respectively. The following three models—ECHAM5/MPIOM, The models used were CGHR, ECHOG, FGOALS, GFCM20, GFCM21, GIEH, HADCM3, CNRM-CM3 and IPSL-CM4—were used. HADGEM, MIHR, MPEH5, MRCGCM, and NCCCSM. 153 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL compared to 2005 production levels. For the Middle East,63 which Scenarios which correspond to a 2°C temperature increase by 2050 in the study also includes Central Asian countries, production was (and 3°C by the end of the century) compared to pre-industrial projected to decrease by 1.66TWh (or –1.43 percent). levels. Demand for space heating in the MENA region is projected The results of the study by Hamududu and Killingtveit (2012) to remain marginal (Cozzi and Gül 2013). need to be interpreted with care, however, as the authors high- lighted two key limitations to their projections. First, the study 4.4.6.4 Synthesis did not take into account changes in the timing of flows (just Based on the current energy mix in the Middle East and North annual river flow). Therefore, the potential impacts of floods and Africa, and the projected impacts of climate change on the region, droughts, which have very significant impacts on hydroelectricity it is difficult to reach a robust conclusion regarding the projected generation and are projected to occur more frequently and with a vulnerability of the region’s energy systems. Whether defining greater intensity in the coming decades, is not taken into account vulnerability as the predisposition to being adversely affected (Bozkurt and Sen 2013; Kaniewski et al. 2012). Second, changes by the impacts of climate change (IPCC 2012) or as a function of in river runoff were calculated at the country level and not at the the system’s sensitivity, exposure, and adaptive capacity (IPCC river basin level. This simplification does not reflect potential 2007), it appears that the MENA countries are vulnerable to dif- spatial variability and changes occurring over short distances. fering extents. Furthermore, in the Middle East and North Africa, surface water Some countries in the region have a very high economic is projected to be significantly affected by changes in precipitation capacity (GDP per capita) that could support the implementa- and decreases in snow coverage at high altitudes (Abdulla et al. tion of adaptive measures, whereas others, such as Yemen, have 2008; Samuels et al. 2010) (see Section 4.4.1, The Agriculture- a weak economic and technical capacity that would not allow Water-Food Security Nexus). River runoff is therefore projected to them to anticipate and cope with the projected impacts of climate decrease or be subject to large fluctuations, which could decrease change on energy systems. In this context, the vulnerability of hydropower supply and reliability. the region should not be regarded as homogeneous, but rather heterogeneous and wideranging. Furthermore, the MENA coun- 4.4.6.3 Climate Change and Energy Demand tries have one of the highest wind and solar energy potentials in As global mean temperatures and the occurrence and intensity of the world (OECD 2013). Exploiting this wind and solar potential heat extremes are projected to increase under climate change (see would strongly help these countries, enabling them to decrease Section 4.3, Regional Patterns of Climate Change), the demand the vulnerability of their existing energy systems to the projected for air conditioning is also expected to rise (Cozzi and Gül 2013). impacts of climate change (OECD 2013). This would also increase At the global level, Isaac and van Vuuren (2009) modeled the electricity production, which is important as demand in the projected demand of cooling as measured in cooling degree days majority of the countries is expected to increase steeply in com- (accumulated daily temperatures above 18°C). For this simula- ing decades due to demographic and economic development as tion, the TIMER/IMAGE model was used under an emissions well as to the increasing need for space cooling as temperatures scenario in which temperatures increase by 3.7°C by the end of rise (Cozzi and Gül 2013). the century. They estimated that by 2100 the number of cooling degree days will rise from 12,800 during the period 1971–1991 to 19,451 in 2100; this corresponds to a 51.9 percent increase com- 4.5 Regional Development Narratives pared to the baseline period. For the same interval, demand for heating (measured in heating degree days) is projected to remain The following implications of climate change are discussed in almost constant. this section in order to relate climate change impacts to existing The International Energy Agency (Cozzi and Gül 2013) esti- and future vulnerabilities in the MENA region: (1) the implica- mates that demand for space cooling in the MENA region could tions for agriculture, water resources, and food security; (2) the increase by seven percent (or 1450 cooling degree days) in 2035 implications for human health and thermal discomfort; and and by 11 percent in 2050 compared to the 2005 energy demand (3) the implications for migration, security, and urban areas. It is for space cooling in the residential sector. These projections are important to note that each development narratives presents only made using the IEA World Energy Model, under the New Policies one of the many possible ways in which climate change can put key development trajectories at risk. Table 4.10 summarizes the 63 key climate change impacts under different warming levels in the Armenia, Azerbaijan, Bahrain, Cyprus, Georgia, Iran, Iraq, Israel, Jordan, Kuwait, Lebanon, Oman, Qatar, Saudi Arabia, Syria, Turkey, the United Arab Emirates, and Middle East and North Africa region and Figure 4.24 summarizes Yemen. the key sub-regional impacts. 154 MI DDLE E AST A ND NOR TH A FRICA Figure 4.24: Sub-regional risks for development in the Middle East and Northern Africa under 4°C warming in 2100 compared to pre-industrial temperatures. e k a s h r M Maghreb &HQWUDO$UDE 3HQLQVXOD 3RSXODWLRQ'HQVLW\ >3HRSOHSHUVTNP@  Southern ² Arab Peninsula ² ² ²  Maghreb Mashrek and Eastern Parts Arabian Peninsula Strong warming reduction in annual Highly unusual heat and decrease in annual Highly unusual heat extremes in central precipitation, increased water stress and precipitation will increase aridity, decrease Arabian Peninsula. In southern parts reduced agricultural productivity. Large in snow water storage and river runoff for relative increase in annual precipitation, but coastal cities exposed to sea level rise. example in Jordon, Euphrates and Tigris. uncertain trend of annual precipitation in Adverse consequences for mostly rain-fed central part. Sea level rise in the Arabian Climate change risks will have severe agricultural and food production. Sea likely higher than in Mediterranean implications on farmers’ livelihoods, country and Atlantic coasts with risk of storm economy, and food security. Exposure of Climate change risks will have severe surges and adverse consequences for critical coastal assets would have impact implications on farmers’ livelihoods, country infrastructure. on the economy, including tourism. There is economy, and food security. There is a risk risk for accelerated migration flows to urban for accelerated migration flows to urban More heat extremes expected to increase areas and social conflict. areas and social conflict. thermal discomfort, posing risk to labor productivity and health. Data sources: Center for International Earth Science Information Network, Columbia University; United Nations Food and Agriculture Programme; and Centro Internacional de Agricultura Tropical—(2005). Gridded Population of the World, Version 3 (GPWv3): Population Count Grid. Palisades, NY: NASA Socioeconomic Data and Applications Center (SEDAC). This map was reproduced by the Map Design Unit of The World Bank. The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of The World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. 4.5.1 Changing Precipitation Patterns and an where rainfall, and thus water availability, are predicted to decline Increase in Extreme Heat Pose High Risks to most strongly. Rainfall is low and water is scarce, while potential Agricultural Production and Regional Food evapotranspiration is high due to high temperatures. Increases Security in average temperatures and in the frequency of heat waves will represent an additional stressor for agricultural systems. Farmers in the MENA region have traditionally coped with harsh As agriculture employs more than one third of the MENA climatic conditions, and climate change adds to the existing chal- population and contributes 13 percent to the region’s GDP, climate lenges. Most agricultural activities in MENA take place in the change impacts have important implications for farmers’ liveli- semi-arid climate zones close to the coast or in the highlands, hoods, national economies, and food security. In addition, as MENA 155 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL countries are heavy food importers, the region also indirectly suffers Climate change will further reduce the availability of water from climate change impacts in distant food-producing regions. for irrigation. Making the most of the limited available water Some countries (e.g. Egypt, Syria, and Iraq) strongly depend, for in irrigated agriculture can contribute to achieving the goals of their water supply and food security, on precipitation in upstream improving the rural economy. Increasing water productivity in countries situated in different climatic zones. irrigated agriculture, in its broadest sense of yielding more (in Climate impacts in the region are likely to be highly unevenly physical and economic terms) with the same amount or less water, distributed, with the poor being most affected by both the direct will be important to any solution to the region’s water scarcity effects of a dwindling resource base and the market effects of problem. Indeed, it is generally accepted that the efficiency of the productivity shocks in remote agricultural regions. Yield reductions irrigation sector in the region is suboptimal. and increased food import dependency may affect everyone, but Management of the water-agriculture nexus in the MENA region urban populations are particularly vulnerable to rising food prices, still has a strong supply-side bias. In particular, large-scale and while poor farmers in rural areas are particularly vulnerable to centralized supply-side measures are seen as solutions to the grow- hunger and malnutrition as a direct consequence of yield losses. ing demand-supply gap. There is less emphasis on more diverse, The sector most vulnerable to negative climate change impacts distributed, decentralized, and small-scale solutions which would is rain-fed agriculture that represents an important source of increase flexibility, diversity, and resilience to climate change and livelihood for rural populations across MENA. Rain-fed agricul- other pressures (Folke 2003; Sowers et al. 2011). ture accounts for 70 percent of the farmed land and representing Large increases in world food prices do not automatically the main occupation and source of income for rural populations lead to increased poverty for low-income groups, as households (particularly in marginal areas). The prevalence of poverty is may adjust their consumption. Thus, even in the low-productivity extensive among highland farmers because of poor infrastructure scenario examined by Hertel et al. (2010), an increase in world and the degradation of natural resources; among small farmers in prices for staple grains by more than 30 percent raised the cost of dryland farming systems because of low rainfall and weak market living at the poverty line by just 6.3 percent. Nonetheless, relatively linkages; and among small herders in pastoral farming systems small increases of this kind would in particular affect urban poor for the same reasons (Dixon et al. 2001). households dependent on wage labor income. In the context of increasing climate change impacts, farmers in Child malnutrition levels could be affected if food production rain-fed areas with moderate poverty are at risk of extreme poverty and poor people’s livelihoods are affected, or if food prices rise without extensive off-farm income opportunities. Yet poverty in sharply. Child malnutrition levels are already high in parts of the the first place results from lack of access to resources, including Middle East and North Africa, with 18 percent of the region’s chil- inequitable sharing of water, food, and energy resources. Land dren under age five stunted (UNICEF 2013); the rates are higher management practices are often inefficient and many policies in some countries (notably Yemen). Evidence from previous food favor urban populations and the supply of cheap food instead of price shocks suggests that child malnutrition levels would be at supporting rural development (Dixon et al. 2001). risk. For example, the prevalence of stunting in children under age MENA countries have traditionally invested in irrigation as a five rose in Egypt from 23 percent in 2000 and 2005 to 29 percent way to improve the performance of the agriculture sector. Irriga- in 2008 (UNICEF 2013). This increase may be attributable to the tion was introduced to complement rainfall and to increase and global food crisis in 2007 (Marcus et al. 2011), itself partially driven stabilize production. About 25 percent of the agricultural land is by weather-related factors. irrigated, using on average 85 percent of water resources in the region while offering a substantial contribution to incomes and 4.5.2 Heat Extremes Will Pose a Significant employment. Countries throughout the region have also focused Challenge for Public Health Across the Region on infrastructural investments to increase irrigation water supply. Further mobilization of water resources will become increasingly The number of days with exceptional high temperatures is expected difficult, however, because of constraints on water availability. While to increase in several capital cities in the MENA region which can there is still some room in the region to increase water supplies cause heat-related illnesses, including heat stress, heat exhaustion, through the exploitation of both conventional and unconventional and heat stroke. This is a particular risk for people with chronic water resources (e.g. desalination), the extent of this increase will diseases, those who are overweight, pregnant women, children, be more limited than in the past. Thus, regional agriculture will the elderly, and people engaged in outdoor manual labor (Bartlett be confronted with mounting water restrictions in favor of other 2008; IPCC 2014c; Kjellstrom and McMichael 2013). Conditions sectors because of the generally relatively low net returns when for workers in the construction industry in the region are already compared to other water uses. very tough today, with workers reported to suffer from heat 156 MI DDLE E AST A ND NOR TH A FRICA exhaustion and dehydration. Measures to limit working hours of the population is currently living in urban areas in Bahrain, during peak heat periods, such as those implemented by Qatar Israel, Jordan, Kuwait, Lebanon, Qatar, Saudi Arabia, and United (Verner, 2012), may become increasingly necessary, with implica- Arab Emirates (UN-DESA 2014). As rural livelihoods are more tions for productivity. Thermal discomfort is expected to increase, affected by climate change, especially where a high proportion of in particular in southern parts of the Arabian Peninsula. Thermal the population is employed in agriculture, the urbanization trend discomfort will challenge people’s health, especially in densely will likely continue. In Algeria, for example, migrants move from populated large cities (because of urban heat island effects) that rural areas to mid-sized towns (Gubert and Nordman 2010).This already experience high maximum temperatures. Heat stress is movement is partly due to slow-onset environmental degrada- likely to be worse for the urban poor who cannot afford cool- tion, including water scarcity, soil erosion, and desertification, ing and may also lack access to electricity (Satterthwaite et al. which threatens agricultural livelihoods in rural Algeria (Wodon 2007), as in Yemen, for example, where only about 40 percent et al. 2014). of the population has access to electricity. Changes in climatic Several studies bring attention to the quality of housing in factors could also affect the incidence of malaria in the MENA urban centers and the capacity of urban infrastructure to accom- region because malaria distribution and seasonal occurrence is modate an increasing population. Migrants to urban areas often linked to temperature, elevation, humidity, and rainfall. Malaria live in marginal land with poor infrastructure, liable to flooding endemicity is low in North Africa and the Middle East, with cases or on unstable slopes, and this at great risk from extreme events. reported in Iran, Iraq, and Saudi Arabia and prevalent in Djibouti The migrants are likely to be poor, face health risks related to and Yemen. Poor children under age five are at greatest risk of low-income urban environments (e.g., overcrowding, poor water mortality from malaria and other vector-borne diseases (Costello quality, poor sanitation). Such areas are also typically at higher risk et al. 2009; WHO 2009). of crime (Black et al. 2012; Hugo 2011). In some areas, migrants also face discrimination based on their ethnicity, making it harder 4.5.3 Climate Change Might Act as a Threat for them to access services and find employment, with the risk that Multiplier for the Security Situation poor migrant children may have more limited access to education than other local children (Marcus et al. 2011). The urban poor Climate change might act as a threat multipler for the security have also been shown to be the worst hit by rising food prices; situation in the MENA region by imposing additional pressure on this poses a particular risk to the MENA region. MENA is likely already scarce resources and by reinforcing pre-existing threats to be exposed to the complex challenges of rising urbanization connected to migration following forced displacement. rates and mounting direct and indirect pressures on its resource base while simultaneously developing a higher vulnerability to 4.5.3.1 Migration global food prices. Rising temperatures and the risk of drought seriously affect rural Social ties play an important role in finding employment and livelihoods, possibly leading to increasing migration flows in the housing in migration destinations. Migrants, especially those coming future. However, whilst many people will be forced to move, oth- from rural areas, are often disadvantaged since they usually have ers will be forced to stay because they lack the financial resources less education and lower language capabilities (e.g., not speaking or social networks that facilitate mobility. This indicates that both French and Arabic in Morocco). Furthermore, many migrants climate-induced migration should be addressed not only within live in overcrowded apartments and in slums. Lack of permanent the framework of climate change, but also within other economic, access to electricity for cooling in the night can lead to loss of cultural, technological or political conditions that might foster or sleep and lower productivity during the day. limit migration (Gemenne 2011). If current migration patterns continue, the majority of migrants Evidence of migration flows as a response to climate change are likely to be men migrating without their families, at least is scarce. The EACH-FOR studies, a series of local qualitative initially. Women left behind in rural areas may thus face more household surveys investigating the motivation to migrate in intensive workloads in agriculture, domestic work, and the man- Egypt, Morocco, and Western Sahara (Afifi, 2009, Hamza et al. agement of scarce water supplies (Verner, 2012). If rural women 2009; Gila et al. 2009) appear to be among the only comprehensive taking on new roles do not have the skills to generate productive studies on migration and climate change in the MENA region. livelihoods, as a result of discriminatory social norms and limited Wodon et al. (2013) also conducted household interviews in five education opportunities, then they are at risk of falling into poverty. Arab countries on migration, climate change, and related topics. Promoting gender equity, by tackling both discriminatory norms Within the MENA region, most migration is internal. Urbaniza- and inequalities in access to resources, is thus a vital component tion is the predominant migration pattern, and more than 80 percent in climate-change adaptation strategies (Verner, 2012). 157 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL 4.5.3.2 Security serve as receiving and transit countries for Sahelian and other Climate change might act as a threat multiplier for the security Sub-Saharan African migrants who leave their home countries situation in the MENA region by placing additional pressures due to poverty, conflicts, and environmental degradation (see on already scarce resources and by reinforcing such preexisting Section 4.4.4, Migration and Security). threats as political instability, poverty, and unemployment. This Building resilience and institutional capacity is crucial in order creates the potential for social uprising and violent conflict. Due to react and adapt to climatic changes, extreme weather events, or to food imports and international migration, MENA is increas- shocks such as climate-variability-related increases in food prices. ingly vulnerable to climate change impacts in other parts of the Institutions weakened by conflicts or crises have a lower capacity continent and the world. to build resilience to climatic changes and extreme events. Further For example, the literature described how droughts in China, research is needed to investigate the strength of links between Russia, and Ukraine, and heavy rainfall in Australia and Canada, long-term climate change, instead of single climatologic hazards, combined with market mechanisms to contribute to high bread to migration and social unrest. prices in Egypt in 2011 (Lagi et al. 2011; Sternberg 2013). The studies contend that this, together with a previous food crisis in 2008, the country’s weak bread subsidy system, and an already unstable food supply situation, made the country more vulnerable to world market prices and reductions in wheat imports. Transit migration places an additional external pressure on scarce resources in MENA. The Maghreb countries in particular 158 4.6 Synthesis Table—Middle East and North Africa Table 4.10: Synthesis table of climate change impacts in MENA under different warming levels. The impacts reported in several impact studies were classified into different warming levels (see Section 6.4, Warming Level Attribution and Classification). OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Regional Warming (summer 1.5°C 2.2°C 2.5°C 4.5°C 7.5°C temperatures) 1 Parts of inland Algeria, Mean summer Libya, and large parts of temperatures of up to 8°C Egypt warm by 3°C warmer in parts of Algeria Warming over the Sahel is Warming over the Sahel is more moderate (2°C) more moderate (5°C) Heat Heat Waves Abadan, Iran: 6 days Abadan, Iran: 43 days Abadan, Iran: 82 days Abadan, Iran: 134 days Extremes Amman, Jordan: 4 days Amman, Jordan: 31 days Amman, Jordan: 62 days Amman, Jordan: 115 days Baghdad, Iraq: 8 days Baghdad, Iraq: 47 days Baghdad, Iraq: 90 days Baghdad, Iraq: 162 days Damascus, Syria: 1 day Damascus, Syria: 36 days Damascus, Syria: 71 days Damascus, Syria: Jerusalem: 7 days Jerusalem: 26 days Jerusalem: 46 days 129 days Riyadh, Saudi Arabia: Riyadh, Saudi Arabia: Riyadh, Saudi Arabia: Jerusalem: 102 days 3 days 81 days 132 days Riyadh, Saudi Arabia: Tehran, Iran: 5 days Tehran, Iran: 48 days Tehran, Iran: 92 days 202 days Tripoli, Libya: 3 days2 Tripoli, Libya: 13 days 3 Tripoli, Libya: 22 days3 Tehran, Iran: 159 days 30 and 40 days in a Tripoli, Libya: 59 days3 year with high maximum 150 and 210 days in temperatures in the a year with maximum Mediterranean Basin and temperatures in the the Sahara respectively4 Mediterranean Basin and the Sahara respectively, with hotspots in the Maghreb, Iran and Yemen Highly unusual 5% of land area 25%–33% of land 30% of land area5 Almost all summer months Heat Extremes Unprecedented Absent 2–5% of land area 5–10% of land area, only 65% of land area by Heat Extremes present in some isolated 2071–2099, 80% of land coastal regions in Egypt, area by 2100 Republic of Yemen, Djibouti, and Oman 159 Table 4.10: Continued. 160 OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Precipitation Summer and Increase (50% more rain) Increase south of 25°N6, Annual south of 25°N6, decrease decrease by up to 40%, in by 20–40%, in parts of parts of area north of 25°N. area north of 25°N. Winter Drier in the eastern part of Drier in the eastern part of the Sahel. the Sahel. Drought < 0.1 months per year < 0.5 months per year with ~1.5 months per year with 6 months per year with with moderate drought7 moderate drought7 moderate drought7 moderate drought7 Regional hotspots Increase in length of Increase in length of Increase in length of of drought exposure the longest period of the longest period of the longest period of (i.e. 7 out of 100,000 consecutive dry days by consecutive dry days by consecutive dry days by inhabitants exposed to 10 days in the Sahara 10 days in the Sahara 15 days in the Sahara drought)8 and by 5 days in the and by 15 days in the and by 20 days in the Mediterranean9 Mediterranean Mediterranean Little change on the Arab 20% increase in days Peninsula, and reduction under drought conditions in future droughts for the for the MNA region, too Maghreb states and the little data for the Arab Middle East10 Peninsula, and > 60% Number of rainy days increase for Morocco and decrease in the Middle the Middle East12 East11 Extreme Precipitation No change in extreme No change in MNA region, No change for North No change in North precipitation13 ~5% increase in Iran13 Africa, 10–20% increases Africa, 10–20% increase No change in return No change in return values in Arabian Peninsula and in Arabian Peninsula and values of baseline of baseline precipitation Iran13 Iran, and >50% increase in precipitation maxima14 maxima14 No change for North the Republic of Yemen13 10% increase in maximum Africa, 10–15 years for the 10 years for North Africa 5-day precipitation15 Arab Peninsula, and Iran.14 and the Arabian Peninsula, 10% increase in maximum 7–10 for the Islamic 5-day precipitation15 Republic of Iran and the Republic of Yemen14 ~ 20% increase in maximum 5-day precipitation15 Aridity 69% of land area is 71% of land area is hyper- 74% of land area is hyper- hyper-arid, 13% arid, arid and reduction in arid, arid (5% increase) and 7% semi-arid, and 2% semi-arid and sub-humid reduction in arid, semi-arid sub-humid16 areas16 and sub-humid areas OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Sea-level Rise Above Sea level changes < 0.1m Mediterranean Sea, Mediterranean Sea, Present (1985–2005) during the 20th century: Atlantic Ocean: 0.20m Atlantic Ocean: 0.38m rise of 1.1–1.3 mm/yr17 –0.57m; –0.96m Arabian Sea: 0.22m Arabian Sea: 0.44m –0.64m; maximum rate –1.04m in 2081–2100, of rise between 6.4mm/yr maximum of 1.24m in and 7.8mm/yr 2100; maximum rate of rise between 20mm/yr to 21.4mm/yr Desertification 65–89% of land critically Strong increase of more arid ecosystems, higher risk of sensitive in Egypt, desertification in North Africa and the Western Mashrek20 39% in Algeria, 28% 16% of Africa’s grassland in Morocco, 0–22% in become desert21 and 23% Iran, and 28% in Iraq18 of Africa’s desert become 83% of land in Oman grassland22 with decreasing biomass trend between 1986–200919 Water Runoff 17% reduction in mean daily runoff for the tributaries of 25–55% decrease Availability the Jordan River23 in eastern Anatolian reduction in discharge exceeding 15% in parts of the region mountains (headwaters Snow Decrease in runoff with 10–50% likelihood in Maghreb24 of Euphrates and Tigris 55% reduction in 77–85% reduction in rivers)25 snow water equivalent snow water equivalent reduction in discharge in Euphrates/Tigris in Euphrates/Tigris exceeding 45% in parts of headwaters25 headwaters25 the region 87% reduction in snow water equivalent in Euphrates/Tigris headwaters25 Groundwater 7% to 37% reduction in 14–44% reduction in the Recharge the West Bank26 West Bank and 21–51% reduction for temperatures above 4°C Crop Areas Crop Growing Decrease of over 8,500 Decrease of over 170,000 and Food Areas km² in viable rain-fed km² viable rain-fed Production agriculture; land over Israel, agricultural land over Israel, Lebanon, Syrian Arab Lebanon, Syrian Arab Republic, Iraq, and Iran at Republic, Iraq, and Iran at 161 a significance level of 0.9 27 a significance level of 0.927 162 Table 4.10: Continued. OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Yields Cereals 3.4% reduction in 3.8% reduction in the southwestern the southwestern Mediterranean and 4.9% Mediterranean and 10.1% reduction in the southeast reduction in the southeast Mediterranean 28 Mediterranean28 10–50% reduction in wheat yields in Tunisia29 1% reduction in wheat yields, 3% reduction in maize yields, and 5% reduction in rice yields in West Asia30 1% increase in barley and sorghum yields in West Asia30 Oil Crops 6% reduction in sunflower yields in West Asia30 Tubers 1% reduction in sugar 1.5% reduction in beet yields, 3% increase the southwestern in potatoes yields in West Mediterranean and 5.7% Asia30 reduction in the southeast 13.3% reduction in Mediterranean28 the southwestern Mediterranean and 4.3% reduction in the southeast Mediterranean28 Legumes 24% reduction in 18.5% reduction in the southwestern the southwestern and southeastern Mediterranean and 30.1% Mediterranean28 reduction in the southeast Mediterranean28 All Crops 26%–39% reduction in Algeria and Morocco33 Growing Shortening of the wheat Shortening of the wheat Shortening of the wheat Period growing period by 10 days growing period by 16 growing period by 30 days in Tunisia.34 days in Tunisia.34 in Tunisia.34 OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Coastal Infrastructure Sea-level rise of 0.2 m $5 billion37 in damages Areas* and Land would result in damages with 1.26 m sea-level rise of $16.5–50.5 billion in in Egypt; losses of 25% Alexandria, and of $1.2–2 of the Nile Delta’s land billion in Benghazi35 area, 10.5% of national 1.8 million and 1.9 million population, and 6.4% of people affected by flooding GDP with 1m sea-level in Morocco and Egypt with rise38 0.44–0.54 m sea-level 2.1 million and 3.6 million rise36 people affected from flooding in Morocco and Egypt with 1.04–1.14 m sea-level rise36 Tourism 50% of sandy beaches in 99.9% of sandy beaches the area between Saidia and 84.5% of tourism and Ras el Ma eroded39 infrastructure in Tangier lost40 70% of sandy beaches in the area between Saidia and Ras el Ma eroded40 Energy Capacity of nuclear and fossil-fueled power plants could decrease due to changes in river water temperature and in river flows. Mean number Systems of days during which electricity production is possible reduces due to the increase in incidence of droughts and extreme river low flow41 Human Vector-borne Malaria present in the No overall increase in malaria incidence44 39–62 million more people Health Diseases Middle East, with cases exposed to risk of malaria46 reported in Iran, Iraq, and Saudi Arabia. It is prevalent in Djibouti and Yemen42 Lymphatic filariasis could Lymphatic filariasis extend to include much present only in Egypt’s of the North African Nile Delta and in Yemen coastline with a 75–100% Zoonotic cutaceous probability45 leishmaniasis (ZCL) increases in Tunisia with changes in rainfall and humidity43 163 164 Table 4.10: Continued. OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Human Diarrheal Climate change was North Africa: Increase Health Disease estimated to be by 6–14%, Middle East: responsible in 2000 for 6–15%48 approximately 2.4% of worldwide cases of diarrhea47 Other Food Increase in the incidence Increase in the incidence and Water- of morbidity associated of morbidity associated borne with food and water-borne with food and water-borne Disease diseases in Beirut of diseases in Beirut of 16–28%49 35–42%49 Thermal 100 thermal discomfort Discomfort days in North Africa and eastern and southern parts of the Arabian Peninsula50 Number of emergency hospital admissions in Tel Aviv increases by 1.47% per 1°C temperature increase51 Migration & Security 40,000 to 60,000 households migrated to urban areas after droughts in the Syrian Arab Republic52 Please note that the years indicate the decade during which warming levels are exceeded with a 50 percent or greater change (generally at start of decade) in a business-as-usual scenario (RCP8.5 scenario), and not in mitigation scenarios limiting warming to these levels, or below; in the latter case, the year of exceeding would always be 2100 or not at all. Exceedance with a likely chance (>66 percent) generally occurs in the second half of the decade cited. Impacts are given for warming levels irrespective of the timeframe (i.e. if a study gives impacts for 2°C warming in 2100, then the impact is given in the 2°C column). Impacts given in the observations column do not necessarily form the baseline for future impacts. Impacts for different warming levels may originate from different studies and therefore may be based on different underlying assumptions— meaning that the impacts are not always fully comparable (e.g., crop yields may decrease more in 3°C than 4°C because underlying the impact at 3°C warming is a study that features very strong precipitation decreases. Moreover, this report did not systematically review observed impacts. It highlights important observed impacts for current warming but does not conduct any formal process to attribute impacts to climate change. * Impacts of sea-level rise on coastal land, infrastructure, and tourism might have been calculated with different warming levels than reported here but have been moved to a lower warming level in the table to be consistent with the sea-level rise reported at the top of the table. Please see Chapter 2—The Global Picture for more information on the reason for different sea-level rise projections and Section 4.4.5, Coastal Infrastructure and Tourism for details on the individual impact studies. MI DDLE E AST A ND NOR TH A FRICA Endnotes 1 Temperature projections for the Middle East and North African land area, for the multi-model mean during the months of June, July and August. 2 Klok and Tank (2009). 3 Lelieveld et al. (2013). Number of days in a year when maximum temperatures are above the 90th percentile for six consecutive days or longer relative to the baseline. Equals the WSDI index. Please note that the numbers given are multi-annual mean values (which is why values below six for the observation period are possible), whereas Lelieveld et al. (2013) derive median values for the non-zero entries only. 4 Sillmann et al. (2013b). Warm spell duration index (WSDI, WSDI, and CSDI count the number of days in a year when TX is above the 90th percentile for six consecutive days or longer relative to the 1961–1990 base period. The Sahara is the region spanning from 18N to 30N latitude and from 20W to 65E longitude. The Mediterranean Basin is the region spanning from 30N to 48N latitude and from 10W to 40E longitude. See Figure 2 in Sillmann et al. (2013b). 5 On average one of the summer months (June, July or August) each year will exceed temperatures warmer than three standard deviations beyond the 1951–1980 mean. 6 Note that these relatively large percentage changes projected for regions south of 25°N occur in a region that is very dry today. Thus, the absolute changes are small. 7 Orlowsky and Seneviratne (2013). Months with Standard Precipitation Index (12 month) in a 20-year windowed analysis <–1 which indicates moderate drought. The region studied is the Mediterranean, comprising southern Europe as well as North Africa and the Middle East. Morocco, Algeria, and Tunisia, as well as the Middle East, appear as hotspots. 8 Physical exposure to drought, a product designed by UNEP/GRID-Europe for the Global Assessment Report on Risk Reduction (GAR). It was modeled using global data. It is based on three sources: (1) a global monthly gridded precipitation dataset obtained from the Climatic Research Unit (University of East Anglia); (2) a GIS modeling of global Standardized Precipitation Index based on Brad Lyon (IRI, Columbia University) methodology. (3) a population grid for the year 2010, provided by LandScanTM Global Population Database (Oak Ridge, TN: Oak Ridge National Laboratory). Unit is expected average annual population exposed (inhabitants). This product was designed by Pascal Peduzzi (UNEP/GRID-Geneva, ISDR). Credit: GIS processing UNEP/GRID-Europe. 9 Sillmann et al. (2013b). Increase in the CMIP5 ensemble median in 2081–2100 compared to the reference period 1981–2000. Dry days are days with precipitation of less than one mm. The Sahara region is defined as a box stretching from 18N–30N latitude and 20W–65E longitude. The Mediterranean region is defined as a box stretching from 30N–48N latitude and 10W–40E longitude (only land grid points). 10 Dai (2012). 11 Lelieveld et al. (2013). 12 Prudhomme et al. (2013). Percentage change in the occurrence of days under drought conditions determined by the runoff deficit index (DI). 13 Kharin et al. (2013). Relative changes (percent) of 20-year return values of annual extremes of daily precipitation rates as simulated by CMIP5 models in 2081–2100. 14 Kharin et al. (2013). The multi-model median of return periods (or waiting times), in years, for late 20th century precipitation extremes as simulated by CMIP5 models in 2081–2100 relative to 1986–2005. Changes are statistically significant at the 5-percent level for the Arab Peninsula but are not statistically significant at the 5-percent level for North Africa. 15 Sillmann et al. (2013b). Projected changes (in percent) in annual maximum 5-day precipitation (RX5day) over the time period 2081–2100 relative to the reference period (1981–2000). 16 See Section 4.3.5. 17 Tsimplis and Baker (2000). 18 Various authors used the MEDALUS model to quantify the vulnerability against desertification in the MNA region. The MEDALUS model integrates regional information on vegetation, soil, climate, erosion, and management and quantifies the sensitivity of a region against desertification in four classes: critical, fragile, moderate, and non-sensitive. 19 Brinkmann et al. (2011). 20 Gao and Giorgi (2008). 21 Scheiter and Higgins (2009). Under the influence of fire. 22 Higgins and Scheiter (2012). Only potential vegetation is considered. 23 Samuels et al. (2010). For the period 2036–2060 relative to 1980–2004. 24 Gerten et al. 2013. 25 Bozkurt and Sen (2013). 26 Mizyed (2008). 27 Evans (2008). Using the 200 mm isohyet to indicate the limit of rain-fed agriculture. 28 Giannakopoulos et al. (2009). Southwest Mediterranean includes Tunisia, Algeria, and Morocco; southeast Mediterranean includes Jordan, Egypt, and Libya. 29 Mougou et al. (2010). A crop modeling study with stylized temperature (1.5°C) and precipitation scenarios (–10 percent). 30 Lobell et al. (2008). Median changes in crop yields. West Asia covers parts of the MENA region, and includes the Arabian Peninsula (except Kuwait, the United Arab Emirates, and Qatar), Iraq, Iran, Jordan, Turkey, Georgia, Azerbaijan, and Armenia. 31 Verner and Breisinger (2013). 32 Al-Bakri et al. (2011). A crop modeling study with stylized temperature (1°C–4°C) and precipitation scenarios (+/–20 percent). 33 Schilling et al. (2012). 34 Mougou et al. (2010). 35 Hallegatte et al. (2013). 36 Brown et al. (2011). 37 Hinkel et al. (2012). 38 Dasgupta et al. (2009). 39 Snoussi et al. (2008). 40 Snoussi et al. (2009). 41 See Section 4.5.5.3. 42 WHO and EMRO (2009); WHO (2013a). 43 Toumi et al. (2012). Incidence increases by 1.8 percent with a one mm increase in rainfall lagged by 12–14 months and by 5 percent when there is a 1-percent increase in humidity from July to September in the same epidemiologic year. 165 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL 44 Ebi (2008). Number of additional malaria cases calculated using relative risk of disease estimations from WHO Global Burden of Disease Study. The results are not directly comparable to van Lieshout et al. (2004) as they do not take into account additional population growth and as they use different emissions scenarios and a different climate model. 45 Slater and Michael (2012). 46 van Lieshout et al. (2004). Increases in population (in millions) exposed to risk of malaria for at least one month of the year including population growth. The population at risk in the B1 scenario is higher than in the B2 scenario because population projections for the WHO region EMR-B are higher for this scenario. No additional population is exposed to risk of malaria for three months or more in any of the SRES scenarios. The WHO region EMR-B includes Bahrain, Iran, Jordan, Kuwait, Lebanon, Libya, Oman, Qatar, Saudi Arabia, Syria, Tunisia, and the United Arab Emirates. 47 WHO (2002). 48 Kolstad and Johansson (2011). Increased relative risk of diarrheal disease. 49 El-Fadel et al. (2012). It was assumed in this study that the climate is evolving slowly and continuously beyond 2050 and that the increase in average yearly temperature is linear within the simulation period 2050–2095. Thus, the average increase in decadal temperatures from 2010 to 2095 was linearly interpolated. 50 Giannakopoulos et al. (2013). The number of days characterized by high thermal discomfort (humidex, an index used to express the temperature perceived by people, > 38°C). 51 Novikov et al. (2012). 52 Wodon et al. (2014). In total, 1.3 million people have been affected and 800,000 people lost their livelihoods. Please note that there has been no determination that the drought of 2010 in Syria was a climate change event. 166 Chapter 5 Europe and Central Asia The Europe and Central Asia region encompasses a wide range of geographic features ranging from the mountains to coasts in the Western Balkans and from the vast plains of Central Asia to Russia’s boreal forests. In the Western Balkans and Central Asia, heat extremes and reduced water availability become threats as temperatures rise toward 4°C. This includes earlier glacier melt in Central Asia and shifts in the timing of water flows, and a higher risk of drought in the Western Balkans, with potential declines for crop yields, urban health, and energy generation. In Macedonia, for example, yield losses are projected of up to 50 percent for maize, wheat, vegetables and grapes at 2°C warming. Flood risk is expected to increase slightly along the Danube, Sava and Tisza rivers, and a slight decrease in 100-year flood events is projected in the southern parts of the Western Balkans. At 2°C warming, methane emissions from melting permafrost could increase by 20–30 percent across Russia in the mid‐21st century. 5.1 Regional Summary GDP, poverty rates that are increasing in recent years, inequalities and relatively poor social services and public infrastructure, the The Europe and Central Asia region in this report covers 12 coun- region is highly vulnerable to climate change impacts. tries64 within Central Asia, the Western Balkans, and the Russian In climatic terms, the region displays a clear dipole: regions in Federation. The region encompasses a wide range of geographic the southwest become drier and regions in the northeast become features ranging from the mountainous and partly coastal Western wetter as the world warms toward 4°C. These warming conditions Balkans to the vast plains of Central Asia and Russia’s boreal for- lead to a high risk of drought in the west and challenges to stable ests. The region is inhabited by 226 million people; the population freshwater supplies in the east, where changes in precipitation is, however, unequally distributed, with Kazakhstan having only combine with glacial melt to affect the seasonality of river discharge. six inhabitants per square kilometer and Kosovo as many as 166 inhabitants per square kilometer. The urbanization rate is about 5.1.1 Regional Patterns of Climate Change 50 percent. The population in Russia and the Western Balkans is projected to decline slightly, while the population of Central Asia 5.1.1.1 Temperature is projected to increase sharply by 2050. Warming over Europe and Central Asia is projected to be above The region’s importance is closely related to its rich natural the global mean land warming. In a 2°C world, the multi-model resources, including gas and oil reserves as well as carbon stored mean warming by the end of the century is about 2.5°C above in the boreal forests (the extraction and maintenance of which the 1951–1980 base period. This level of warming is reached by affect worldwide climate mitigation goals). Due to the geographical mid-century and then remains constant until the end of the century exposure as well as a relatively high share of agriculture in regional in a 2°C world. In contrast, in a 4°C world, summer warming continues almost linearly until the end of the century, reaching about 8.5°C above the 1951–1980 baseline by 2100 for the region’s 64 The Europe and Central Asia region in this report includes the following countries: land area (Figure 5.1). The most pronounced warming is projected Albania, Bosnia and Herzegovina, Kazakhstan, Kosovo, the Kyrgyz Republic, the Former Yugoslav Republic of Macedonia, Montenegro, the Russian Federation, Serbia, to occur in Northern Russia in the region bordering the Barents- Tajikistan, Turkmenistan, and Uzbekistan. Kara Sea, along the Black Sea coast (including the Balkans), and 169 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 5.1: Multi-model mean temperature anomaly for RCP2.6 (2˚C world, left) and RCP8.5 (4˚C world, right) for the months of June-July-August for the Europe and Central Asia region. Temperature anomalies in degrees Celsius are averaged over the time period 2071–2099 relative to 1951–1980. in Northern China and Mongolia. In these areas, mean summer storm tracks. The increase in precipitation is far more pronounced temperatures by 2071–2099 will increase by about 3.5°C in a 2°C during the winter than during the summer. world and by about 7.5°C in a 4°C world. Despite an overall negative trend in extreme precipitation events, regional and seasonal projections for the Balkans remain 5.1.1.2 Heat Extremes inconclusive in a 2°C world. However, 20–30 percent reductions One of the clearest climate change signals is the strong increase in are projected for a 4°C world. Although projections of precipita- threshold-exceeding heat extremes65 in the region surrounding the tion for the Central Asian countries suffer from substantial model Black Sea, (and, in particular, the Balkans). Here, even in a 2°C uncertainties, the overall trend for heavy precipitation intensity is world, highly unusual heat extremes, with temperatures warmer below the global average. than three standard deviations beyond the baseline average, will Central and Eastern Siberia is one of the regions expected to occur in about 20–30 percent of summer months by 2100, and experience the strongest increase in heavy precipitation events. unprecedented heat extremes will occur between 5–10 percent of Heavy precipitation events with a 20-year return time are projected summer months. For the whole region, about 15 percent of the land to intensify by over 30 percent in this region and the return time area is projected to be affected by highly unusual heat extremes in of such extremes from the 20-year reference period (1986–2005) in a 2°C world by the end of the century, while unprecedented heat a 4°C world is projected to fall below five years by the end of the extremes will remain almost absent. In contrast, in a 4°C world, 21st century. Changes are much weaker (greater than 10 percent 85 percent of land area in the region is projected to be affected by increase in intensity and 10–15 year return times) in a 2°C world. highly unusual heat extremes; 55 percent of the area is projected to be affected by unprecedented heat extremes by 2100. Most of the 5.1.1.4 Drought and Aridity heat extremes will occur south of approximately 50°N, stretching In a 2°C world around five percent more land in the region will from the Balkans all the way to Japan. The number of tropical be affected by aridity; in a 4°C world, the land area classified as nights south of approximately 50°N is expected to increase by hyper-arid, arid, or semi-arid will increase by more than 30 per- 20–30 days in a 2°C world and by 50–60 days in a 4°C world. cent (Figure 5.2). The Western Balkans is projected to suffer from increased drought conditions. Though changes in annual 5.1.1.3 Precipitation precipitation are weak, the Balkans and the region surround- The basic concept of the “dry-getting-drier and wet-getting wetter” ing the Caspian Sea are projected to become more arid due to under climate change is a good first order estimate for Europe and warming-induced drying. Central Asia. The relative wetting of the Northeast, (i.e., Siberia) Projections for future drought also mimic the overall trend is the most pronounced signal, possibly associated with a shift in toward a wetter climate. Some projections even show a nega- tive change in drought risk for the eastern Siberia under a 4°C 65 In this report, highly unusual heat extremes refer to 3-sigma events and unprec- world. Projections for central and eastern Russia, meanwhile, are edented heat extremes to 5-sigma events (see Appendix). inconclusive. 170 E UR O PE A ND CENTRA L A S IA Figure 5.2: Multi-model mean of the percentage change in the aridity index (AI) for RCP2.6 (2˚C world ) (left) and RCP8.5 (4˚C world) (right) for the Europe and Central Asia region by 2071–2099 relative to 1951–1980. Hatched areas indicate uncertain results, with two or more out of five models disagreeing on the direction of change. Note that a negative change cor- responds to a shift to more arid conditions.66 5.1.2 Regional Sea-level Rise The principal driver behind the glacier volume and snow cover change is air temperature. Projections show approximately The countries of the ECA region considered here (excluding Rus- 50 percent (31–66 percent) glacier volume loss in Central Asia in sia) cover a relatively short stretch of coast that is affected by sea a 2°C world and approximately 67 percent (50–78 percent) gla- level rise. The sea-level rise in the region is projected to reach cier volume loss in a 4°C world. A temperature rise higher than 0.52 m on average (0.37–0.9 m) in a 4°C world from 2081–2100 1.1°C will cause the small glaciers of the Balkans (Albanian Alps above the 1986–2005 baseline, with rates of increase of 10.1 and Montenegrin Durmitor) to melt completely within decades. mm per year (5.9–19.6 mm/yr) from 2081–2100. This is slightly below the global mean. One of the most vulnerable coasts in 5.1.3.2 Water the region is the Drini-Mati River Delta in Albania. The sea level River flows in Central Asia will in general be lower during the in the Caspian Sea, that is completely isolated from the global summer months when the vegetation is present, while winter ocean, is projected to fall by 4.5 m by the end of the century due runoff may increase. Climate change in the region is likely to to increased evaporation.66 have consequences for runoff seasonality, and a shift in the peak flows from summer to spring can be expected due to earlier snow 5.1.3 Sector-based and Thematic Impacts melt. This may increase water stress in summer, in particular in unregulated catchments. The annual amount of water in rivers 5.1.3.1 Glaciers and Snow is not likely to decrease considerably, at least until the middle of The enhanced runoff from the glaciers is expected to continue the century when glacier depletion will cause a distinct decrease over the 21st century. Projections of glacier change use different in water volume of Central Asian rivers. Over the short-term, scenarios applied to different geographical regions for different enhanced glacier melt rates will provide an inflow of additional reference periods, making direct scenario comparisons rather water into the rivers, though in the more remote future, when difficult. In all projections, however, glaciers are expected to lose glaciers are shrinking, their buffer effect will disappear. This effect more than half of their volume by 2100. The loss of stored water will be more pronounced for the Amu Darya, because of its actual implies increased runoff in the coming decades, followed by a higher share of glacier melt water, than for the Syr Darya. significant shortage until the store is completely emptied. Very few scientific studies about regional impacts on water resources and river runoff levels are available for the Western Balkan countries, with most projections done on a broader Euro- 66 Some individual grid cells have noticeably different values than their direct pean level. In particular, there is a lack of area-wide hydrological neighbors. This is due to the fact that the aridity index is defined as a fraction of data, especially since the 1990s. Water availability over summer total annual precipitation divided by potential evapotranspiration (see Appendix). months In the Balkans is assumed to decrease considerably until It therefore behaves in a strongly non-linear fashion and year-to-year fluctuations can be large. As the results are averaged over a relatively small number of model the end of the century. In the northern parts of the Balkans, spring simulations, this can result in local jumps. and winter riverine flood risk can increase. Results from a global 171 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL study show severe decreases in annual discharge in the Western anticipated that some of these will be affected by such climatic Balkans of more than 45 percent in a 4°C world. changes as increased temperatures and more frequent and intense rainfall and drought events. A lack of certainty about 5.1.3.3 Agriculture the mechanisms through which climate change affects the inci- Central Asia’s agricultural sector is highly dependent on irriga- dence of diseases, however, prevents strong claims about future tion water availability, and the impact climate change will have trends. In general, however, higher temperatures correlate to an on agriculture in both Central Asia and the Western Balkans is increased occurrence of tick-borne encephalitis and mosquito significant. Changing precipitation patterns, reduced runoff in the transmitted malaria and dengue fever. Malaria is endemic in major river basins, and increasing temperatures will put additional Tajikistan; since the 1990s, it has reoccurred in Uzbekistan, pressure on available water resources (and, at the same time, the Kyrgyz Republic, and Turkmenistan. Furthermore, there is increase agricultural water demand). Prolonged periods of above evidence providing stronger indications of an increased risk of average temperatures will exacerbate heat stress of agricultural dengue in the Western Balkans. crops, leading to decreasing plant productivity. Droughts, mean- Historical observations show that increased temperatures, as while, are very likely to increase desertification in the Kyrgyz well as extreme weather events such as floods can lead to drink- Republic and Kazakhstan. ing water contamination, salmonellosis, cholera, typhoid, and • Yields. Yields for a few crops, including alfalfa, grasslands, dysentery. Evidence from Albania and Macedonia in the Western and wheat in parts of the region are projected to increase in Balkans, as well as Tajikistan and Kazakhstan in Central Asia, parts of the region. The overwhelming majority of results, show an increased vulnerability of heat related strokes and mor- however, point toward decreasing crop yields. Climate change talities. Severe floods, like those that occurred in recent years in is also likely to increase heat stress and change river runoff Serbia, as well as glacial outbursts in the mountains of Tajikistan, reducing agricultural yields in the long term. In the Western Uzbekistan and the Kyrgyz Republic, increase vulnerability to Balkans, the increasing occurrence of droughts will be a injuries and drowning. major threat to agricultural production under climate change; 5.1.3.5 Energy conversely so will the increasing appearance of extreme rain Climate change will have a strong impact on the region’s energy and flood events. sector. In Central Asia, the demand for electricity is expected • Livestock. Increasing temperatures and reduced water avail- to rise as a consequence of population growth, and current and ability will negatively impact livestock production. Pasture projected economic growth. Hydroelectricity can play a major growth and regeneration rates are expected to decline in parts role in the future energy mix of the Central Asian countries, as of Central Asia. If producers react to the changes by increasing only 8 percent of the hydropower potential of the region has livestock numbers, pastures might be at added risk from over- been developed. Changes in climate and melting of glaciers grazing and erosion. In the areas where productivity of alfalfa generally mean that the amount of water available for power and grasslands is projected to increase (e.g., in Uzbekistan), generation could increase, but the new pattern of intra-annual the indirect effect of climate change on livestock production runoff distribution means that less water will be available for might be positive. energy generation in the summer. Changes in reservoir manage- • Food Security. The rural population in Central Asia is at a ment and the need to balance water requirements for agriculture particular risk of food insecurity, and there have been recent may also have a negative impact on energy availability over the cases of a direct hunger threat. Rising food prices that might summer months. follow production declines will affect the poorest social groups Due to changes in river water temperatures and river flows, the (i.e., people who spend a large portion of their income on capacity of nuclear and fossil-fueled power plants in Southern and food). There are, however, opportunities to increase regional Eastern Europe could decrease from 6.3 percent to 19 percent in agricultural production efficiency by, for example, improving Europe from 2031–2060 compared to the production levels observed agricultural policies and institutions as well as by improv- from 1971–2000. Furthermore, due to the increased incidence of ing production infrastructure and technology. Finally, while droughts and extreme river low flows, the mean number of days access to international food markets could lead to higher food during which electricity production will be reduced by more than security and lower prices, the region is not well integrated into 90 percent is projected to increase threefold; from 0.5 days per international trade networks. year (in present days) to 1.5 days per year from 2031–2060 under 1.5˚C global warming. The challenge to meet growing energy 5.1.3.4 Human Health demands in the Western Balkans will be further intensified by a A number of diseases and adverse health conditions are already reduction in energy generation from hydropower sources as the present across Eastern Europe and Central Asia, and it is result of decreases in precipitation. 172 E UR O PE A ND CENTRA L A S IA 5.1.3.6 Security and Migration 5.1.4 Overview of Regional Development Climate change impacts will intensify in Central Asia and contrib- Narratives ute to increasing the population’s overall physical, economic, and environmental insecurity. A key vulnerability is the high exposure The development narratives build on the climate change impacts of the densely populated, agriculturally productive Fergana Valley analyzed in this report (cf., Table 5.7) and are presented in more region to catastrophic floods and mudflows as a result of glacial detail in Section 5.5. Increasing climate variability and changing lake outbursts. climate are expected to threaten agricultural and energy produc- Forecasting migration patterns is a challenge because of both tion in the region by changing the hydrological snow, and glacial the complexity of these phenomena and the low reliability of and regimes. Furthermore, climate change in interaction with vegetation significant gaps in existing datasets particularly with respect to shifts and fires threaten forest productivity and carbon storage in information on environmental problems (including disasters) and the Eurasian forests. The exposure to climatic changes in combi- environmentally induced migration. nation with the regional social vulnerability patterns could have The Western Balkans, especially those nations bordering the sea, negative consequences on key development trends. are projected to experience sea-level rise and hotter temperatures; • Water resources in Central Asia are projected to increase this is expected to result in growing numbers of people moving during the first half of the century and decline thereafter, from coastal zones to cooler mountainous zones. Migration in the amplifying the challenge of accommodating competing Western Balkans has already led to severe demographic changes, water demands for agricultural production and hydropower which coupled with an aging population is expected to lead to generation. The timing of river flows is projected to shift from further increased regional climate change sensitivity as a result summer to spring, with adverse consequences for water avail- of decreased adaptive capacity. ability in critical crop-growing periods. An intensification of the In Central Asia, the majority of the population lives in climate runoff variability is expected to increase in all river basins in the hotspot areas, with projected increase in the intensity and frequency region. The competition for water resources between key sectors of extreme events (e.g. forest fires, heat waves, floods). The rural (e.g., agriculture and energy), as well as between upstream and population is among those that is the most vulnerable, and an downstream water users, can therefore be expected to intensify. increased rural-to-urban migration could be expected. Women are Until 2030 the contribution of glacial melt water to river runoff particularly vulnerable, since they typically remain behind in the might lead to an increase in river runoff and partially offset the countryside to manage their households as men migrate to urban runoff variability. In the second half of the century, however, areas. Taking into account the urbanization trends in Central Asia, runoff generation of melt water in the mountainous parts of the vulnerability of cities to catastrophes might increase. the river basins is likely to decline substantially. An increasing population, followed by increased water and energy demand, 5.1.3.7 Forests of the Russian Federation will put an additional pressure on scarce resources. Improving Russia’s forest covers a large area with a huge amount of carbon irrigation water management and efficiency of irrigation infra- stored in the soil and vegetation. Future projections highlight structure, institutional and technical advancements in agriculture, changes in productivity (both increasing and decreasing, depending integrated transboundary river management, and new employ- on species, region, site and so forth) and vegetation composition ment opportunities outside agriculture could counterbalance the which will typically be stronger under a 4°C world than a 2°C negative impacts of these environmental changes. world. Changes in species composition toward better-adapted tree • Climate extremes in the Western Balkans pose major risks species may buffer productivity losses, but they will also lead to to agricultural systems, energy and human health. The vul- a change in the forest structure and biodiversity. nerability of the Western Balkans to climatic changes is mainly The region includes a large forest area affected by permafrost related to rain-fed agricultural production and the high share of which contains large stock of carbon and methane. In general, the population that is dependent on income from agriculture. changes in the carbon, water and energy fluxes of Russia’s forests There are, however, projections showing production increases may strongly affect local, regional, and global forest resource avail- for irrigated crops in parts of the region (for example, C4 ability, ecosystem functioning, services such as carbon storage summer crops and tubers in Serbia). Increased temperatures and biodiversity, and even feedback on the global climate system. as well as both droughts and extreme river flows could pose Substantial research gaps exist, for example, regarding the effect further challenges to energy production. Recent floods and of disturbances such as fire and insect outbreaks on vegetation landslides illustrate the threats of extreme events to human cover or carbon stocks and how climate change will change forest health and well-being. In addition, the climatic conditions in productivity under concomitant changes of growing conditions, the region are becoming increasingly suitable to dengue fever disturbance regimes, and forest management practices. and other vector-transmitted diseases such as dengue fever. 173 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL • The responses of the permafrost and the boreal forests of Russia across the region. Central Asia is less densely populated, with to climate change have consequences for timber productivity Kazakhstan (with six people per square kilometer) and Russia and global carbon stocks. Changes to carbon fluxes in response (with nine) being the least densely populated and Uzbekistan to rising temperatures, changing precipitation patterns, and inter- (with 70 people per square kilometer) and Tajikistan (with 57) actions with disturbance regimes in the forest and permafrost being the more densely populated. In the Western Balkans, the areas in the region can have far-reaching repercussions—affecting least densely populated country is Montenegro (with 46 people the global carbon stock and having an effect on albedo in the per square kilometer) and the most densely populated countries northern hemisphere. While climate change can increase the are Kosovo (with 166 people) and Albania (with 102 people per productivity of some tree species, heat waves, water stress, forest square kilometer). On average, half of the region’s population fires, and an increased incidence of tree pests and diseases could lives in urban areas (World Bank 2013b). In most countries in counterbalance any positive effects. Improving forest manage- the region population numbers have stabilized in recent years ment and sustainable wood extraction are of key importance and population projections show on average 27 percent popula- as is sustainable and far-sighted management of Russian forest tion declines for Russia and Eastern Europe (including Western ecosystems, including addressing key research gaps. Balkans) and 50 percent population increases in Central Asia by 2050 (Lutz 2010). 5.2 Introduction ECA is a region with relatively low levels of per-capita annual GDP, ranging from $800 in Tajikistan to $14,000 in Russia. Agri- 5.2.1 General Characteristics cultural production contributes an important share to the local economies, especially in Tajikistan, the Kyrgyz Republic, Uzbeki- The report covers 12 countries located in the Europe and Central stan, and Albania (World Bank 2013b)(see Table 5.1). Asia region (ECA) in three sub-regions: • Central Asia: Kazakhstan, the Kyrgyz Republic, Tajikistan, 5.2.2 Socioeconomic Profile of ECA Turkmenistan, and Uzbekistan. • Western Balkans: Albania, Bosnia and Herzegovina, Kosovo, All countries in the ECA region underwent a transition at the the Former Republic of Macedonia, Montenegro, and Serbia. end of the 20th century from various types of closed, plan-based economies to more open and free market-based ones. This transi- • Russia. tion, with the dissolution of trade networks and production shifts, The total population of the region is 226 million people, with was accompanied by a steep increase in poverty and inequality the highest share living in Russia and the fewest people living in within the region. In 1990, only 1.9 percent of the population Uzbekistan and Montenegro. The population is unevenly distributed in the ECA region was affected by poverty; this number grew to Russian Federation Bosnia and Herzegovina Kazakhstan Serbia Kosovo Uzbekistan Kyrgyz Rep. IBRD 41280 OCTOBER 2014 FYR Macedonia Turkmenistan Tajikistan This map was produced by the Map Design Unit of The World Bank. Albania The boundaries, colors, denominations and any other information Montenegro shown on this map do not imply, on the part of The World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. 174 E UR O PE A ND CENTRA L A S IA Table 5.1: Basic socioeconomic indicators in ECA countries. URBAN LIFE URBAN POPULATION AGRICULTURE, EXPECTANCY INDICATOR POPULATION POPULATION GROWTH GDP PER CAPITA VALUE ADDED1 AT BIRTH² % OF CURRENT UNIT MILLION POPULATION ANNUAL % 1000 US$ % OF GDP YEARS YEAR 2011 2012 2012 2012 2009–2010 2011 ID SP.POP.TOTL SP.URB.TOTL.IN.ZS SP.URB.GROW NY.GDP.PCAP.CD NV.AGR.TOTL.ZS SP.DYN.LE00.IN Central Asia Kazakhstan 16.5 54 1.3 11.9 4.8 69 Kyrgyz Republic 5.5 35 1.5 1.2 19.4 70 Tajikistan 7.8 27 2.7 0.8 21.2 67 Turkmenistan 5.1 49 2.0 5.5 14.5 65 Uzbekistan 29.3 36 1.6 1.7 19.3 68 Western Balkans Albania 3.1 54 2.2 4.4 19.1 77 Bosnia and Herzegovina 3.8 49 1.0 4.4 7.6 76 Kosovo 1.8 – – 3.4 12.0 70 Macedonia, FYR 2.1 59 0.3 4.6 11.5 75 Montenegro 0.6 63 0.4 6.8 9.3 75 Serbia 7.2 57 0.1 5.2 9.0 75 Russian Federation 142.9 74 0.6 14 4.0 69 1 Agriculture corresponds to ISIC divisions 1–5 and includes forestry, hunting, and fishing, as well as cultivation of crops and livestock production. Value added is the net output of a sector after adding up all outputs and subtracting intermediate inputs. It is calculated without making deductions for depreciation of fabricated assets, or depletion and degradation of natural resources. ²Life expectancy at birth indicates the number of years a newborn infant would live if prevailing patterns of mortality at the time of its birth were to stay the same throughout its life. Source: World Bank (2013b). 7 percent by 2010. The highest poverty rates are currently observed poverty, unemployment (particularly among women and youth), in the Kyrgyz Republic (38 percent in 2012), Uzbekistan (17.7 per- a shortage of adequate living conditions, poor medical care, and cent in 2010), and Albania (14.3 percent in 2010). Meanwhile, deficient water management infrastructure (Lioubimtseva and Kazakhstan (3.8 percent in 2012) has the lowest poverty rate in Henebry 2009). Adaptive capacity in the region is negatively the region. However, the number of people that are affected by affected by weak governance. A few countries in Central Asia are extreme poverty has been declining since 1999. Due to very cold among 20 percent of countries performing the worst in the World winters that require people to have a higher caloric intake (and, Governance Indicator. For example Kazakhstan, Russia, Tajikistan, hence, higher food expenditures), higher spending on clothing, and Turkmenistan are among 20 percent of countries that have the energy, and transportation, and differences in economic and social lowest ranks in the voice and accountability indicator that captures development, the regional poverty line was set at a $2.5–5 per day perceptions of the extent to which a country’s citizens are able to (World Bank 2011d). The total number of people living on less participate in selecting their government, as well as freedom of than $5 per day fell from 240 million in 1999 to 91 million in 2010. expression, freedom of association, and a free media. Moreover, There are also pronounced inequalities in the region. In the Turkmenistan, Uzbekistan, Tajikistan, and Kyrgyz Republic are Kyrgyz Republic in 2009, for example, the income share of the among the 20 percent of countries that have the lowest ranks in lowest 10 percent of the population was 2.8 percent compared the rule of law indicator that captures perceptions of the extent to to an income share of 27.79 percent for the highest 10 percent which agents have confidence in and abide by the rules of society, of the population (World Bank 2013t). Other issues that increase and in particular the quality of contract enforcement, property the vulnerability of the population in the region include rural rights, the police, and the courts, as well as the likelihood of crime 175 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL and violence (Worldwide Governance Indicators, 2014). The lack Figure 5.3: Temperature projections for the European and of good governance is one of the key challenges that will have to Central Asian region, compared to the 1951–1980 baseline for be addressed by the future mitigation and adaptation strategies the multi-model mean (thick line) and individual models (thin in the region. Strengthening local institutions and citizen engage- lines) under RCP2.6 (2°C world) and RCP8.5 (4°C world) for the ment is very important in water management and biodiversity months of JJA. protection and is relevant for the sectors such as agriculture and forestry (McEvoy et al. 2010; Otto et al. 2011). People and poverty in ECA are located along a spatial spec- trum with sparsely populated rural areas and dense urban areas at the ends. However, most poor are concentrated in smaller and medium-sized towns. In Albania, 42 percent of the popula- tion is urban but resides almost entirely in very small, small, and medium-sized towns. The share of the poor in urban areas is 31 percent, and nearly all of the poor reside in similar sized towns. Similarly, in Kazakhstan, 57 percent of the population and 43 percent of the poor live in urban areas—but very large cities house only eight percent of the population and one percent of the poor (Global Monitoring Report, 2013). Generally, urbanization can be a force that helps in the achievement of the Millennium Development Goals but the slums in the Kyrgyz Republic show The multi-model mean has been smoothed to give the climatological trend. that it can also worsen poverty. The failure of rural migrants to find better paying urban jobs forces them to live in informal settle- ments where without proof of residence they cannot have access than the Eurasian mean. Thus, in these regions, mean summer to health services (Global Monitoring Report, 2013). warming by 2071–2099 will be about 3.5°C in a 2°C world, and 5.3 Regional Patterns of Climate Change about 7.5°C in a 4°C world. The normalized warming (i.e., the warming expressed in terms 5.3.1 Projected Temperature Changes of the local year-to-year natural variability—see Section 6.1) is a useful diagnostic tool as it indicates how unusual the projected Figure 5.3 shows projected boreal summer (June, July, August or warming is compared to fluctuations experienced in the past (Cou- JJA) temperatures for the European and Central Asian land area in mou and Robinson 2013; Hansen et al. 2012; Mora and Frazier et al. a 2°C and 4°C world. Warming across the Northern Hemisphere 2013). The geographical patterns of normalized warming (see the land area is projected to be somewhat more than the global mean. lower panels in Figure 5.4) show that the southern hotspot regions In the 2°C warming scenario, the multi-model mean warming by (i.e., the Black Sea coastal region and northern China/Mongolia) the end of the century is about 2.5°C above the 1951–1980 base- experience the strongest shifts. In a 2°C world, the monthly tem- line (i.e., about 0.5°C more than the global mean) (World Bank perature distribution here shifts by 2–3 standard deviations toward 2013). This level of warming is reached by mid-century and then warmer conditions. In a 4°C world, these southerly regions see a remains constant until the end of the century. In contrast, in a shift of up to six standard deviations. Such a large shift implies that 4°C world, summer temperatures continue to increase up to and summer temperatures in these regions will move to a new climatic beyond the end of century in an almost linear trend, reaching about regime by the end of the century. The northern regions will see 8.5°C above the 1951–1980 baseline by 2100. The multi-model a less pronounced shift in normalized temperature because the mean warming for the 2071–2099 period reaches about 6.5°C standard deviation of the natural year-to-year variability is larger (see Figure 5.3) and is reduced due to one model which warms (i.e., temperatures are already naturally more variable) (Coumou far less than the others. Because the transient climate sensitivity and Robinson 2013). Nevertheless, a shift by at least 1-sigma (in of this model is lower than the others, results can therefore show the 2°C world) or 2-sigma (in the 4°C world) is projected to occur pronounced regional differences. here during the 21st century. The most pronounced warming is projected to occur in three Warming in southern Siberia during the 20th century is already distinct regions: (1) Northern Russia bordering the Barents-Kara evident, and 1990–1999 was the warmest decade in the last century. Sea, (2) the Black Sea coastal region, including the Balkans, and Average summer temperatures increased in the observed regions (3) northern China and Mongolia. In these hotspot regions (see by between 0–0.5°C from 1960 to 1999, but with a significantly Figure 5.4) summer warming is projected to be roughly 1°C higher higher increase of 1–2°C in the 1990s. Average winter temperatures 176 E UR O PE A ND CENTRA L A S IA Figure 5.4: Multi-model mean temperature anomaly for RCP2.6 (2˚C world, left) and RCP8.5 (4˚C world, right) for the months of JJA for the European and Central Asian region. Temperature anomalies in degrees Celsius (top row) are averaged over the time period 2071–2099 relative to 1951–1980, and normalized by the local standard deviation (bottom row). increased between 1–4.5°C within the 40 years analyzed (1960– Most studies focusing on Central Asia agree that the warm- 1999), with an increase greater than 2–3°C in the 1990s (Soja et al. 2007). ing trend in mean annual temperatures is less pronounced in the Precipitation patterns have also become more diverse. Whereas high altitudes than in the lower elevation plains and protected precipitation on windward slopes in the Urals and the Altai has intramontane valleys (Unger-Shayesteh et al. 2013). For the winter increased by up to 130–260 mm per year within the 40 years that months, a stronger warming trend can be detected at higher eleva- were analyzed, a drastic decrease in precipitation of 230 mm per tions of the Tien Shan mountains (Kriegel et al. 2013; Mannig et al. year between 1960 and 1999 has been noted on leeward slopes of 2013; Zhang et al. 2009a) the Sayans in the interior of Siberia (Soja et al. 2007). For the ECA region, IPCC (2001) predicted a warming of 5.3.2 Heat Extremes between 2 and 10°C in the 21st century as compared to the 20th century; warming rates are expected to be the largest in Figure 5.5 shows the projections of the percentage of boreal sum- 10,000 years. Recent climate projections for the 21st century range mer months warmer than 3-sigma and 5-sigma (see Section 6.1) from –0.1°C to 12°C mean winter temperature and from 0°C to over the ECA region from 2071–2099 for both 2°C and 4°C warm- 8°C mean summer temperature for all climate scenarios in this ing. One of the clearest signals identified is the strong increase region (IPCC 2013). All projections suggest precipitation increases in threshold-exceeding heat extremes in the region surrounding for the entire region of up to 58 percent in autumn and only up to the Black Sea, and in particularly in the Balkans. Here, even in a 35 percent in spring. While projections of local temperature and 2°C world, 20–30 percent of summer months are expected to rise precipitation changes are unevenly distributed across the region, beyond the 3-sigma threshold by the end of the century; 5-sigma both factors are projected to be more pronounced in the northern events are also expected to occur (between 5–10 percent of sum- area. Temperature projections for south-central Siberia range from mer months). The Balkan region is thereby expected to experi- 4–6°C (Gustafson et al. 2010), although winter temperatures are ence substantially higher frequencies of extreme heat events than expected to increase by up to 10°C. those projected for the ECA region as a whole for which about 177 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 5.5: Multi-model mean of the percentage of boreal summer months (JJA) in the time period 2071–2099 with temperatures greater than 3-sigma (top row) and 5-sigma (bottom row) for scenario RCP2.6 (2°C world, left) and RCP8.5 (4°C world, right) over the European and Central Asian region. 10–15 percent of the land area is projected to be affected by 3-sigma The increase in the frequency of summer months warmer events by the end of the century and 5-sigma events essentially than 3-sigma or 5-sigma, as shown in Figure 5.6, is quantitatively remaining absent in a 2°C world (see Figure 5.6). Similar to the consistent, even on the country scale, with published results ana- Balkans, Northern China/Mongolia will also see a substantial lyzing the full set of climate models (Coumou and Robinson 2013). increase in 3-sigma events (~15 percent of summer months) but The published literature also clearly indicates a strong increase in 5-sigma events are not expected to occur. The stronger increase heat extremes south of 50°N and a much more moderate increase in summer heat extremes in these two hotspots is thus consistent to the north of that latitude (Sillmann et al. 2013b). In fact, the with the broader shift in the mean of the normalized temperature decrease in frequency of cold extremes in this northern region distribution (Figure 5.3). may potentially have beneficial effects. Sillmann et al. (2013b) Compared to the 2°C world, the 4°C world will see a much more reported that, over Russia, the minimum nighttime temperatures pronounced increase in the frequency of summer months warmer in boreal winter are projected to increase by 3–4°C in a 2˚C world, than 3- and 5-sigma. Whereas in the 2°C world the increase in fre- and by 10°C in a 4˚C world. Likewise, the number of frost days in quency levels-off by mid-century, it continues in the 4°C world as the European part of Russia is expected to be reduced by between seen in projections of the mean summer warming (see Figure 5.3 approximately 25 days (2˚C world) and 65 days (4˚C world). This and Figure 5.4). The multi-model mean projects 85 percent of land reduces the cold spell duration in this region by up to seven days area to be affected by events hotter than 3-sigma and 55 percent in a 4°C world (Sillmann et al. 2013b). of land area to be affected by hotter than 5-sigma events by 2100 South of approximately 50°N, the projections of temperature (Figure 5.6). The bulk of these events occur in a widespread region extremes provide a totally different picture. Here the number of south of approximately 50°N, stretching from the Balkans all the tropical nights increases by 20–30 days in a 2˚C world and by way to Japan. Here, over the 2071–2099 period, about 80 percent of 50–60 days under a 4˚C world (Sillmann et al. 2013b). Temperatures summer months will be beyond 3-sigma, and 45–55 percent beyond experienced during the warmest 10 percent of summer nights during 5-sigma. Although the 3-sigma threshold level will become the new the 1961–1990 period are expected to occur in about 30 percent normal in regions north of 60°N (being exceeded in about half of the (2˚C world) or 90 percent (4˚C world) of summer nights by the summer months), 5-sigma heat extremes will remain largely absent. end of the century. These changes will cause a strong increase in 178 E UR O PE A ND CENTRA L A S IA Figure 5.6: Multi-model mean (thick line) and individual agreement, however, that under the high-emissions scenario (4°C models (thin lines) of the percentage of land area in the world), the Balkans, the Caucasus region, and Turkmenistan will European and Central Asian region warmer than 3-sigma (top) receive less rain, with the multi-model mean annual precipitation and 5-sigma (bottom) during boreal summer months (JJA) for dropping by about 20 percent. scenarios RCP2.6 (2˚C world) and RCP8.5 (4˚C world). 5.3.4 Extreme Precipitation and Droughts The footprint of climate change on climatological extremes in the 21st century is very different for the three sub-regions in ECA. Central Asia Despite a robust warming trend over Central Asia, no clear trend for precipitation extremes emerges from the observational record (Dai 2012; Donat, Alexander et al. 2013). While uncertainties are large, the overall trend regarding heavy precipitation intensity is below the global average (Kharin et al. 2013; Sillmann et al. 2013b). A similar picture emerges from the projections for future droughts. A moderate increase in drought risk for Central Asia is generally projected (Dai 2012; Prudhomme et al. 2013), but confidence in the projections is very low (Sillmann et al. 2013b). Although drought projections remain vague, regional water avail- ability will be strongly affected by changes in river runoff due to glacier melting (see Section 5.4.1, Water Resources). Western Balkans For the Western Balkans region, little to no increase in extreme precipitation events is projected over the 21st century (Kharin et al. 2013; Sillmann et al. 2013b) despite a global increase of about 20 percent in heavy precipitation event intensity in a 4°C world (Kharin et al. 2013). Regional models suggest, however, that the region’s complex topography may strongly influence extreme precipitation (Gao et al. 2006). Thus, despite an overall negative trend in extreme precipitation events, regional and seasonal pro- jections for this region remain inconclusive. the length of warm spells, by up to 90–150 days in a 4˚C world This picture changes for drought projections. The Western (Sillmann et al. 2013b). Balkans are robustly projected to suffer from an increase in 5.3.3 Regional Precipitation Projections drought conditions based on global analysis; this is similar to that expected for the greater Mediterranean region, as discussed Figure 5.7 shows that future changes in annual precipitation in the MENA section (Dai 2012; Orlowsky and Seneviratne 2013; exhibit a southwest-northeast dipole pattern, with regions in the Prudhomme et al. 2013; Sillmann et al. 2013b). Prudhomme et al. southwest becoming drier and regions in the northeast becoming (2013) project a 20 percent increase in the number of drought days wetter. Thus the basic concept of the “dry-getting-drier and wet- in a 4°C world. However, regionally resolved climate projections getting-wetter” under climate change is a good first order estima- suggest that the Western Balkans might be less affected, with no tor for the ECA region. The relative wetting of the northeast (i.e., significant increase in drought risk (Gao and Giorgi 2008), while Siberia) is the most pronounced signal, possibly associated with the greater Mediterranean region is considered a global hotspot a shift in storm tracks. The increase in precipitation is far more for future drought projections Dai (2012). pronounced during the winter (DJF) than during summer (JJA). The Russian Federation The multi-model mean drying signal in the southwest, includ- Within Russia, the projections of changes in future climatological ing the Balkans and the Caucasus region, is very weak (almost extremes are diverse. Central and Eastern Siberia is one of the flat) under low-emissions scenarios (2°C world), and the models regions expected to experience the strongest increase in heavy disagree about the direction of change. There is robust model 179 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 5.7: Multi-model mean of the percentage change in winter (DJF, top), summer (JJA, middle), and annual (bottom) precipitation for RCP2.6 (2˚C world, left) and RCP8.5 (4˚C world, right) for the European and Central Asian region by 2071–2099 relative to 1951–1980. Hatched areas indicate uncertainty regions with two or more out of five models disagreeing on the direction of change. precipitation events (Sillmann et al. 2013b). Heavy precipitation in the absolute amount of snow-water equivalent (Callaghan et events with a 20-year return time are projected to intensify by al. 2011; Shkol’nik et al. 2012), with far reaching consequences over 30 percent in this region, and the return time is projected to for the regional hydrological cycle. fall below five years by the end of the 21st century under a 4°C warming scenario (Kharin et al. 2013). The changes are much 5.3.5 Aridity weaker (less than 10 percent increase in intensity and 10–15 years return time) under a 2°C warming scenario (Kharin et al. 2013). Figure 5.8 shows the relative changes in the aridity index (AI), Projections for central and western Russia point in a similar which captures the long-term balance between water supply (pre- direction, although the intensification of heavy precipitation events cipitation) and demand (evapotranspiration) (see Section 6.1) for with a 20-year return time is less pronounced (between 10 and the ECA region by 2071–2099. The AI is defined as the total annual 30 percent). While an increase in extreme precipitation is projected precipitation divided by the annual potential evapotranspiration, throughout Russia for all seasons in a 4°C world, it is strongest and fundamentally determines whether ecosystems and agricul- in boreal winter (DJF). This would lead to a substantial increase tural systems are able to thrive in a certain area. A decrease in AI 180 E UR O PE A ND CENTRA L A S IA Figure 5.8: Multi-model mean of the percentage change in the annual-mean of monthly potential evapotranspiration for RCP2.6 (2˚C world, left) and RCP8.5 (4˚C world, right) for the European and Central Asian region by 2071–2099 relative to 1951–1980. Hatched areas indicate uncertain results, with two or more out of five models disagreeing on the direction of change. Figure 5.9: Multi-model mean of the percentage change in the aridity index (AI) for RCP2.6 (2˚C world, left) and RCP8.5 (4˚C world, right) for the ECA region by 2071–2099 relative to 1951–1980. Hatched areas indicate uncertain results, with two or more out of five models disagreeing on the direction of change. Note that a negative change cor- responds to a shift to more arid conditions.67 value indicates that water becomes more scarce (i.e., more arid Table 5.2: Multi-model mean of the percentage of land area conditions), with areas classified as hyper-arid, arid, semi-arid in the European and Central Asian region which is classified and sub-humid as specified in Table 5.2. 67 as Hyper-Arid, Arid, Semi-Arid and Sub-Humid for 1951–1980 The geographical patterns of the relative change in the annual and 2071–2099 for both 2˚C and 4˚C degrees warming levels. mean AI, as shown in Figure 5.9, are similar to those for precipita- 2071–2099 2071–2099 tion. Thus shifts in annual mean precipitation primarily determine 1951–1980 (RCP2.6) (RCP8.5) which regions become more or less arid. In the 4˚C warming Hyper-Arid 2.2 2.1 2.7 scenario, this drying region expands further to the east, covering Kazakhstan, Uzbekistan, and Turkmenistan. Because these regions Arid 3.7 4.1 5.1 are already drought-prone, this could have major consequences Semi-Arid 7.3 7.7 9.7 Sub-Humid 4.0 4.3 4.8 67 Some individual grid cells have noticeably different values than their direct neighbors. This is due to the fact that the aridity index is defined as a fraction of total annual precipitation divided by potential evapotranspiration (see Appendix). It therefore behaves in a strongly non-linear fashion and year-to-year fluctuations can be large. As the results are averaged over a relatively small number of model simulations, this can result in local jumps. 181 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL for water scarcity. Trends in AI over the Southeastern region (i.e., Figure 5.10: Sea-level rise projection for Drini-Mati River Delta Northern China and Southern Mongolia) are weaker, and models in Albania. disagree over the direction of change (partly due to similar uncer- tainty in projections of precipitation changes). Northern regions will see an increase in the AI (i.e., wetter conditions) as would be expected from the projected increase in annual-mean precipitation. The signal is especially strong over the Northeast (i.e., the Asian part of Russia), with a 20–30 per- cent increase forecast under RCP2.6 and a 50 percent forecast in a 4˚C world. The shift in AI shown in Figure 5.9 causes some regions to be classified in a different aridity class. In a 4˚C world, the area of land classified as hyper-arid, arid, or semi-arid will grow from about 13 percent in 1951–1980 to 17.5 percent in 2071–2099, which is an increase of more than 30 percent (see Table 5.2). In a 2˚C world, this increase in arid regions is much more limited (only about five percent larger). 5.3.6 Regional Sea-level Rise Time series for sea-level rise for the two scenarios RCP2.6 (blue, 1.5˚C world) and RCP8.5 (green, 4˚C world). Median estimates are given as full thick lines and the lower and upper bound given as shading. Full thin lines The region’s coastal area is relatively limited (if excluding the are global median sea-level rise, with dashed lines as lower and upper Russian Federation), and this section focuses in on the Western bound. Vertical and horizontal black lines indicate the reference period and reference (zero) level. Balkan coastline. This is where the Drini-Mati River Delta in Albania was identified as a vulnerable area in a recent UNDP report (Le Tissier 2013), and it serves as an example for the whole Western Balkan coastline (see Figure 5.10). This analysis projects a sea-level rise of 0.52 m (0.37–0.9 m) in a 4˚C world in 2081–2100 above century of due to increased evaporation in a warming climate the 1986–2005 baseline, with rates of rise of 10.1 mm per year (Renssen et al. 2007). (5.9–19.6 mm per year) (see Figure 5.10 and Table 5.3). This is slightly below the global mean. The Caspian Sea, which is isolated 5.4 Regional Impacts from the ocean, exhibits a completely different behavior, with variations of several meters over the past millennia (Naderi Beni 5.4.1 Water Resources et al. 2013) and a projected 4.5 m sea-level fall by the end of the 5.4.1.1 Central Asia Glaciers: Current Situation and Observed Changes Central Asian glaciers cover four percent of the Kyrgyz Republic (Tien Shan) and six percent of Tajikistan (Pamir); some glaciers Table 5.3: Sea-level rise (SLR) projection for the Drini-Mati River Delta. also exist in Kazakhstan and Uzbekistan (Figure 5.11). The total glaciered area is about 10,300 km2 in the Amu Darya basin and RCP2.6 RCP8.5 1,600 km2 in the Syr Darya basin (Arendt et al. 2012; Lutz et al. (1.5˚C WORLD) (4˚C WORLD) 2013a). This corresponds to a total volume of frozen water of about SLR in 2081–2100 0.32 (0.21, 0.54) 0.52 (0.37, 0.9) 1,000 km3—the equivalent of about 10 years of water flowing SLR in 2046–2065 0.21 (0.17, 0.32) 0.26 (0.21, 0.39) down the rivers Amu Darya and Syr Darya (Novikov et al. 2009). Figure 5.12 shows the loss of glacier area in the Altai-Sayan, Rate of SLR in 2081–2100 3.0 (–1.5, 5.8) 10.1 (5.9, 19.6) Pamir and Tien Shan in the period from the 1960s to 2008. As Rate of SLR in 2046–2065 4.6 (0.6, 7.1) 7.6 (5.3, 12.1) ice stocks decline, a reliable water resource is disappearing, and The sea-level rise (SLR) is expressed in meters above the 1986–2005 additional reservoirs and an improved water management may be baseline period, while rate of SLR refers to a linear trend over the period needed. In the Amu Darya and Syr Darya Basin, however, most of indicated in the table, in mm/yr. Numbers in parentheses refer to lower and the potential for reservoirs has already been exploited (Immerzeel upper bounds (see Section 6.2, Sea-Level Rise Projections for an explana- tion of the 1.5° world). and Bierkens 2012). This implies a high risk of water scarcity in the future once the peak in glacial melt runoff has passed. 182 E UR O PE A ND CENTRA L A S IA Figure 5.11: Upstream parts of the Amu and Syr Darya river basin (green and pale blue), the river basin (blue lines), and the glacierized fraction of each 1 km model grid cell (red shades). KYRGYZ REP. Source: Lutz et al. (2013a), Figure 1. Snow Cover: Current Situation and Observed Changes a global assessment (Christensen et al. 2007), a smaller fraction An increase in air temperature can generally change the proportion of precipitation is expected to fall as snow, as the snow line rises of precipitation falling as either rain or snow and shorten the dura- by about 150 m per 1°C of warming. The depth and duration of tion of seasonal snow cover. The impact of snowfall changes on seasonal snow cover are also expected to decrease, with a shift the Central Asian rivers is very high, since the seasonal snowmelt in the onset of snowmelt toward earlier spring. Through a reduc- is a key source of water. Snow reserves in the mountainous river tion in the snow-albedo feedback, the reduced snow coverage is basins will respond differently to an increase in air temperatures expected to affect both the melting rate and the regional climate, depending on the elevation and topography of the mountain area. thereby reinforcing the warming trend in Central Asia (Unger- While glaciers store water over decades and centuries, the seasonal Shayesteh et al. 2013). snowpack stores water mainly at an intra-annual time scale. At its maximum annual extent in late winter, the snow cover in the Projections of Glacial Volume Loss Aral Sea basins extends over major parts of the Amu Darya and The Fifth Assessment Report (AR5) of the Intergovernmental Syr Darya basins and contributes to a larger share of the mean Panel on Climate Change (Church et al. 2013) states that current annual runoff than glaciers (Ososkova et al. 2000). glacier extents are out of balance with current climatic conditions and indicates with high confidence that glaciers will continue to Projected Snow Cover Changes shrink in the future, even without further temperature increases. The IPCC refers to an expected decrease in Northern hemisphere In a 2°C warming scenario, projections show glacier volume snow cover of 25 percent under a high warming scenario (IPCC losses of about 50 percent (31–66 percent) for Central Asia, which 2013 AR5 WGI). Despite the high relevance of snow cover for the represents a mass of about 2800 Gt (Marzeion et al. 2012). water management of Central Asia, only a few studies address There are only a limited number of regional studies currently the future evolution of snow cover with respect to a warming available that address the timing and evolution of projected glacier climate, and these studies only provide general trends. Based on shrinkage and related changes in runoff in Central Asia. Siegfried 183 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 5.12: Losses of glacier area in the Altai-Sayan, Pamir, and Tien Shan. Remote sensing data analysis is from the 1960s through to 2008. Source: Hijioka et al. (2014), Chapter 24, Figure 24–3. et al. (2012) performed projections for the Syr Darya basin under net annual mass loss from glacial melt will peak in the middle 2°C warming by 2050; they projected a loss of mass of 31 ±4 per- of the 21st century. cent (50 Gt) in this region compared to 2010 (see Figure 5.13). For a 4°C world, Marzeion et al. (2012) projected that glacier However, the signal varies greatly across the different catchments mass loss in Central Asia will reach 67 percent (50–78 percent) (Siegfried et al. 2012). by the year 2100, which is a mass loss of about 3800 Gt (also Lutz et al. (2013a) investigated the model spread for projections stated in Hijioka et al. 2014). For a slightly different sub-region, of glacial retreat in the Amu and Syr Darya region and projected Radic´ et al. (2013) projected a mass loss of 75 percent, which is a retreat in glacial extent in the range of 54–65 percent for the the equivalent of about 4,300 Gt of ice. period 2007–2050 (see Figure 5.14). In a 3°C world, the glaciers of the region are projected to lose Impacts on River Flow and Riverine Floods about 57 percent (37–71 percent) of their current mass, equivalent The large transboundary rivers, the Amu Darya and Syr Darya, to 3200 Gt (Marzeion et al. 2012). Radic ´ et al. (2013) considered a are the main freshwater suppliers for the arid and semi-arid areas different sub-region of Central Asia (including Tibet but excluding of Central Asia (see Figure 5.15). The volume of water in these Altai and Sayan, and comprising 5,830 Gt of ice in total). They rivers strongly depends on conditions in the headwater catch- inferred a 55 percent loss for the period 2006–2100, corresponding ments, located in the mountains of Tien Shan and Pamir-Alay to 3150 ±900 Gt of ice. Giesen and Oerlemans (2013) obtained (Tajikistan, Kyrgyz Republic, and Afghanistan), where winter comparable results for the same sub-regions. precipitation is stored in snow and ice and released during the Bliss et al. (2014) went a step further and projected monthly spring and summer months (Aus der Beek et al. 2011; Krysanova glacier runoff through 2100 from all mountain glaciers in Central et al. 2010). As the water supply (e.g., for irrigated agriculture) in Asia. They inferred a 41 percent decrease in average annual runoff, the arid downstream areas largely relies on the rivers, changes in from 136 Gt per year to 80 Gt per year, for the reference periods the volume and seasonality of river runoff have major implication 2003–2022 and 2081–2100. The study further indicates that the for the region’s water management (Unger-Shayesteh et al. 2013). 184 E UR O PE A ND CENTRA L A S IA Figure 5.13: Map of Syr Darya catchment showing mean percentage loss of glacier ice by 2049 relative to 2010 for sub-regions. Source: Armstrong et al. (2005) as cited in Siegfried et al. (2012), Figure 8. Figure 5.14: Decrease in total glacier area in the Amu Darya and Syr Darya basins combined for 2008–2050 based on the CMIP3 (left panel) and CMIP5 (right panel) model runs for the median and extreme values of temperature and precipitation change. Source: Lutz et al. (2013a), Figure 13. 185 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 5.15: Water resources of the Aral Sea basin. K Y R G Y Z R E P. I.R. OF IRAN Source: ENVSEC and UNEP (2011), p.15. The Amu Darya River is characterized by two major high- Mountains, hence altering the hydrological regimes of the major flow periods, which often merge into a 4–5 month flood season. Central Asian rivers. More than 80 percent of the annual runoff With more than 90 percent of the total annual runoff of the of Amu Darya and Syr Darya is formed by snow and glacier melt Amu-Darya forming in the mountain catchment (Pamir-Alay), (Dukhovny and Schutter 2011). The Amu Darya has a higher annual the river runoff has a very high share of melt water (Hagg et share of glacier melt water (>20 percent), and this is the main al. 2013). The first high-flow period of the Amu Darya usually cause of the annual summer flood. The Syr Darya, meanwhile, takes place in the late spring, when snow melts in the lower is less influenced by glacier melt water. As observed in the last mountain areas and the spring rains fall. The second high-flow decades, higher surface temperatures are leading to higher glacier period is induced by melt water from snow and glaciers in the melt rates and significant glacier shrinkage; this trend is expected alpine Pamir Mountains. The Tien Shan Mountains, where the to continue into the future. headwaters of the Syr Darya River originate, have fewer glaciers By 2030, river runoff is expected either to increase slightly or than the Pamir. Hence, the Syr Darya usually has just one mainly not to change beyond the natural runoff variability, even in the snowmelt-induced flood season, which occurs in late spring case of potentially higher precipitation rates (Main Administra- (Dukhovny and Schutter 2011). tion of Hydrometeorology 2009). Model projections for 2055, for a headwater catchment (Panj) of the Amu Darya river, revealed Projected Changes in River Flows and Seasonality a seasonal shift in peak river flow rates from summer to spring The impact that climate change will have on river runoff rates in (Hagg et al. 2013). The study indicates that, in the near future, the Central Asia is unclear due to the uncertainty of future precipita- reduction in glacial area will be partly compensated by enhanced tion patterns (Davletkeldiev et al. 2009; Dukhovny and Schutter melt rates in a warmer atmosphere, leading to only slight changes 2011; Krysanova et al. 2010). In the next decades, enhanced glacier in total annual river flow. Under a 3.1°C warming scenario, by 2055 melt rates are expected to somewhat counterbalance increasing runoff will increase in spring and early summer due to an earlier evaporation rates. Climate change will most likely affect snow and intensified snowmelt. Under these conditions, the peak flow and glacier storage and melting rates in the Pamir and Tian Shan will shift from July to June, leading to a reduction in discharge in 186 E UR O PE A ND CENTRA L A S IA July and August of approximately 25 percent, which will further Figure 5.16: Climate change impact on flow of large rivers in limit water availability in the summer (Hagg et al. 2013). Central Asia. Climate change will also influence the snow regime in the mountains. This will further contribute to a shift of the spring floods to earlier periods. River flow will be lower in the vegetation period and the winter runoff may increase. Siegfried et al. (2012) found for Central Asia that climate changes are likely to affect runoff seasonality due to earlier snowmelt. Based on a model set up for 2050 for the Syr Darya, they projected a shift in the peak flows from summer to spring. This may increase water stress in the sum- mer, particularly in unregulated catchments (Siegfried et al. 2012). By the end the 21st century, climate change is expected to lead to a distinct decrease in the water volume of the Syr Darya (see Figure 5.16) and an even more distinct decrease in the Amu Darya River due to its higher share of glacier melt water (Davletkeldiev et al. 2009; Main Administration of Hydrometeorology 2009). This is due to decreasing precipitation and enhanced glacial retreat (see Figure 5.17 for an example). Glacier retreat will continue to diminish the stock of water stored in the high mountain areas as snow and ice; this will provide enhanced runoff during the next few decades, followed by a severe water shortage as the stock becomes depleted. In particular, the summer floods of the Amu Darya, highly important for irrigation, are expected to decline substantially by 2100. A further reduction in surface water flow is projected to be caused by an increase in evaporation rates, due to higher temperatures. By the end of the 21th century, runoff generation rates in the mountainous areas of Central Asia are likely to decline substantially (Main Administration of Hydrometeorology 2009). Impacts on the Aral Sea and Major Lakes More than 50 years of unsustainable water use for irrigation in arid deserts has led to a profound depletion of water resources in the Aral Sea basin, with consequences for society, the economy and nature. Source: Novikov et al. (2009). Starting in the 1960s, the Aral Sea has shrunk drastically, mainly due to water withdrawals for irrigation purposes and the construc- tion of water reservoirs along the Syr Darya and Amu Darya Rivers. Climate change is expected to affect the Aral Sea indirectly water are being released into the highland rivers. While this does not through changes in river contributions from the Amu Darya and directly cause flooding, melt water may become trapped behind the Syr Darya as well as directly through water evaporation and pre- glaciers’ terminal moraines so that water pressure builds up until the cipitation changes (Cretaux et al. 2013). natural moraine dam bursts (see also Nayar (2009) for a related discus- Projections of the future development of lake volumes and sion on the Himalayas). Glacier lake outbursts can cause catastrophic levels for Central Asia are scarce. The national communication flooding downstream. During the last 50 years, more than 70 glacial report of the Kyrgyz Republic displays modeling results for the lake outbursts have been reported in the Kyrgyz Republic alone, with water level of Lake Issyk-Kul in the Kyrgyz Republic under the the largest one in 1998 at Ikedavan lake resulting in more than 100 B2-MESSAGE scenario and indicates a decrease in the average deaths and causing damage to five villages (Slay and Hughes 2011). water level of the lake of between 5 m (2°C world) and 15 m (4°C In the southwest Pamir region (Tajikistan), meanwhile, any glacier world) (Davletkeldiev et al. 2009). lake outbursts have been observed (Mergili and Schneider 2011). The potential for outburst floods is expected to increase with Geohazards and other Water-related Impacts rising temperatures as well as with a rising number and size of Climate change is contributing to an increased risk of floods and moraine-dammed lakes (Armstrong 2010; Bolch et al. 2011; Mar- landslides in Central Asia. As glaciers retreat, large volumes of melt zeion et al. 2012). This is associated with an increased risk in 187 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 5.17: Dynamics of surface water-flow structure [in km3] by 2025, and approximately 19 percent by 2100, due to a projected for the Kyrgyz Republic (all rivers) for different temperature-rise decrease in annual precipitation and an increase in temperatures. scenarios calculated from the difference between the annual Albania’s water resources are projected to decline by between sum of atmospheric precipitation and annual evaporation; 14 percent (Chenoweth et al. 2011) and 40 percent (Dakova 2005) m-annual sum of precipitation compared to the baseline by the end of the 21st century. Results from a global study show period 1961–1990 (climate scenario B2-MESSAGE). severe decreases in annual discharge in the Western Balkans of up to 15 percent in a 2°C world and more than 45 percent in a 4°C world (Schewe et al. 2013). Climate change, through rising winter temperatures, directly affects snow accumulation and snowmelt by elevating the winter 0°C isotherm (snow line), leading to a decrease in the accumula- tion of snow in the headwater catchments. More rainfall in the winter months will increase winter runoff and decrease the snowmelt flood in spring (Islami et al. 2009). Model projections show a reduction of up to 20 days in the snow cover duration across the Balkans by 2050 and of up to 50 days Source: Davletkeldiev et al. (2009), Figure 5.4. in the Dinaric Alps (Schneider et al. 2013). Snow-fed river basins are very sensitive to climate change, as snow responds rapidly to slight variations in temperature and precipitation. In snowmelt- inhabited areas, such as the densely populated and agriculturally driven river regimes like the Sava river, climate change is therefore productive Fergana Valley region. The Fergana Valley is particularly expected to result in earlier spring floods and, in some cases, higher exposed to these geohazards because glaciers surround the valley winter runoff (Arnell and Gosling 2013; García-Ruiz et al. 2011). to the south, the east, and the north (Bernauer and Siegfried 2012; Changes in temperature and precipitation patterns can also Siegfried et al. 2012). affect the timing, frequency, and intensity of flood, droughts, and other extreme events (Dankers and Feyen 2009). Regions under 5.4.1.2 Western Balkans Mediterranean climate conditions are expected to experience longer Impacts on Water Resources low-flow periods during the summer season and a distinct reduc- The Western Balkans is currently one of the most water-rich tion in low-flow magnitudes (Arnell and Gosling 2013; Dakova regions in Europe and has relatively abundant freshwater resources. 2005; Dankers and Feyen 2009; Schneider et al. 2013). On the other Changes in water availability are especially relevant in areas where hand, projections also suggest an increase in riverine flood risk, water is a limiting factor for agriculture, industry, and livelihoods mainly in spring and winter, caused by more intense snowmelt and in general. In the Mediterranean areas, water is scarce and mainly increased rainfall in the winter months. Modeling results, at the depends on runoff from the mountainous headwaters (García-Ruiz European scale, indicate that more floods with a current 100-year et al. 2011)(see Figure 5.18). return time will occur by the end of the 21st century. Dankers and Feyen (2009) project a slight increase (less than 20 percent) in the Projected Trends in Water Resources frequency of 100-year floods for large rivers such as the Danube, Changes in the temperature and precipitation regime directly Sava, and Tisza in the northern parts of Serbia and Bosnia and affect the amount of water that reaches the soil, and eventually Herzegovina, and a slight decrease in 100-year flood events in the the magnitude and seasonality of river discharge. The available southern parts of the Balkans. For highly snowmelt-influenced rivers, studies indicate that a progressive decline in water availability, the time of greatest flood risk may shift, with peak flows occurring especially for the summer months, is expected in the near future earlier in spring or in late winter. For FYR Macedonia, with a more and will become more pronounced by the end of the 21st cen- Mediterranean climate, a slight reduction in the magnitude of peak tury (Arnell and Gosling 2013; García-Ruiz et al. 2011; Ministry flows is projected (Schneider et al. 2013). for Spatial Planning Construction and Ecology 2013; Ministry of Environment and Spatial Planning 2010). 5.4.1.3 Synthesis Schneider et al. (2013) found that under a regional warming of For the region of Central Asia, there exist several peer-reviewed ~2°C by the 2050s, the impacts of climate change on the natural studies examining the observed changes in water availability flow characteristics of most Balkan rivers will be “medium”; for (Lioubimtseva and Henebry 2009; Unger-Shayesteh et al. 2013). the rivers of southern Serbia, Kosovo, and FYR Macedonia, the Reliable model-based projections are very scarce, however, and impacts will be “severe.” Dakova (2005) projected a decrease in no multi-model approaches have yet been applied (Hagg et al. long-term annual mean runoff for Serbia of approximately 12 percent 2013; Siegfried et al. 2012). 188 E UR O PE A ND CENTRA L A S IA Figure 5.18: River water discharge in the Western Balkans. KO SO VO FYR Source: ENVSEC and UNEP (2012). River flows in Central Asia are expected to be lower in the For the Western Balkan countries, only a few scientific studies summer vegetation period while the winter runoff is projected on the regional impacts of climate change on water resources and to increase. Siegfried et al. (2012) found that, for Central Asia, river runoff levels are available. There is a lack of comprehensive climate change is likely to affect the seasonality of river runoff region-wide hydrological data (Dankers and Feyen 2009; García- due to earlier annual snowmelt. Using a climate, land-ice, and Ruiz et al. 2011; Schneider et al. 2013). The available scientific rainfall-runoff model for the Syr Darya basin, they projected that studies suggest that across the Balkans water availability over the by 2050 there will be a shift in peak river flows from summer to summer months is expected to decrease considerably by the end of spring. This may increase water stress in summer, in particular in the century. In the northern parts of the Balkans, however, spring unregulated catchments (Siegfried et al. 2012). However, the total and winter riverine flood risk is expected to increase. Results from annual river runoff is not likely to decrease considerably until at a global study show severe decreases in annual discharge in the least the middle of the century when glacier depletion is projected Western Balkans of more than 45 percent in a 4°C world. to cause a distinct decrease in the water volume of the Central Asian rivers. Over the short-term, enhanced glacier melt rates 5.4.2 Agricultural Production and Food Security will provide an inflow of additional water into rivers (Hagg et al. 2013). In the medium to long term, however, as glaciers shrink, 5.4.2.1 Central Asia this buffer effect will be reduced and will eventually disappear. Projected Impacts of Climate Change on Agriculture This effect will be more pronounced for the Amu Darya than for Sutton et al. (2013a) analyzed the potential impact of climate change the Syr Darya because of its higher share of glacier melt water. on Uzbekistan’s agricultural sector. Without implementing adaptation 189 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL measures and technological progress, yields for almost all crops are Degradation of soils and increased risk from desertification are expected to drop by as much as 20–50 percent (in comparison to more likely to occur when surface water availability decreases the 2000–2009 baseline) by 2050 in a 2°C world due to heat and (World Bank 2013w). The runoff in the Syr Darya River is expected water stress. Under a lower warming scenario (1.42°C warming), to decrease by 2–5 percent by 2050. At the same time, irrigation the declines are projected to be less pronounced, with wheat yields demand in Uzbekistan could increase by up to 16 percent by 2080, expected to decline by up to 13 percent with the exception of eastern increasing the competition for water and imposing risks on current parts of the country where yield increases of up to 13 percent are agricultural production systems. Uzbekistan’s crop yields could be possible. For cotton, yield decreases of 0–6 percent are projected. reduced by as much as 10–25 percent by 2050 (World Bank 2013x). Crops which might benefit from changing climatic conditions are By 2050, a shift in peak river flow and the increasing appearance alfalfa and grasslands. When including the effects of reduced water of extreme events (e.g., floods and droughts) are expected for the availability, yield decreases are much more pronounced. Irrigation river basins of the Syr Darya and Amu Darya (Schlüter et al. 2010). water demand is also likely to increase by up to 25 percent by the Many water management systems are not equipped to deal with middle of the century, while water availability could decline by the increasing occurrence of drought events (Schlüter et al. 2010). up to 30–40 percent during the same period (Sutton et al. 2013a). In the Fergana Valley, climate change is likely to affect the According to Sommer et al. (2013), due to the high irrigation water availability for large-scale irrigation. Siegfried et al. (2012), rates in Central Asia, agriculture is less dependent on precipitation for example, modeled changes in runoff for the year 2050. They than on surface water availability. The authors concluded that concluded that the runoff peak of the Syr Darya will shift by 30 wheat yields across all periods and scenarios increase by an aver- to 60 days, leading to water deficits in the vegetation period of age of 12 percent, this ranges from four percent to 27 percent. It is the Fergana Valley (Siegfried et al. 2012). necessary to note that their simulations did not include changes Livestock in irrigation water availability. Sommer et al. (2013) argued that The direct effects of climate change on livestock are likely to irrigation water demand does not necessarily increase under the be negative. In particular, increasing temperatures and reduced influence of climate change. Instead, they stated that yield increases water availability will put pressure on the sector. With changing are a consequence of higher winter and spring temperatures, less precipitation patterns and increasing temperatures, growth and frost damage and CO2 fertilization. regeneration of pastures for livestock grazing will decline in the In an assessment of global hotspots, Teixeira et al. (2013) came Tian-Shan and Alai valleys as well as in other regions of Central to a different conclusion concerning the influence of heat stress Asia (World Bank 2013r). Moreover, as water demand for livestock on yields; their study does not include the CO2 fertilization effect. increases with rising temperatures, this will put pressure on exist- The authors simulated the risk of heat stress for wheat, maize, ing water resources in water-scarce regions (Thornton et al. 2009). rice, and soybeans for the period 2071–2100 relative to the baseline The indirect effects of climate change on livestock may be posi- period of 1971–2000. They concluded that Central Asia, especially tive in some cases, such as in Uzbekistan, where the productivity Kazakhstan, is likely to be a major future hotspot of heat stress of alfalfa and grasslands is expected to increase under warming for wheat in a 3°C world (Teixeira et al. 2013). conditions (Sutton et al. 2013b). In Tajikistan, increased water stress due to climate change If livestock productivity decreases, producers could react by will be the main influencing factor for the agricultural sector. enlarging livestock numbers in order to maintain current levels According to the World Bank (2013u), yields could drop by up to of production. This could lead to overgrazing and degradation of 30 percent by 2100 in some parts of the country. pastures as well as to erosion, thereby increasing the negative Desertification is already a problem in Kazakhstan, affecting impact on the region’s ecosystems (Fay et al. 2010). up to 66 percent of the country’s land area. A projected tempera- ture increase of up to 4.6°C in 2085, and highly heterogonous 5.4.2.2 Western Balkans changes in precipitation patterns, could increase desertification Projected Impacts of Climate Change on Agriculture and threaten agricultural production, especially winter wheat A study by Sutton et al. (2013a) analyzed two countries in the (World Bank 2013v). Similar to the impacts in Kazakhstan, the Western Balkans: Albania and the Former Yugoslav Republic Kyrgyz Republic is likely to suffer from increasing desertification. of Macedonia. Albania regularly suffers from floods, which are The country’s arid and semi-arid deserts could spread, covering up problematic for agriculture when they delay the planting of crops to 23–49 percent of the country’s territory by 2100, in comparison or destroy harvests. Projections up to 2050 for Albania indicate to roughly 15 percent in 2000 (World Bank 2013r). that flooding events could increase in both frequency and intensity In Turkmenistan climate change is likely to impact the Amu under the influence of climate change. Yield changes in the Alba- Darya River, reducing its runoff by 10–15 percent by 2050 and put- nian agricultural sector are projected to be most severe for rain-fed ting pressure on existing irrigation systems and crop production. grapes and olives, with yield declines of up to 20 and 21 percent 190 E UR O PE A ND CENTRA L A S IA respectively compared to the baseline (2000–2009) under 1.81°C of 2012; Peyrouse, 2013). The access to international markets however warming. However, under this scenario wheat yields are projected is problematic due to complex regional trade linkages such as to increase by up to 24 percent due to an extended growing season the trade blockages, exports/imports bans, and quotas (Bravi & and more moderate temperatures in the winter (Sutton et al. 2013a). Solbrandt, 2011; Chabot & Tondel, 2011). However, according to Precipitation changes in FYR Macedonia are similar to those in Headey (2013), the Central Asian region is not a hotspot of climate Albania. Yield effects are relatively heterogonous, with yield declines change induced food insecurity, even though the region is to a of up to 50 percent for maize, wheat, vegetables and grapes under certain extent dependent on food imports. The country with the 1.62°C warming in the Mediterranean and Continental parts of the highest risk of food insecurity is Turkmenistan (Headey, 2013). country, whereas in the alpine areas of FYR Macedonia, wheat yields The various threats to food security from climate change in are projected to increase considerably by more than 50 percent Central Asia and the Western Balkans are indicated in the litera- (Sutton et al. 2013a). Low average temperatures and short grow- ture as follows: ing seasons, both of which will be influenced by climate change, • Increasing temperatures and changes in precipitation and pat- characterize the alpine region. Yield declines in grape production terns of river runoff are serious risks for agricultural production are especially problematic, as grapes are the most important cash (Meyers et al. 2012). crop in FYR Macedonia. Wheat, on the other hand, is the most • The availability of suitable arable land and water resources important crop in terms of production area (World Bank 2010b). is expected to decline simultaneously, changes in the type Giannakopoulos et al. (2009b) assessed the impact of climate and intensity of pests and diseases will occur, influencing the change on Mediterranean agriculture for the period 2030–2061 available productive and adapted crop varieties and animal with warming above 1.5°C (reference period 1961–1990) and dif- breeds (Meyers et al. 2012). ferentiated between the impacts on irrigated crops (C4 summer • The availability of irrigation water in Central Asia is crucial in crops, tubers) and rain-fed crops (legumes, C3 summer crops, order to maintain or expand agricultural production. As the and cereals). For irrigated crops, yields in Serbia are projected to available water is likely to decline, competition for the remaining increase by 3–4 percent for C4 summer crops and by 2–5 percent water resources among agriculture, industry, and people (for for tubers. For rain-fed crops, which dominate Serbian agriculture, human consumption) could increase (Hanjra and Qureshi 2010). yields are mostly projected to decline. Cereal yields in Serbia show a variation from a 2 percent decline to a 3 percent increase depend- • Heavy rainfall events and storms may occur more often, ing on the climate scenario used. When introducing adaptation increasing the risk of erosion from wind and water and lead- measures (e.g., early sowing dates, longer growing cycle cultivars), ing to the degradation and desertification of scarce, valuable the negative impacts of climate change could be reduced or even arable land (Christmann et al. 2009). reversed. However, the authors warned that the adaptation options • Sensitivity thresholds of crops might be exceeded more often they analyzed require up to 40 percent more irrigation water. with rising average temperatures and the increased risk of temperature extremes (Lioubimtseva and Henebry 2012; 5.4.2.3 Food Security Teixeira et al. 2013). Of the 65 million people living in the five Central Asian countries, • Eastern Europe and parts of the Balkan region will suffer roughly five million lack reliable access to food (Peyrouse 2013). from decreasing water availability, leading to yield reductions The Central Asian population is expected to increase significantly (Meyers et al. 2012). in the future (up to 95 million people by 2050), compounding the pressure induced by climate change on land and water resources 5.4.2.4 Synthesis (Lutz 2010). The rural population is the most exposed to food Climate change will have a significant impact on agriculture in insecurity. In Tajikistan roughly 2 million people suffer from food Central Asia and the Western Balkans. Central Asia’s agricultural insecurity, of which 800,000 were directly threatened by hunger sector is highly dependent on the availability of water for irriga- in 2009. In the Kyrgyz Republic one million people are affected tion. Changing precipitation patterns, reduced runoff in the major by food insecurity (Peyrouse, 2013). Rising food prices can have river basins, and increasing temperatures will simultaneously put severe effects on the Central Asian population, as large percent- additional pressure on available water resources and increase ages of household incomes are spent on food and many countries agricultural water demand (Schlüter et al. 2010; Siegfried et al. in the region are highly dependent on food imports. For example, 2012; World Bank 2013r; u; v). Prolonged periods of above aver- inhabitants of Tajikistan and Uzbekistan spend 80 percent of their age temperatures will exacerbate the heat stress of agricultural household incomes on food (Peyrouse, 2013). Tajikistan produces crops, leading to decreasing plant productivity (Mannig et al. 2013; only 31 percent of the nation’s food domestically. This indicates Teixeira et al. 2013). Increasingly frequent and intense droughts that Central Asian countries are exposed to fluctuations in inter- are very likely to increase desertification in the Kyrgyz Republic national food prices (Meyers, Ziolkowska, Tothova, & Goychuk, and Kazakhstan (World Bank 2013r; v). 191 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Because of the high importance of irrigation agriculture in large The livestock sector is particularly underrepresented in current parts of southern Central Asia and the discussed inefficiency of climate impact research on Central Asia and the Western Balkans. many irrigation systems, an improvement in irrigation techniques While the possible direct and indirect effects of climate change would be very helpful in reducing pressure on existing water on agriculture are discussed in numerous studies (Fay et al. 2010; resources (Lioubimtseva and Henebry 2009). Miraglia et al. 2009; Sutton et al. 2013a; Thornton et al. 2009; In the Western Balkans, the increasing occurrence of droughts UNDP 2014), there is almost no regional modeling on climate has been identified as a major threat to agricultural production change and agriculture. For example, the effects of climate change under climate change (Giannakopoulos et al. 2009b; Gocic and on livestock diseases and livestock biodiversity lack coverage in Trajkovic 2013, 2014; Kos et al. 2013; UNDP 2014). The risk of the scientific literature. More research on the impact of climate increasing droughts for this region was also cited in the latest change on agricultural and livestock productivity, as well as on IPCC publication (Kovats et al. 2014). The dominance of rain-fed food security in this region, is needed. agriculture in the Western Balkans makes the agricultural sec- tor especially vulnerable to changing precipitation patterns and 5.4.3 Energy Systems increasing temperatures. The increasing appearance of extreme rain and flood events also poses risks to agriculture in the region 5.4.3.1 Current Energy Access and Systems (Sutton et al. 2013a). Currently, the agricultural productivity of Situation in the ECA Region the Western Balkan states is comparably low—even though large The Central Asian and Balkan countries have very different energy parts of the region are suitable for agricultural production (Swin- mixes and varying climate change vulnerabilities. Some Central nen et al. 2010; Volk 2010). Recent years have seen a rise in yields, Asian countries heavily rely on hydroelectricity (e.g., close to mostly based on improved production techniques (Volk 2010), and 100 percent in Tajikistan and the Kyrgyz Republic). Other countries there is potential for further yield increases. While climate change in Central Asia, and the Balkan countries (with the exception of could pose risks to increasing agricultural productivity, the poor Albania), have access to electricity primarily via thermal electric execution of agricultural policy reforms, inefficient institutions, sources and, to a lesser extent, hydroelectricity. Table 5.4 dis- and missing infrastructure also threaten increases in productivity plays the shares of electricity production by source, and the total (Swinnen et al. 2010; Volk 2010). electrical power consumption per capita, in the different Central While some climate impact studies on agriculture are avail- Asian countries. able for the countries of Central Asia (Mannig et al. 2013; Schlüter et al. 2010; Siegfried et al. 2012; Sommer et al. 2013; Sutton et 5.4.3.2 Hydropower in Central Asia: al. 2013a), the Western Balkan countries lack coverage in recent More Conflicting Demands research (Meyers et al. 2012). The Western Balkan countries are Hydropower infrastructure plays a key role in Central Asia not at times the subjects of climate change analyses focusing on only for electricity generation but also for river flow regulation Europe or the European Union (Giannakopoulos et al. 2009b), and irrigation. Tajikistan and the Kyrgyz Republic, which are but very few studies concentrate on the regional climate impacts located upstream of the Syr Darya and Amu Darya, respectively, on agriculture (Ruml et al. 2012). produce 98.8 percent and 93.3 percent of their total electricity Table 5.4: Electricity production from hydroelectric and thermoelectric sources, including natural gas, oil, coal, and nuclear, in 2011 in the Central Asian countries. ELECTRICITY ELECTRICITY ELECTRICITY ELECTRICITY PRODUCTION FROM PRODUCTION FROM PRODUCTION POWER HYDROELECTRIC THERMOELECTRIC FROM OTHER CONSUMPTION SOURCES SOURCES SOURCES COUNTRY NAME (kWh PER CAPITA) (% OF TOTAL) (% OF TOTAL) (% OF TOTAL) Kazakhstan 4,892 9 91 0 Kyrgyz Republic 1,642 93 7 0 Tajikistan 1,714 99 1 0 Turkmenistan 2,444 0 100 0 Uzbekistan 1,626 19.5 80.5 0 Sources: World Bank (2013e, f, g, h, i, j). 192 E UR O PE A ND CENTRA L A S IA from hydropower (World Bank 2013p). By contrast, the riparian 5.4.3.3 Energy Systems in the Western Balkans downstream countries of Kazakhstan, Uzbekistan, and Turkmeni- The energy mix of the Western Balkans countries is very het- stan produce, respectively, 9.1 percent, 19.5 percent, and close erogeneous. Albania produces almost all of its electricity from to 0 percent of their electricity from hydropower (World Bank hydropower; at the other end of the scale, Kosovo only produces 2013p). Despite the relatively high reliance on hydroelectricity in a two percent of its electricity from hydroelectric sources (World few countries in the region, in total only 8 percent of the regional Bank 2013p). FYR Macedonia, Serbia, Montenegro, and Bosnia hydropower potential has been developed (Granit et al. 2010). Tak- and Herzegovina produce between 20 and 45 percent of their ing into account the steeply growing population of Central Asian electricity from hydroelectric sources. countries (World Bank 2013h) and their current and projected As many of the Balkan countries rely on thermal electricity economic growth (World Bank 2013s), demand for energy is pro- sources, the assessment of climate change impacts on this sec- jected to rise. However, the impacts of climate change on river tor is particularly relevant. These projected impacts on southern runoff and seasonality could affect hydropower generation and and eastern European countries were studied by van Vliet et al. increase conflicting water demands for hydropower and irrigation. (2012); their study took into account the effects of the changes A global study by Hamududu and Killingtveit (2012) projected in river water temperature and river flows on thermal electricity both future river runoff and the electricity produced by hydroelec- production. They found that the capacity of nuclear and fossil- tric installations at the national level, aggregated at the regional fueled power plants in Europe could face a 6–19 percent decrease level, for Africa, Asia, Europe, North and Central America, South in from 2031–2060 compared to the production levels observed America, and Australasia/Oceania. The authors estimated percent- from 1971–2000. Furthermore, due to the expected increase in age changes in hydropower generation up to 2050 under 2.3˚C the incidence of droughts and extreme river low flows, the mean global warming. For Central Asia, they found that production is number of days in which electricity production will be reduced by expected to increase by 2.29TWh, or 2.58 percent, compared to more than 90 percent is projected to increase threefold compared the 2005 production level. Other available projections show that to present levels, from 0.5 days per year (at present) to 1.5 days the potential of installed small hydropower plants is projected to per year from 2031–2060 under 1.5˚C global warming. Table 5.5 decrease by the 2050s under 2°C warming by around 13 percent summarizes the results of their projections. in Turkmenistan and 19 percent in the Kyrgyz Republic, and to It has also been projected that decreased production and power increase by nearly 7 percent in Kazakhstan (WorleyParsons 2012). generation disruption induced by lower runoff and increased air Siegfried et al. (2012) projected the impacts of climate change and water temperatures will lead to an increase in electricity prices on Syr Darya river runoff and seasonality changes for the middle (McDermott and Nilsen 2014). As the majority of the countries of the 21st century (2040–2049) under 1.8˚C of global warming. in the region are strongly dependent on thermal electric produc- Their projections took into account glacier melt, precipitation, tion, climate change is projected to increase their vulnerability by and temperature at six different locations in Central Asia: the affecting the supply of electricity to both households and industry. Fergana Valley (Uzbekistan, Kyrgyz Republic, and Tajikistan); In addition, economic development and a growing population the Toktogul Reservoir (Kyrgyz Republic); the Andijan Reser- are expected to increase energy demand, thereby putting thermal voir (Kyrgyz Republic); the Charvak reservoir (Uzbekistan); electric power plants under increasing pressure. In the absence of the Kayrakum reservoir (Tajikistan); and the Chardara reservoir adaptation measures, climate change, economic development, and (Uzbekistan). The authors concluded that the most significant population growth may together contribute to a rise in electricity consequence of climate change will be the 30–60 days seasonal- prices and increase the risk of electricity shortages in the region. ity shift in river runoff. This shift in seasonality is projected to In addition to the gradual slow-onset climate changes expected have major consequences for both upstream and downstream in the region, climate change may also increase the intensity and reservoir management, and the shift could lead to a major water frequency of extreme weather events. The number of studies demand deficit. For the Fergana Valley, where approximately assessing the impacts of extreme weather events, such as floods, 22 million people depend on irrigation for their livelihoods, the droughts and storms, on thermal electricity production plants at water demand deficit is projected to increase due to an increase the regional and global level is still very limited, however; this in evapotranspiration and changes in runoff seasonality. Accord- does not allow for a comprehensive analysis of this issue. ing to the projections, maximum demand deficits are projected Regarding hydropower generation, a study by Hamududu and to occur in the early growing season, which is the most sensitive Killingtveit (2012) found that, for Southern Europe (including the for plant growth. This increased demand deficit for irrigation Balkan countries), overall hydropower production is expected to is projected to put more pressure on hydroelectricity, which is decrease by 1.66TWh, or 1.43 percent compared to 2005 produc- likely to be accentuated by a projected population growth and tion levels. will contribute to a reduction in per capita water availability. 193 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Table 5.5: Reduction in usable capacity (expressed in KWmax) of thermal power plants in Europe. SCENARIO / PROJECTED KWMAX REDUCTION > 25% KWMAX REDUCTION > 50% KWMAX REDUCTION > 90% REDUCTION IN USABLE (IN MEAN NUMBER (IN MEAN NUMBER (IN MEAN NUMBER CAPACITY OF DAYS PER YEAR) OF DAYS PER YEAR) OF DAYS PER YEAR) Once-through or combination cooling thermal power plants 1971–2000 64 31 0.5 B1 (2031–2060)(1.4˚C global warming) 84 44 1.4 A2 (2031–2060)(1.5˚C global warming) 90 50 1.5 Recirculation (tower) cooling thermal power plants 1971–2000 14 9 0.02 B1 (2031–2060)(1.4˚C global warming) 18 10 0.09 A2 (2031–2060)(1.5˚C global warming) 19 11 0.08 Source: van Vliet et al. (2012). Despite the limitations of this study, the projections for this 5.4.4.1 Vector-Borne Diseases region are supported by a more detailed local study in Croatia. The Balkans and parts of Kazakhstan and the Kyrgyz Republic fall Pasicko et al. (2012) projected that energy generation from hydro- into the endemic zone for tick-borne encephalitis (TBE), transmit- power plants could decrease by 15–35 percent in a 4˚C world. The ted by the Ixodes genus of ticks (Lindquist and Vapalahti 2008). projected decrease in hydropower production in Croatia originates Although climate is only one factor among several that influence from the projected 35 percent reduction in future precipitation, the TBE transmission, and climate change could also disrupt the affecting the major Croatian rivers basins in the summer months conditions required for the disease transmission (Randolph and from 2080–2100 as compared to the reference period 1961–1990. Rogers 2000), the spread of TBE appears to be a real risk associ- ated with rising temperatures across the region. 5.4.3.4 Synthesis The reemergence of malaria in Tajikistan—following its near- Energy demand in the ECA region is projected to rise together eradication by the end of the 1950s in the USSR (Lioubimtseva and with population growth and economic development. Warmer Henebry 2009)—has happened in conjunction with an increase winter temperature can be expected to lead to decreased energy in mean temperatures (Ministry of Nature Protection 2003). The consumption for heating; however, this trend will be counterbal- disease is currently endemic to Tajikistan, with the risk classified anced by higher energy consumption for cooling purposes during as very high in much of the Khatlon region in the southwest of the summers. Energy generation in the ECA region will be affected country and the Sogd region in the north, and there are currently mainly by changes in the river flows and water temperatures. more than 150 days of the year suitable for malaria transmission In Central Asia hydropower generation has a potential to play a according to the Tajik meteorological agency (Ministry of Nature major role in the future energy mix, however, the new pattern of Protection 2003). Since the early 1990s, malaria has also reap- intra-annual runoff distribution will mean that there will be less peared in Uzbekistan, the Kyrgyz Republic, and Turkmenistan, water available for energy generation in summer months. In the and locally transmitted cases have also been reported in Russia Western Balkans, hydropower generation potential could decrease and Kazakhstan (Lioubimtseva and Henebry 2009). due to less precipitation in the region. Furthermore, the capacity Dengue fever and Chikungunya fever, transmitted by Aedes of nuclear and fossil-fuelled power plants in the sub-region could mosquitoes, are already present in Europe (ECDC 2013). Climatic decrease due to increased water temperature. conditions in the Balkans have become more suitable over the last two decades for one of the potential vectors of dengue and Chi- 5.4.4 Human Health kungunya, A. albopictus, also known as the Asian tiger mosquito (Caminade et al. 2012). It is currently found in most of Albania A number of diseases and health conditions are already present and Montenegro, and in northwestern areas of Serbia and Bosnia across Eastern Europe and Central Asia, some of which will be and Herzogovina. This is related to wetter and warmer condi- affected by climatic changes such as increased temperatures and tions favoring the winter survival of the mosquito. A study from more frequent and intense rainfall and drought events. 194 E UR O PE A ND CENTRA L A S IA the European Centre for Disease Prevention and Control (2012), Second National Communication, most parts of that country have points to increasing climatic suitability for A. albopictus in the seen a doubling in the frequency of heat waves and a decrease in Western Balkans with climate change. Caminade et al. (2012) also the duration of cold waves (BMU and WHO-Europe 2009). projected an increased suitability from 2030–2050 in the Balkans with about 1.5°C global warming, and an associated lengthening Flooding of the mosquito’s activity window. In the Western Balkans, northern Serbia is considered particularly vulnerable to flooding (ENVSEC and UNEP 2012). Further east, 5.4.4.2 Food- and Water-Borne Diseases severe flooding has been experienced in Tajikistan in recent years. Food-borne diseases, including salmonellosis, display a distinct Injuries and fatalities as a result of glacial outburst flooding in the seasonal pattern that has been associated with including increased mountains of Tajikistan, Uzbekistan, and the Kyrgyz Republic have temperatures, heat waves, and flooding (Semenza et al. 2012). also been observed (Novikov et al. 2009); glacial lake outburst flood- A time-series analysis of 10 European countries undertaken by ing poses a mounting danger as glaciers retreat with regional warm- Kovats et al. (2004) showed a clear relationship in most of the ing. Mudflows in Kazakhstan, meanwhile, are expected to increase countries studied between increases in ambient temperature and tenfold with a warming of 2–3˚C (BMU and WHO-Europe 2009). increases in the incidence of sporadic salmonella poisoning. While the overall incidence of salmonellosis is in fact declining in most 5.4.4.4 Synthesis European countries (ECDC 2013), it is likely that, with increased Little modeling of climate change impacts on human health in the temperatures, climate change will increase the risk of outbreaks. Western Balkans and Central Asian regions has been undertaken. In Central Asia, Novikov et al. (Novikov et al. 2009) noted A lack of certainty about the mechanisms through which climate that salmonellosis could become a greater problem due to warmer change affects the incidence of tick-and mosquito-borne diseases, temperatures and the contamination of communal water sources for example, prevents strong claims about future trends. An excep- exacerbated by either drier conditions or flooding. Contaminated tion, perhaps, is dengue fever: stronger indications of an increased water supplies are also associated with cholera, typhoid, and risk of dengue in the Western Balkans have been provided by dysentery. The reproductive rates of flies, which often play a sig- the European Centre for Disease Prevention and Control (2012) nificant role in transmitting food and water-borne diseases, may and Caminade et al. (2012). In addition, historical observations be increased at elevated temperatures, leading to higher incidence of extreme events, including glacial outburst flooding, may offer rates and longer disease seasons (Lioubimtseva and Henebry 2009). some clues as to what the increased risks of such events under In Tajikistan, for example, there is a risk of choleric reservoirs climate change could mean for human health in the region. developing in the lower reaches of the Vakhsh, Kafirnigan, and Syrdarya rivers (Ministry of Nature Protection 2003). 5.4.5 Security and Migration 5.4.4.3 Impacts of Extreme Weather Events 5.4.5.1 Climate-related Drivers of Migration One of the impacts of climate change on the socioeconomic and Heat Waves political stability of the ECA region will be manifest from the Heat waves can impact human health in direct ways (e.g., heat modification and/or intensification of migratory movements stress) and indirectly (e.g., aggravating respiratory and cardiovas- within and from the Central Asian and Western Balkans countries. cular conditions). The increased incidence and intensity of extreme Concerns associated with population trends and climate change heat events could cause the seasonality of temperature-related have so far been largely treated in separation (Lutz 2010). While mortality across continental Europe to shift from winter to sum- recent research is gradually trying to bridge this gap (e.g., Drabo mer, with fewer cold-related deaths and more heat-related ones. and Mbaye 2011; Reuveny 2007), the knowledge base remains While the decreasing trend in cold-related deaths is expected to extremely weak in terms of forecasting migratory patterns. initially cancel out the increasing trend in heat-related deaths, the Frequently cited figures estimate that, globally by 2050, the net total number of deaths is projected to increase for the period number of people forced to move primarily because of climate 2050–2100 under 3˚C global warming (Ballester et al. 2011). In change will be within a wide range of 200 million and one billion the Western Balkans, Albania and FYR Macedonia are considered (Tacoli 2009, 514). It is likely that both extreme weather events and particularly vulnerable to heat waves (ENVSEC and UNEP 2012). changes in mean temperatures, in precipitation, and in sea levels Central Asia has seen an increased incidence of heat-related will contribute to increasing levels of mobility (Islami et al. 2009; strokes and mortalities. For example, Tajikistan’s Second National Tacoli 2009). It is difficult to predict with precision, however, the Communication to the UNFCCC reported that the number of days extent to which these factors will impact population distributions with extremely hot weather has doubled since 1940 (cited in BMU and movements (Kniveton et al. 2008). and WHO-Europe 2009). In addition, according to Kazakhstan’s 195 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL According to Lutz (2010), future migration within the ECA Balkans has already led to severe demographic changes; coupled region is likely to be determined by political changes and security with the general demographic trend toward an aging population, problems in certain countries (linked, for instance, to the pres- this is expected to cause increased regional climate change sen- ence of ethnic minorities) in addition to environmental stresses; sitivity and decreased adaptive capacity as an aging population these are all factors that are very hard to predict with certitude is more sensitive to heat (EEA 2012). (Lutz 2010). The projected increase in the intensity and frequency According to the Asian Development Bank (2011), a large part of natural catastrophes (e.g., forest fires, heat waves, floods and of the region’s population already lives in areas at high risk of landslides) (Adger et al. 2014) is likely to result in population increased water stress due to climate change. Population growth movements that, in turn, could generate frictions in such politi- is another potential push factor contributing to both internal and cally sensitive countries as Albania, Bosnia and Herzegovina, and external migration. For example, the population in the Aral Sea Kosovo (Maas et al. 2010). In Central Asia, the biggest potential basin is expected to grow by about 20 million over the next 40 years threat is an increase in landslides and avalanches, especially in (about a 30 percent increase relative to today), with Uzbekistan the Fergana Valley, which would impact on regional livelihoods contributing about 50 percent and Tajikistan about 25 percent to and food security (Siegfried et al. 2012). the expected growth (UN World Population Database in Siegfried If it is assumed that the Western Balkans will exhibit similar et al. 2010). reactions to those of Southeast and Eastern Europe, the increased Population growth in climate change hotspots in the Central risk of disasters will result in decreasing economic opportunities and Asian countries indicates that, almost all of the population in the provide incentives for migration (Maas et al. 2010). The European sub-region is living in areas at risk from the impacts of climate Union could be among the primary destinations. Migration could change. By 2050, a 77.2 percent increase in the population living also take place within the region, eventually further aggravating the in hotspots is expected for Tajikistan, 55.4 percent in Uzbekistan, economic situation. In Central Asia, migratory patterns are likely 41.3 percent for Turkmenistan, and 31.3 percent for the Kyrgyz to continue to flow toward more productive and secure areas (e.g., Republic (with respect to the values for the year 2000) (see Russia, Europe, and the United States, from which the majority Table 5.6). of remittances are currently received) (Asian Development Bank A study commissioned by the International Organization 2011; Maas et al. 2010; Schubert et al. 2007). of Migration (IOM) in 2005 found that internal displacements in Central Asia amounted to about half of the total migrant population (Kniveton et al. 2008). Among internal migrants, a 5.4.5.2 Rural-Urban Migratory Patterns significant share moved due to environmental reasons (Jaeger as a Consequence of Climate Change et al 2009). Reasons for these displacements included mudslides In Central Asia today, migration plays an important role in the and landslides, floods, hazardous waste and desertification development of the region, notably through remittances (Asian (particularly around the Aral Sea) (Asian Development Bank Development Bank 2011). Of the countries in the ECA region, 2011). In Tajikistan, for example, there is a clear trend of moving Albania, Kazakhstan, and Georgia are among the main sending out of rural areas in the period 1995–2003, with most educated countries of migrants (Fay et al. 2010). Migration in the Western employable people moving to cities, especially Dushanbe (Asian Table 5.6: Central Asia: projected number of people facing multiple risks from climate change. 2000 2020 2030 2050 % CHANGE OF NATIONAL POPULATION IN POPULATION IN HOTSPOTS COUNTY HOTSPOTS (%) PEOPLE (MILLIONS) 2000–2050 (%) Kazakhstan 32.1 5.3 5.4 5.5 5.6 6.5 Kyrgyz Republic 99.9 5.0 6.0 6.3 6.6 31.3 Tajikistan 100.0 6.1 8.3 9.4 10.8 77.2 Turkmenistan 80.9 3.9 4.7 5.1 5.5 41.3 Uzbekistan 100.0 24.7 32.4 35.2 38.3 55.4 Source: Asian Development Bank (2011), page 52. 196 E UR O PE A ND CENTRA L A S IA Development Bank 2011; Khakimov and Mahmadbekov 2009). 5.4.5.4 Synthesis This is expected to be driven by worsening agricultural conditions Projected climatic changes related to temperatures and water in the southern latitudes and improving conditions in the north, availability, together with an increased risk of climate extremes, but it is unclear whether this push from southern rural areas will contribute to increased mobility in the ECA region. The exact and the pull into northern areas will translate into rural to rural migration patterns are difficult to estimate due to low data avail- migration, or will be associated with rural to urban migration ability as well as due to the fact that the decision to migrate is into cities in the north (Lutz 2010). usually an outcome of several processes that might be related to According to a study by the World Bank, ISDR and CAREC bio-physical environmental changes but might also be strongly under the Central Asia and Caucasus Disaster Risk Management influenced by the social and political context. Initiative (CAC DRMI), the urban population as a percentage of The projected increase in the intensity and frequency of extreme the total is expected to remain roughly constant for most countries events in the ECA region is likely to lead to migration. In Central until 2015, with the exception of Turkmenistan and Kazakhstan, Asia almost all of the population lives in climate change hot-spot where these figures are expected to increase (World Bank et al. areas and the population of these areas is expected to increase in the 2012). However, between 2015 and 2020 the figures for Central future. In general terms, climate change may contribute to a revival Asia are expected to start increasing again, up to a rate of around of internal migration movements in Central Asia, as well as from 6 percent a year by 2050. Therefore, taking the overall growth Central Asia to Russia. Due to the projected increased urbanization, patterns into account, the total urban population is expected to the vulnerability of cities to extreme events might rise, together with increase by 10 million by 2025 and by 27 million by 2050. Such the increasing population and the concentration of infrastructure. an increase in the number of people living in urban areas will The adverse effects of climate change will be borne by those significantly impact the levels of vulnerability associated with who are already more vulnerable: women, children and older high population concentrations, especially in case of disasters people, disabled people, urban poor, as well as those that are (World Bank et al. 2012). In fact, because urban areas have higher dependent on rain-fed agricultural production or pastoralism. population densities, more concentrated infrastructure, and are Furthermore, population movements could generate friction in key contributors to economic growth, the consequences of a cata- politically sensitive countries such as Albania, Bosnia and Her- strophic event there will generally be greater than in rural areas zegovina, and Kosovo. (World Bank et al. 2012). 5.4.6 Russia’s Forests: A Potential Tipping Point? 5.4.5.3 Other Consequences of Climate-Induced Migration The forests of Russia cover 882 million ha (FAO 2012b). They have The adverse effects of climate change will be felt most acutely a growing stock of 79,977,200,000 m³ and are of crucial relevance by those parts of the population that are already more vulner- for the regional and global timber supply, despite the fact that able owing to their gender, age, and disability. Moreover, climate only half of the annual wood increment is actually available for change is likely to compound existing food security issues and use (FAO 2012b). impact heavily upon those dependent on the agricultural economy. The boreal region in Russia is characterized by markedly Its distributional effects, therefore, are more likely to fall upon greater shifts to warmer temperatures and altered precipitation those involved in rain-fed subsistence agriculture or pastoralism patterns than the global mean. In a 4°C world, for example, local (Government of the Republic of Tajikistan 2011). temperature increases in Russian boreal forests are projected to In Tajikistan, for example, women and children, who consti- be almost twice as high as globally (see Section 5.3.1, Projected tute the majority of the country’s poor, are especially vulnerable Temperature Changes). Anthropogenic climate change is altering to the impacts of climate change as they are often charged with the Russian forest ecosystems and interacting with other changes the responsibility to secure water, food, and fuel for cooking and (e.g., the abandonment of agricultural lands). This threatens the heating. provision of such ecosystem services as carbon storage and timber According to a Climate Risk Assessment (CRA) study conducted production. Moreover, there is a risk that the boreal forest may by Camp Alatoo in the Kyrgyz Republic in 2013, the country’s female cross a tipping point and shift to an alternative state (Lenton et al. population is likely to experience higher climate change risks and 2008; Scheffer et al. 2012). vulnerabilities in several situations. Women in the Talas, Chui, Physiological responses to changes in climate depend strongly Naryn, and Issyk-Kul Oblasts were less impacted by landslides, on the limiting factors of forest growth, in particular low sum- but suffered more in cases of snowfall and other natural events, mer temperatures and nitrogen availability. Multiple interacting as well as from the impacts of climate change on crop production factors such as drought and heat stress, together with a changing (Camp Alatoo 2013). background climate, could lead to forest diseases or insect pest outbreaks—and to increased tree mortality. If, as a response to 197 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL climate change, tree mortality continues to increase more rapidly arctic tree line (Berner et al. 2013; Devi et al. 2008; MacDonald et al. than growth, and if the permafrost is melting, the carbon balance 2008). At the same time, water stress, forest fires, insect pests, and may be substantially altered. This is significant because Russian diseases have led to increased tree mortality—counteracting forest boreal ecosystems store a massive amount of carbon, especially in growth stimulation. Tchebakova et al. (2009) suggested that growth soils, permafrost regions, and wetlands (Tarnocai et al. 2009). In depends on water availability and, in the case of larch forests, on addition, Russian forests store about 26.25 billion tons of carbon other factors potentially related to permafrost dynamics and wildfires. in aboveground biomass (FAO 2012b). Sharmina et al. (2013) Recent analyses of Normalized Differenced Vegetation Index reviewed the Russian-language climate change impact literature (NDVI) data, used as a proxy for terrestrial gross primary produc- and found that the key climate change impacts expected to affect tion, explored the spatial and temporal variability of greening forest ecosystems are changes in vegetation zones, more frequent and browning patterns in the boreal zone (Beck et al. 2011; Bunn and intensive wildfires, and terrestrial CO2 fertilization. and Goetz 2006; Goetz et al. 2007; de Jong et al. 2011). Here, The sensitivity of the vast Russian forests to warming has the positive trend of seasonal photosynthetic activity is mostly significant implications for the climate system as a result of the confined to tundra ecosystems; a number of boreal forests in the biosphere-atmosphere exchange of water, carbon, and energy. continental interior showed a negative trend (especially after The two dominant feedbacks are changes in carbon cycling and 1990). Furthermore, studies of tree rings have identified complex changes in reflectance and energy exchange (albedo) that result patterns of tree growth in response to climate variability (Lloyd from the loss or gain of evergreen coniferous vegetation at high and Bunn 2007). Kharuk et al. (2006) found that the radial incre- latitudes (Betts 2000; Bonan 2008; O’Halloran et al. 2012). Russian ment of larch depends strongly on summer temperatures and the boreal ecosystems contribute strongly to the northern terrestrial amount of precipitation in both summer and winter. carbon sink, and are estimated to represent around half of the terrestrial global sink, which was estimated at 1.3 ±0.15 PgC Productivity Decline per year between 2000–2009 (Dolman et al. 2012; Schaphoff et Continuing warming may offset the benefits of an earlier spring al. 2013). This estimate takes into consideration emissions from onset and a delayed end of the growing season. Browning trends land use change. are shown to occur predominantly in the interior of the forest in Global warming has the potential to reduce the carbon sink late summer, despite positive trends in photosynthetic activity at capacity of the boreal zone (Koven et al. 2011; Schaphoff et al. the beginning of the vegetation period (Bunn and Goetz 2006). 2013). For Eurasia, climate change in interaction with vegetation Productivity declines in the boreal forests of the continental interior shifts and fires has the potential to turn the Eurasian carbon sink appear to be related to warming-induced drought stress (Barber into a source in a 4°C world (Kicklighter et al. 2014). In a 1.5°C et al. 2000; Dulamsuren et al. 2013; McDowell 2011). Browning is world, Eurasia would remain a small carbon sink (Kicklighter et associated with already-warmer areas (Berner et al. 2013). Many al. 2014). Observations suggest that increased temperatures could trees exhibited a general downward trend in basal area increment stimulate photosynthesis (Magnani et al. 2007; Myneni et al. after the mid-20th century (Berner et al. 2013). Lloyd and Bunn 1997, 2001). Warszawski et al. (2013) and Ostberg et al. (2013) (2007) pointed out that the highest frequency of browning occurs found that across a range of Global Vegetation Model and GCM in the most recent time period during which greening occurs at projections the boreal forests are at particular risk of biosphere the lowest frequency. Kharuk et al. (2013) suggest that soil mois- changes, including changes in the type and distribution of vegeta- ture stress is the main factor in forest mortality in the eastern tion, carbon pool changes, and carbon and water flux changes. Kuznetzky Alatau Mountains of South Siberia. They found that In combination, these changes would alter the biosphere-climate most Siberian pine mortality was detected on steep slopes; birch interactions. and aspen trees in the same area did not, however, show drought stress. Siberian pine is an important forest species in the region 5.4.6.1 Observed Changes in Forest Productivity and its decline has great significance for forestry. Forest growth in the northern latitudes depends on a variety of cli- matic and non-climatic factors. Increased radiation and atmospheric Productivity Increase CO2 concentrations as well as temperature increases leading to a Warming has increased net primary production (NPP) north of lengthening of the growing season have stimulated forest growth 47.5° over the past decade (2000–2009) in spite of a concurrent (Berner et al. 2013; Ichii et al. 2013; Myneni et al. 1997). Berner et al. drying trend (Zhao and Running 2010). However, over the Sibe- (2013) stressed that, in addition to temperature, water availability rian forest this increase was heterogeneous with an extensive and the seasonality of precipitation also affect growth. Increased negative trend over the western part (Zhao and Running 2010). A precipitation has promoted vegetation greening in some regions satellite imagery-based study by Jeong et al. (2011) observed an (Ichii et al. 2013). If water availability is sufficient, future warming earlier onset and a delayed end of the growing season in Eurasia. could promote plant growth and forest expansion along the Russian The authors estimated that from 1982–1999 the growing season 198 E UR O PE A ND CENTRA L A S IA increased by more than 0.8 days per year in accordance with a 2010 using modified forest inventory data and including remote significant warming of more than 0.25°C per year. An evaluation sensing data. Shvidenko and Nilsson (2002, 2003) found carbon of satellite remote sensing data shows positive greening trends in balance estimates for 1961–1998 based on forest inventories that the transition zone to tundra and wetlands during the summertime indicated a large carbon sink ranging from 180–322 TgC per year. growing season since the 1990s (Beck and Goetz 2012; Bunn and Dolman et al. (2012) estimated a carbon sink of 653 TgC per year Goetz 2006). Similarly, Lloyd et al. (2011) proved that warming from 1998–2008 that implies a substantial increase of the Rus- has a more positive effect in the northern sites. Greening is more sian boreal sink in the last 15 years. Tarnocai et al. (2009) have often observed in colder areas, and it was most evident in areas of reported 331 PgC in permafrost areas of Eurasia in the first meter low tree cover (Berner et al. 2013). Therefore, the positive trend in and another 162.8 PgC in peats. Schepaschenko et al. (2013) Normalized Difference Vegetation Index data might reflect enhanced estimated similar amounts of soil carbon in Russia; they dif- understory growth rather than higher tree growth (Berner et al. ferentiated Asian (189 PgC) and European (42 PgC) forest areas 2011); it might also indicate a changing allocation pattern from and excluded the tundra. The large range in reported values here woody parts to green plant material (Lapenis et al. 2005). depends on whether vegetation and soil carbon are considered or just vegetation carbon as well as whether or not disturbances 5.4.6.2 Observed Forest Cover Changes are taken into account (Balshi et al. 2007). and Vegetation Redistribution Northern latitude areas have a high potential to become carbon Russia showed the highest forest cover loss globally from 2000–2012 sources. Balshi et al. (2007) found conflicting estimates of mean (Hansen et al. 2013). Fires and subsequent recovery in Russian annual changes in carbon storage for Eurasia north of 45°N for forests also led to large gains in forest cover over the same period; the period 1996–2002. Using a process-based ecosystem model, due to the slow regrowth dynamics, however, the gains typi- they simulated either a sink of 280.2 TgC per year with CO2 fertil- cally occurred in different areas than the losses (Hansen et al. ization or a source of 29.4 TgC per year without CO2 fertilization. 2013). By analyzing Landsat remote-sensing data, Kharuk et al. Changes in permafrost dynamics, soil-vegetation carbon dynam- (2006) estimated an increase in the density of larch forests of ics, and vegetation distribution could cause long-term changes in about 65 percent and an advancement of the northern treeline by the biosphere at high latitudes. These could also have a large impact 90–300 m over the period 1973–2000. Devi et al. (2008) reported on the climate system. Permafrost thawing is most pronounced an altitudinal expansion into the formerly tree-free tundra during within the discontinuous permafrost zone but is also reported in the last century of about 20–60 m in altitude, as well as increasing the continuous permafrost zone (Romanovsky et al. 2010). tree ages and high sapling densities. At the trailing end of forest distribution, Kharuk et al. (2013) described a decline in Russian 5.4.6.4 Observed Disturbances birch stands in the southeastern Siberian forest-steppe. Disturbances play an important role in Russian forests. Fire is the Vegetation redistribution affects species composition and forest single most important forest disturbance, but a larger area is also structure and may also lead to biodiversity loss. Evidence from affected by various pests and diseases (FAO 2012b). warming experiments suggests that climate change may cause a decline in biodiversity in the tundra, as warming promotes Pests and Diseases increased height and cover of deciduous shrubs and graminoids About 13 million ha of East Siberian forest area, representing a and, consequently, a decrease in mosses and lichens (and, ulti- loss of 2 billion m³ of growing stock, are believed to have been mately, less species diversity) (Walker et al. 2006). An invasion of destroyed by the Siberian silk moth from 1880–1969 (Shvidenko southern conifers was also reported in the zone of larch dominance et al. 2013). A single outbreak in 2001 affected an area of almost (Kharuk et al. 2006). 10 million ha in a larch forest that had been thus far unaffected (Shvidenko et al. 2013). The area affected by biogenic agents in 5.4.6.3 Observed Carbon Budget, Russian forests is increasing from an average of 2.73 million ha Carbon Balance, and Permafrost during 1973–1987 to 5.48 million ha during 1998–2010 (Shvidenko The forests of Russia are of great importance for the global carbon et al. 2013). The study also pointed out that a warmer and drier cycle. The greenhouse gas inventory approach of Pan et al. (2011) climate would induce large-scale outbreaks. estimated a carbon sink of 0.26 PgC per year in Asian boreal Projected climate changes in the boreal zone could increase forests from 1990–2010 (and 0.15 PgC per year in the 1990s and the frequency and intensity of pest outbreaks. Studies of the 0.20 PgC per year overall in the last century). Myneni et al. (2001) Canadian boreal forest show that insect disturbances can turn the found a contribution of more than 40 percent to the Northern forest from a carbon sink into a carbon source (Kurz et al. 2008). carbon sink in 1995–1999. Pan et al. (2011) found that Russian It is important to note that there are considerably more studies forests stored 34.9 PgC in live biomass in 1990 and 37.5 PgC in on the effects of forest fires than on pests and diseases. What is 2007. Thurner et al. (2013) estimated about 32 PgC per year in clear, however, is that climate change will lead to northward shifts, 199 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL longer summer seasons, and warmer temperatures for the growth Figure 5.19: Dynamics of total area of wildfires in Russia’s and reproduction of forest insects (Bale et al. 2002). forests according to (1) GFDE3 (global fire emissions database); (2) refined data provided by the Institute of Forest, Fire Russian Academy of Sciences; (3) Space Research Institute, The uncertainty regarding the amount of carbon released through Russian Academy of Sciences; and (4) Vivchar et al. (2010). fires is large, and estimates from several studies compiled by Balshi et al. (2007) range from 58 TgC per year up to 520 TgC per year for boreal Russia/Siberia for different time periods within 1971–2002. A large part of the uncertainty relates to how burn severity is accounted for (Balshi et al. 2007). An estimated 59.3 percent of vegetation fires occurred in forest areas and accounted for about 82 TgC per year of emissions on average during 1998–2010 (Shvidenko et al. 2012). A later study (Shvidenko and Schepaschenko 2013) estimated the total amount of carbon burnt in 1998–2010 at 121±28 TgC per year, with 76 percent (i.e., 92±18 TgC per year) occurring on forest lands. Shvidenko and Schepaschenko (2013) also stressed that post-fire dieback is uncertain and may amount to 90–100 TgC per year. Shvidenko and Schepaschenko (2013) highlighted that the Source: Shvidenko and Schepaschenko (2013), Figure 1. types of fires have been changing, with a 1.5–2 times higher share of crown and underground fires seen of late. They found that the forest area affected by fires for 1998–2010 is 8.2 million ha per year (or 9.2 million ha per year using data from a global dataset); for of abandoned farmlands leading to declining numbers of forest 2000–2010, that number is 8.5 million ha per year. Soja et al. (2007) managers, forest firefighters, and less-efficient forest protection pointed out that 22 percent of the annual burned area is made up systems. Similarly, Flannigan et al. (2009) noted that, since the of severe crown fires with strong impacts on forest productivity breakdown of the Soviet system, the effectiveness of the Russian and carbon pools (Chertov et al. 2013), and that, in severe crown firefighting system has decreased—which has led to larger areas fires make up 50 percent of all fires in extreme fire years. Soja being burned. et al. (2007) found an increase in fire severity during the years 1998–2006, which can be connected to warmer conditions. In 5.4.6.5 Projections of Vegetation Redistribution, addition, the area burnt in the 1990s is 29 percent greater than in Forest Productivity, and Carbon Budget Changes the 1980s and 19 percent more than reported for a 47-year mean While climate change is expected to have a great impact on veg- (Soja et al. 2007). More recent analysis of both forest statistics etation distribution, changes in vegetation distribution also feed and remote sensing data reveal that, despite large variability, the back onto the climate. Enhanced warming of the dark forest (as area affected by fire seems to increase, as shown in Figure 5.19 compared to other vegetation) results in an elevated sensible heat (Shvidenko and Schepaschenko 2013). flux. Northward movement of the boreal forest, with its relatively There are important feedbacks between fire and climate. Rand- low albedo and the resulting replacement of higher albedo tundra, erson et al. (2006) found that the long-term effects of boreal forest can cause a significant increase in regional and global tempera- fires on climate warming are uncertain since positive feedbacks tures (Foley et al. 2003). This climate forcing could have an effect (enhancing warming) from increasing greenhouse gas emissions of 25.9 W per m2 (Chapin et al. 2005). Such a shift could also may be offset by changes in surface albedo (decreasing warming increase carbon storage by the same magnitude (Field et al. 2007). due to loss of canopy and more snow exposure). For the whole of the Eurasian continent, Kicklighter et al. It is important to note that forest fires are also affected by (2014) projected that biomes will shift northward as a consequence socioeconomic changes. Ivanova et al. (2010) have shown that of climate change, with boreal forest encroaching into the north- extant climate change in combination with socioeconomic changes ern tundra zone, temperate forests encroaching into the present (e.g., reduced firefighting funds) has resulted in an increase in boreal zone, and steppes encroaching into temperate forests. In fire intensity and area burned (but not fire frequency) in the Tuva a 4°C world, this would result in a reduction in the boreal forest region in southern Siberia. Moreover, Isaev and Korovin (2014) area of 19 percent and an increase in the temperate forest area of highlighted that the large forest fires that occurred in 2010 were 258 percent; in a 1.5°C world, boreal forest area would decrease due not only to unusual meteorological conditions but also to by 2 percent and temperate forest area increase by 140 percent. poor forest governance and management and an increasing area 200 E UR O PE A ND CENTRA L A S IA Figure 5.20: Vegetation distribution in Siberia in 2080 from HadCM3 A1FI (leading to a 4°C world) and B1 (leading to a 3°C world) climate change scenarios. Simulated hotspots of forest-to-steppe change (1—yellow), tundra-to-forest change (3—green), and no change in major vegetation classes (2—light gray; 0—water) in 2080. Source: Tchebakova et al. (2009). This would lead to a 7 percent net gain in forest in a 4°C world, to note that several key climate change effects, including heat and a 12 percent gain in a 1.5°C world (Kicklighter et al. 2014). stress and CO2 fertilization, were not considered in this study. Several studies using a bioclimatic model also suggest that Furthermore, in those simulations that included the effect of fire, vegetation zones will shift northward under climate change the climate-change-induced increase in carbon stock was offset (Tchebakova et al. 2009, 2011; Tchebakova and Parfenova 2012). by higher intensity fires (Shanin et al. 2011). Tchebakova et al. (2009) showed that, for Siberia, changes in Permafrost is projected to be highly vulnerable to warming, vegetation will start as early as the 2020s under all climate change and thawing is projected to be very pronounced (Koven et al. 2011; scenarios. Vegetation shifts are projected to remain moderate in Schaefer et al. 2011; Schaphoff et al. 2013)—but how exactly carbon a 3°C world, but are expected to be substantial in a 4°C world stocks will be affected is still uncertain. Koven et al. (2011) and (see Figure 5.20). Forest-steppe and steppe ecosystems are even Schaphoff et al. (2013) stressed that enhanced plant productivity predicted to become dominant across large areas of the Siberian could increase biomass input at different soil depths which can tundra (Schaphoff et al. 2006; Tchebakova et al. 2009). balance out carbon release due to permafrost thawing until the Study results from eastern Eurasia suggest that only a small late 21st century. This depends strongly, however, on the warm- range of climate change (with warming of no more than 2°C) is ing level. Schaefer et al. (2011) estimated a carbon stock loss of tolerable in order to maintain current forest structure and biomass, 190± PgC by 2200. Anisimov (2007) estimated that methane emis- (Shuman et al. 2011; Tchebakova et al. 2009; Zhang et al. 2009b, sions from melting permafrost might increase by 20–30 percent 2011). Above this level of warming, potential changes include with a global mean temperature rise of 2°C, congruent with an permafrost-thaw and changes in forest structure whereby broad- enhanced permafrost thawing rate of 10–15 percent over Russia leaved deciduous trees could increase their spread over Eastern for the mid-21st century. These fluxes mostly originate in the West Eurasia and coniferous area could decrease (Lucht et al. 2006; Siberian wetlands. Schaphoff et al. 2013; Zhang et al. 2009b). In a study of larch Anisimov and Reneva (2006) highlighted the risk to engineered forests in the region, Zhang et al. (2011a) found that such forests structures in regions affected by permafrost thawing. They pro- could not be sustained under warming of more than 2°C. jected a reduction in the permafrost area, down to 76–81 percent For a forest area in the Kostroma region 450 km northeast of the present day value by 2080. Furthermore, Schaphoff et al. of Moscow, Shanin et al. (2011) projected an increase in carbon (2013) and Schaefer et al. (2011) showed that, due to inertia in stock in trees from 125 tons per ha to 150 tons per ha in a 4°C the climate system, carbon release from permafrost thawing will world; this implies strong regional warming of 7.2°C by 2100. The continue even if warming ceases. productivity of the stands was projected to increase as well due Projections of carbon stock changes in the boreal forest eco- to the enhanced availability of nitrogen in the soil. However, soil systems under climate change are generally uncertain. Simulations and deadwood carbon stocks were projected to decrease under show that, as a result of vegetation shifts, the potential carbon this climate change scenario (98–99 tons per ha without climate gains from the expansion of boreal forests in the north are likely change vs. 33–35 tons per ha with climate change). It is important to be offset by losses in the south (Friend et al. 2014; Schaphoff 201 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL et al. 2013). Furthermore, increases in tree growth from climate moderate to high fire risk increases by up to 12 days in a 3°C warming may be limited by decreased soil fertility in northern world. Mokhov et al. (2006) projected that, under the B2 scenario and eastern regions (Lawrence et al. 2005). In addition, model (implying a moderate global warming), fire hazard will increase projections of forest ecosystem change in response to anthropo- in the southern latitudes and the northwestern areas of European genic climate changes are dominated by plant physiological CO2 Russia and decrease in much of the rest of Russia. The decrease effects (Friend et al. 2014). Moreover, the stability of ecosystems in fire hazard in their projections can be explained by increasing in response to such extreme events as flooding and drought is precipitation and minor summer warming in the climate model unpredictable (Bale et al. 2002). The interplay of disturbances and scenario used; the increasing fire risk is associated with (e.g. fire) and vegetation shifts, as well as the effects of climatic precipitation decreases. feedbacks, determine the future of the carbon stored in and the It is important to note that these studies are based on an goods and services provided by boreal forests. analysis of climatic data only and do not consider current or future forest composition and structure nor interacting disturbance 5.4.6.6 Future Impacts on Timber Harvesting regimes. Shvidenko and Schepaschenko (2013) argued that a more The FAO Forest Sector Outlook Study (FAO 2012b) does not project thorough integration of such factors might show that Russian for- any radical changes in Russian forests in the next 10–20 years due est cover changes even more strongly. Tchebakova et al. (2009) to climate change, but does highlight the potential for substantial projected an increase of an average of 10 days and 20–30 days impacts beyond 2030. Lutz et al. (2013b) used a forest gap model in the annual number of high fire danger days in a 3°C and 4°C to project that, under local warming of 2°C in 2100, larch and pine world, respectively (see Figure 5.21). They explicitly considered forests are expected to have higher productivity and subsequent fire risk within a bioclimatic vegetation model and showed that timber harvests than under the baseline climate scenario. When climate-change-induced forest-to-steppe transitions interact with changing from a 2°C to a 4°C local warming scenario, however, fire activity and promote the risks of large fires, especially in productivity levels compared to the baseline were mostly negative southern Siberia and Central Yakutia. (with the exception of larch forests in central Russia). Modeled spruce and fir forests showed small or negative responses to 2°C 5.4.6.8 Risk of a Boreal Forest Tipping Point local warming; their response to 4°C local warming was consistently Lenton et al. (2008) identified the boreal forest as a tipping element negative. The deciduous and more species-rich forests of the kind in the Earth system. They argued that, under an estimated 3–5°C found in northwestern and far eastern Russia showed a projected of global warming, water and peak summer heat stress leading decrease in productivity under 2°C local warming and increasing to tree mortality, to increased vulnerability to diseases and fire, productivity and harvests under 4°C. This counterintuitive response and to decreased reproduction rates could lead to a large-scale pattern was induced by changes in the dominant species toward forest dieback and a transition to open woodlands or grasslands. more heat tolerant species and constitutes a substantial shift in Analyzing satellite data, Scheffer et al. (2012) suggested that the forest composition and forest resources. The analysis of Lutz et only possible ecosystem state at the northern edge and at the al. (2013b) also showed that although harvests are still profitable dry continental southern edge are treeless tundra and steppe. under 4°C local warming, this is partly due to an increased harvest Their study also found a broad intermediate temperature range in the first 50 simulation years (2010–2060) that compensates for where treeless ecosystems states coexist with boreal forest (about strong declines thereafter. Moreover, the study did not include 75 percent tree cover). Tree covers of 10 percent, 30 percent, and the effects of disturbances (e.g., insects or fire). The exclusion 60 percent are relatively rare. Scheffer et al. (2012) therefore sug- of fire may explain the somewhat contradictory results regarding gest that these may represent unstable states. Such sparse tree increasing larch productivity (Lutz et al. 2013b) and decreasing cover occurs especially in continental permafrost-affected areas larch productivity (Zhang et al. 2011a). and on saturated soils. Scheffer et al. (2012) suggest that boreal forest may be less resilient than assumed (and thus potentially shift 5.4.6.7 Projections of Disturbances into a sparse woodland or treeless state) while tundra may shift Due to a lack of studies exploring the correlation between climate abruptly to a more abundant tree cover state. The mechanisms change and forest pests and diseases, this section focuses on the which could explain such unstable states are not clear, however, future risk from fire. In general, higher temperatures lead to drier and uncertainty surrounding these findings is high. fuels and hence higher fire risk (Flannigan et al. 2009). Stocks et al. (1998) projected an earlier start to and later end of the fire 5.4.6.9 Synthesis season as well as larger areas affected by higher fire danger in a Russia’s forests cover a large area and provide important ecosystem doubled CO2 scenario with 4 GCMs. Similarly, Malevsky-Malevich services. Besides supplying timber, they store huge amounts of et al. (2008) found that the area of maximum fire risk doubles carbon in soil and vegetation. The evidence for a tipping point of the by the middle of the century and that the number of days with boreal forest is unclear; already, however, under current conditions 202 E UR O PE A ND CENTRA L A S IA Figure 5.21: Modeled distributions of annual number of high fire danger days across Siberia in the current climate (a) and during the 21st century (b) for HadCM3 A1FI and B1 climate change scenarios. Fire danger days key: 0—non-forest area; 1—<30 days; 2—40 days; 3—50 days; 4—60 days; 5—70 days; 6—80 days; 7—90 days; 8—100 days; 9—110 days; 10—120 days; and 11—>120 days. Source: Tchebakova et al. (2009), Figure 4. the impacts of disturbances such as fire and pest outbreaks are Transition zones, from forest to steppe, are very vulnerable substantial—and projected climate change impacts could be both to climate change; in particular, increases in atmospheric water large-scale and disastrous. Future projections highlight changes demand could lead to water stress and higher tree mortality. in productivity, vegetation distribution, and composition that will Increased occurrences of disturbances could also affect vegetation typically be stronger in a 4°C world than in a 2°C world—often distribution in transition zones. in non-linear ways. The impacts of climate change are often overlaid with other It is important to also highlight that change in species compo- environmental and societal changes; this could exacerbate both sition toward better adapted tree species may buffer productivity existing and projected challenges. These changes may strongly losses, although they will also lead to a strong change in the forest affect local, regional, and global forest resource availability, ecosys- landscape and associated uses. Projected climate change may also tem functioning, services such as carbon storage and biodiversity induce an increase in fire danger and fire intensity. Defoliators support, and even feedback on the global climate system. and other pests and diseases could be stimulated by a warmer Russia also contains an extensive area of forested perma- and drier climate. frost. Changes here are already among the largest and they could 203 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL accelerate as a result of permafrost thawing. This has the potential reduce hydro-system storage through changes in snowpack, earlier to affect the hydrological regimes of vast territories beyond the snow melt, and glacial melt. changes in hydrology expected from precipitation changes alone, Expected glacier loss and reductions in snow pack pose serious and could affect critical carbon, water, and energy fluxes. Forest threats to freshwater resources, which rely on water storage in ice dieback and thawing of permafrost threaten to amplify global and snow. Critically, as ice and snow melt earlier during the year warming as stored carbon and methane are released into the due to rising temperatures, the timing of river flow is projected to atmosphere, giving rise to a self-amplifying feedback loop. shift within the next few decades. Peak flows are expected to shift Finally, it is important to note that substantial research gaps from summer to spring, with adverse consequences for agricultural exist, including the effect of disturbances on vegetation cover water demand during critical crop growing periods. Furthermore, and how climate change will affect forest productivity under an intensification of the runoff variability is expected in all river concomitant changes in growing conditions, disturbances, and basins, increasing the risk of floods, mudslides, and droughts (Main forest management practices. Administration of Hydrometeorology 2009). These events already have considerable social impacts—for example, economic losses from individual mudslide events have been as high as $150 mil- 5.5 Regional Development Narratives lion, while over 7000 people have migrated from landslide zones in Kyrgyz Republic alone since 1992. Flooding in Tajikistan in 2005 The report covers 12 countries located in the Europe and Central led to notable reductions in agricultural production (e.g., 70 percent Asia region (ECA) that split into three sub-regions: Central Asia, reductions in grain production and 95 percent reduction in grape Western Balkans, and Russia. The negative consequences for key production) with 71 percent of affected people stating that they development trends that may be triggered from such exposure had experienced a loss in income (Thurman 2011). to climatic changes that are described in the following develop- The impacts on water resources are distinctly different for ment narratives. It is important to note that each development the next few decades compared to the end of the century. In the narratives presents only one of the many possible ways in which coming decades, the contribution of melt water to river runoff is climate change can put key development trajectories at risk. expected to increase and may lead to an increase in river runoff— Table 5.7 summarizes the key climate change impacts under increasingly high evaporation rates, however, are expected to different warming levels in the Europe and Central Asia region counterbalance this effect (Davletkeldiev et al. 2009). By 2030, and Figure 5.22 summarizes the key sub-regional impacts. river runoff is expected either to remain unchanged or to increase slightly, even in the case of slightly higher precipitation rates (Main 5.5.1 Impacts on Water Resources in Central Administration of Hydrometeorology 2009). The picture changes Asia Increase the Challenge of Accommodating in the second half of the century: by the end of the 21st century, Competing Water Demands for Agricultural runoff generation rates in the mountainous areas of Central Asia Production and Hydropower Generation are likely to decline substantially (Main Administration of Hydro- meteorology 2009). The scientific basis for observed climate changes and their impacts Changes to natural water stocks are expected to severely in Central Asia is overall weak or lacking (Hijioka et al. 2014). affect irrigated agriculture. This impact will be compounded However, as mentioned in Section 5.4.1, sub-regionally pronounced under conditions that further increase the water demands of winter warming has been observed for Southern Siberia and the crop production systems, as rising temperatures are expected to Tien Shan mountains, concurrent with glacier volume change by lead to an around 30 percent increase in potential evapotranspi- about a third from the beginning of the 20th century. ration (see Section 5.3.5, Aridity). Increasing temperatures and Five Central Asian countries (Kazakhstan, Kyrgyz Republic, water demand also affect rain-fed agriculture, which accounts Tajikistan, Turkmenistan, and Uzbekistan) are particularly vulner- for more than 90 percent of arable land in Kazakhstan (FAO- able to climate change compared to the other ECA countries (Fay AQUASTAT 2012). et al. 2010). They face common climate challenges that affect such Uncertainties in precipitation projections translate into an key resources and sectors as water, land, biodiversity and ecosys- uncertain future for rain-fed crop production in the region. Fur- tems, agriculture, energy, and health. Water resource systems in ther, prolonged periods of above-average temperatures and heat Central Asia are sensitive to climate change and variability, and extremes exacerbate the heat stress of agricultural crops, leading to climate impacts on water supplies will reverberate across the agri- decreased plant productivity and high climatic risks for the sector cultural and energy sectors. Increasing temperatures are expected (Mannig et al. 2013; Teixeira et al. 2013). Such risks may limit a to increase both crop water requirements and evaporation and to projected increase in agricultural areas or crop yields. In addition, 204 E UR O PE A ND CENTRA L A S IA rising temperatures and seasonally reduced water availability puts There are many opportunities to improve climate resilience pressure on livestock directly and indirectly through limiting the in the region. Strengthening capacity and responsiveness of local regeneration potential of pastures. and national disaster risk management institutions in the region Losses in agricultural productivity and employment oppor- could significantly boost resilience to rapid-onset effects of climate tunities would add pressure to the labor markets and challenge change, such as those related to more rapid spring snow-melt poverty reduction for affected population groups. They could (UNISDR 2009). Upgrading of worn-out infrastructure and more further stimulate increased migration from affected areas to those effective poverty reduction and social protection are also priori- with stronger economies, potentially following already established ties (Thurman 2011). migration routes, such as from poorer areas of Central Asia to There is also room to improve water-saving irrigation tech- Russia, more locally to cities with better job opportunities (IOM niques in the region. Approximately 87 percent of the region’s 2011). Groups unable to migrate (older people, people with dis- extracted water is used in agriculture (FAO-AQUASTAT 2012), and abilities, in some cultural contexts, such as parts of Central Asia, in many Central Asian countries the water irrigation systems are women), are at greater risk of being trapped in poverty (Black et al. inefficient. In Uzbekistan, for example, 70 percent of the irriga- 2011) particularly if remittance flows are limited or unpredictable. tion water is lost between the river and the crop (Rakhmatullaev Further, decreased levels of agricultural production as well as et al. 2012). Changes in reservoir management and the need to lower levels of certainty over future yields are likely to contribute meet water requirements for agriculture can also have a negative to an increase in food prices. Rising food prices may have severe impact on energy availability over the summer months (Siegfried effects on the Central Asian population since a large percentage et al. 2012). of household income is spent on food; up to 80 percent in Uzbeki- stan and Tajikistan, and 58 percent in the Kyrgyz Republic (Bravi 5.5.2 Climate Extremes in the Western Balkans and Solbrandt, 2012). In addition, agricultural impacts in other Pose Major Risks to Agricultural Systems, large, food-producing regions, including Russia, can have negative Energy and Human Health repercussions on the Central Asian region as some countries (e.g., Tajikistan, Kyrgyz Republic, Uzbekistan, and Turkmenistan) are For the Western Balkans, a pronounced drying trend is projected, largely dependent on food imports and are exposed to fluctuations concurrent with strongly rising temperatures and prevalence of in food prices (Meyers et al. 2012; Peyrouse 2013). heat extremes. Such changes pose high risks to agricultural pro- A further risk to development stems from the energy sector, ductivity in the region. The majority of arable land in the Western which relies on stable water supplies. Despite an expected decrease Balkans is rain-fed and, therefore, highly vulnerable to changes in in heating loads during the winter (WorleyParsons 2012), overall climatic conditions (UNDP 2014). In the Former Yugoslav Republic energy demand is projected to rise together with population and of Macedonia, 90 percent of the agricultural area is rain-fed; in economic growth (World Bank 2013s). Changes in climate, reduced Albania, irrigation agriculture is practiced on roughly 50 percent of snow accumulation, and accelerated melting of snow and glaciers arable land (World Bank 2010b, 2011c). The share of the working increase uncertainty in the timing and amount of water available population of the Western Balkans that is employed in agricul- for power generation. Tajikistan and the Kyrgyz Republic, which ture varies from 18 to 58 percent depending on the country, and are located upstream of the Syr Darya and Amu Darya, respec- agriculture is directly responsible for 17 percent of the region’s tively, produce 98.8 percent and 93.3 percent of the total electricity GDP (Hughes 2012). Cereals and fruits (predominantly grapes) are consumed from hydropower (World Bank 2013p). Hydroelectricity the most important agricultural products in terms of production can also play a major role in the future energy mix of the Central area and economic output, with Serbia being the biggest producer Asian countries, as only eight percent of the hydropower potential (Mizik 2010; Volk 2010). of the region has been developed (Granit et al. 2010). The region is highly vulnerable to the effects of droughts, Climate change impacts will result in high variability of inflows as the 2012 drought in Serbia illustrates. It led to yield declines and a shift in the historical water flow patterns. Upstream countries of up to 50 percent and severe economic losses (Maslac 2012). would have to manage the impact of this change in their hydropower Such events need to be expected to occur more often under generation systems, which is the back bone of their electricity sec- climate change and absent preventive adaptation. In addition, tor. The downstream countries would see increased demand from pasture yields and grassland ecosystems for livestock grazing both the agriculture and energy generation sectors. Efficiency in are expected to decline and change for large parts of Eastern water use for irrigation and for energy generation would be critical Europe and the Western Balkans (Sutton et al. 2013a, b, c). Feed for managing this impact. Energy efficiency measures would help quality could also be negatively affected by changing climate reduce growth in water demand for energy generation. conditions (Miraglia et al. 2009). Declines in fodder production 205 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Figure 5.22: Sub-regional risks for development for Europe and Central Asia at 4°C warming in 2100 compared to pre-industrial temperatures. s i a n F e d e r a t i R u s o n 3RSXODWLRQ'HQVLW\ >3HRSOHSHUVTNP@  l Asia ² Centra Western ² Balkans ² ²  Western Balkans Central Asia Boreal Forests of the Russian Federation Increase in droughts, unusual heat Increasing glacial melt alters river runoff. extremes and flooding. High risks for Risks of glacial lake outbursts, flooding Unusual heat extremes and annual agriculture, human health and stable and seasonal water shortages. Increasing precipitation increase, rising risks of forest hydropower generation. competition for water resources due to fires and spread of pests leading to tree rising agricultural water demand and mortality and decreasing forest productivity. Risks for human health, food and energy demand for energy production. Possible northward shift of treeline and security. changes in species composition. Risks of Risks for poor through rising food prices permafrost melt and methane release. particularly affecting women, children and the urban poor. Risks for human health Risk for timber production and ecosystem due to spreading disease, heat waves and services, including carbon capture. Risks of flooding. substantial carbon and methane emissions. Data sources: Center for International Earth Science Information Network, Columbia University; United Nations Food and Agriculture Programme; and Centro Internacional de Agricultura Tropical—(2005). Gridded Population of the World, Version 3 (GPWv3): Population Count Grid. Palisades, NY: NASA Socioeconomic Data and Applications Center (SEDAC). This map was reproduced by the Map Design Unit of The World Bank. The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of The World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. could impact feed prices and potentially lead to greater price risk of disease outbreaks following the event was particularly volatility (World Bank 2010b). high in overcrowded refugee centers and as people returned to At the same time high vulnerability to flooding events exists. their homes too early (Holt 2014). This flood came against the Despite large uncertainties about future precipitation extremes backdrop of slow-onset risks to human health as climatic condi- in the region high risks of riverine flooding are expected mostly tions become increasingly suitable for vectors transmitting such due to more intense snow melt. The flood of 2014 illustrated the diseases as dengue fever (Caminade et al. 2012). Flooding also region’s vulnerability to such events. The torrential rains led to poses high risks to agriculture, as evidenced by a 2010 event that floods and landslides, killing 51 people and rendering over 31,000 caused damages of $450 million in Albania, Bosnia and Herze- people temporarily homeless. According to health officials, the govina, and FYR Macedonia (Hughes 2012). 206 E UR O PE A ND CENTRA L A S IA A joint assessment report on the social impacts of the 2014 temperatures and changing precipitation patterns, as well as Floods in Serbia by the United Nations, World Bank and EU sug- interactions with disturbance regimes, would have far-reaching gests that about 51,800 jobs were temporarily lost because of repercussions affecting the global carbon stock and planetary interruption of productive activities in the affected municipalities. albedo across large parts of the northern hemisphere. Overall, the floods were estimated to reduce economic growth by The boreal permafrost zones are a further potentially large 0.5 percent and to lead to a decrease in the balance of payments source of greenhouse gases. Already, today, large formerly undis- equivalent to 1 percent of GDP, with a further 1 percent loss related turbed permafrost zones are affected by thawing due to current to lower tax revenue and post-disaster expenditure. Furthermore, warming (Romanovsky et al. 2010), leading to some subsidence of the floods are estimated to have pushed 125,000 people below the housing, disruption of infrastructure and affecting the livelihoods poverty line, an increase of nearly 7 percent over the number of of indigenous minorities living in the Russian Arctic (Crate 2013). people living in poverty in 2013. This increase comprised not only Future thawing is projected to be very pronounced. people living close to the poverty line, but also a much better- Russian forests contributed 1.3 percent of GDP and 3.7 per- off group that lost its productive capital as a result of the floods. cent of industrial production in 2010 (FAO 2012b). The forestry The most affected people were typically those who faced specific sector employs one percent of Russia’s population and produces challenges before the floods, such as access to employment, low 2.4 percent of export revenues (FAO 2012b). In 2010, 32 million security of tenure, and limited social support networks. About m³ of wood raw material was used for biofuels, and that number 12 percent of the 1.6 million people affected by the floods were is expected to double by 2030 (FAO 2012b). Under future climate in groups considered particularly vulnerable, such as Roma and change the productivity of those forests might be at risk. people with disabilities. Many women were also doubly affected: in Climate projections for Russia show above-average tempera- addition to sustaining livelihood losses, they have had to increase ture rise and an overall increase in annual precipitation. While the non-paid time they devote to take care of their family (United fire outbreaks are often currently caused by anthropogenic factors Nations Serbia et al. 2014). rather than climate change, the risk of fire increases as higher tem- The 2014 flooding also led to a reported 40 percent cut in peratures increase the biomass potential of burning as it becomes Serbia’s electricity production and disrupted power supplies drier (Flannigan et al. 2009). In the future, much larger areas will (Sito-Sucic 2014); this highlights the energy sector’s vulnerability be exposed to forest fires (Stocks et al. 1998). The risk of fire is to climate impacts and extreme events. Changes in river water further enhanced by projected changes in vegetation distribution. temperature and river flows can also impact on thermal electricity If exposed to continuous water stress, the forest at lower latitudes production and reduce the capacity of nuclear and fossil-fuelled may give way to steppe ecosystems if the CO2 fertilization effect power plants through changes in cooling water. Critically, reduc- does not sufficiently compensate for this stress through enhanced tions would be concurrent with an increase in energy-intensive efficiency of water use. This would likely promote larger fires, cooling demand, which is projected to increase by 49 percent particularly in southern Siberia and Central Yakutia (Tchebakova (Isaac and van Vuuren 2009). et al. 2009). While slow-onset changes in mean temperature and aver- age precipitation will ultimately affect the broad patterns of 5.5.3 Responses of Permafrost and the Boreal future species distribution, the impacts of climate variability in Forests of the Russian Federation to Climate terms of extremes is an important driver of tree responses and Change Have Consequences for Timber vulnerability to climate change (Reyer et al. 2013). Even under Productivity and Global Carbon Stocks increased annual average precipitation, an increase in heat extremes may lead to strong ecosystem responses. Heat-stressed Russia has the largest forest area in the world, representing trees, for example, may be more susceptible to pest outbreaks. 20 percent of the global forest area (FAO 2012b). Its forests store Such increased vulnerability would, in turn, interact with an enormous amounts of carbon and deliver important ecosystem expected but understudied expansion of areas at risk of pest services, including through timber production. These services outbreaks (Bale et al. 2002). In combination, these stressors could may be compromised or even lost under high levels of warming. lead to increasing risks for neighboring forest stands or even a There is a risk that the boreal forest may cross a tipping point threshold behavior whereby ecosystems shift into an alternate and shift to an alternative state (e.g., steppe grasslands) (Lenton state (Lenton et al. 2008). et al. 2008; Scheffer et al. 2012). Moreover, the effects of biome Different responses and vulnerabilities of tree species to cli- shifts and damages from forest fires would not be confined to mate change could result in altered forest compositions. There are the region itself. Changes to carbon fluxes in response to rising indications, however, that the largest effects on forest composition 207 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL and fragmentation are due to timber harvesting (Gustafson et al. Climate-related effects on forest productivity can lead to both 2010). In line with this, Isaev and Korovin (2014) highlighted that an increase and a decrease in productivity. For now, it remains the large 2010 forest fire was due not only to unusual meteorologi- uncertain at what CO2 concentration and temperature increase cal conditions but also to poor forest governance and manage- a reversal from sink to source would occur. However, if such a ment—and to an increasing area of abandoned farmlands leading reversal were to occur it would affect the global carbon budget to declining numbers of forest managers, firefighters, and overall and thereby regions well beyond Russia itself. less efficient forest protection systems. Similarly, Flannigan et al. The sustainable and farsighted management of Russian eco- (2009) highlighted that, since the breakdown of the Soviet system, systems is of global importance. If pushed beyond tolerance limits the effectiveness of the Russian firefighting system has decreased; and into positive feedback mechanisms with regional and global this has led to larger areas being burned. Similarly, a weakening warming, major carbon and methane stocks in the boreal forests and of regulatory capacities has facilitated illegal and unsustainable permafrost zones may be released into the atmosphere. Critically, logging practices (Vandergert and Newell 2003) that are likely to carbon release would continue even if warming ceased (Schaefer undermine the forest’s resilience to climatic stresses. For example, et al. 2011; Schaphoff et al. 2013). Even without such threshold if logging is done in an unsustainable manner and high amounts effects, southern boreal carbon loss as a result of ecosystem shifts of dead or damaged trees are left in the forest, decaying trees may is likely to offset carbon gains from potential northern boreal for- trigger pest outbreaks and increase the risk of forest fires. Such est expansion (Friend et al. 2014; Schaphoff et al. 2013). Given negative ecosystem impacts might also undermine the livelihoods the large uncertainties in the overall response of Russian forests of groups dependent on the forest for timber and non-wood forest to effects such as future pest outbreaks and heat stress, manage- products (e.g. berries, mushrooms, medicinal plants), traditional ment practices need to be deployed that foster the resilience of agriculture and hunting (FAO, 2014d). It is thus the interaction of forests in order both to minimize the risks for those who depend climate changes, disturbances, and unsustainable forest manage- on them and to support global climate protection. ment and governance that will drive the vulnerability of forests, and people whose lives and livelihoods depend on them, to climate-change-related stresses. 208 5.6 Synthesis Table—Europe Central Asia Table 5.7: Synthesis table of climate change impacts in ECA under different warming levels. The impacts reported in several impact studies were classified into different warming levels (see Appendix for details). OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Heat Extremes Highly Unusual 5% of land area 10% of land area 15% of land area 50% of land area 85% of land area Heat Extremes Hotspot in the Balkans Unprecedented Absent Absent Almost absent 10% of land area 55% of land area Heat Extremes Precipitation Central Asia 20% increase 10% decrease to 10% increase (west to east) Western Uncertain 20–30% decrease Balkans Russian Return period of 20–30% precipitation 20–60% increase Federation maximum 1995 increase Return period of maximum precipitation: Return period of 1995 precipitation: 5–7 years 10–15 years1 maximum 1995 for Eastern Russia, 7–10 years precipitation: for Central Russia and less 10–15 years, and than 5 years for Central and 7–10 years for East Eastern Siberia1 Siberia1 Drought Drought duration Drought duration 20% more drought days for the (CDD) in the Balkans: (CDD) in the Balkans: Balkans, uncertain for Central 1–5 days2 5–10 days2 Asia and Central and Eastern Russia3 Drought duration (CDD) in the Balkans: 5–15 days2 Aridity Central Asia Uncertain Uncertain in the northern parts, up to 60% increase in the West and up to 60% decrease in aridity in the East Western 60% increase in aridity Balkans Russian 10–40% decrease in 20–60% decrease in aridity Federation aridity 209 210 Table 5.7: Continued. OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Glaciers 11% Central Asian volume Balkan glaciers About 50% (31– 54–57% of Central 50–78% of glacial volume loss between 1980 and melting within 66 percent) of Central Asian glacier volume shrinkage in Central Asia10 2011, 3–14% reduction in decades6 Asian glacier volume loss 9 area since 1960s4 31% (50 Gt = 56 km3 loss5 35.5% of glacier volume of ice) of Tien Shan 31% mass loss in loss in Central Asia between glaciers melting7 Syr Darya basin7 1901–20005 41% drop in the annual runoff per year8 Water Runoff Zerafshan river: shift from In Syr Darya basin, Increased spring Very significant decline of runoff Availability summer to spring and shifts of 30–60 days and summer runoff formation in the mountainous winter11 from the current in Central Asia; shift areas of Central Asia16 The Aral Sea volume spring/early summer of peak flow from 15 m Issyk-Kul lake level decrease due to toward a late winter/ July to June; 25% decrease14 climate change without early spring runoff discharge reduction up to 40% runoff decrease in anthropogenic water regime7 in July and August Albania17 abstractions was ~13.7% 5m Issyk-Kul lake level Hagg et al. (2013). 45–75% of increased water between 1958–200212 decrease14 discharge in Northeastern Strong river runoff 15–45% reduction in Russia; 15% increased in reduction of natural rivers water resources15 central Siberia15 of the Balkans, Sava, and More than 45% decrease Danube13 in annual discharge in the Western Balkans15 Groundwater Slight increase Recharge in groundwater recharge in Central Asia18 Crop Growing Areas and Food Desertification affecting 66% 10–15% reduced Increased desertification Production of Kazakhstan19 runoff in the Amu threatening wheat production in Severe droughts in Daraya River, putting Kazakhstan7 2000/2001 leading to pressure on the Increased aridity and 112,600 ha of cereal loss irrigation systems desertification in Kyrgyz in Tajikistan and $50 million and crop production; Republic affecting up to 49% of loss in Uzbekistan20 increased degradation the country’s territory23 $2 billion lost as a of soils22 consequence of 2012 Water deficits during drought in Serbia21 the vegetation period in the Fergana Valley7 OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Yields All Crops 50% yield declines following Up to 13% 20–50% yield loss 20% yield loss of 30% yield drops in some parts drought in 2012 in Serbia21 yield decline in in Uzbekistan due rain-fed grapes and of Tajikistan28 70% yield declines following Uzbekistan; in to heat and water olives in Albania25 drought in 2012 in Bosnia eastern parts of the stress26 and Herzegovina24 country yield increase 10–25% lower yields up to 13% possible25 in Uzbekistan due to Higher productivity of decreasing runoff in alfalfa and grasslands the Syr Daraya River in Uzbekistan26 and increased water use competition27 Up to 21% reduced yields of olives and 20% reduced yields of grapes in Albania; up to 69% reduced yield in grapes and up to 56% reduced yield in vegetables in Macedonia, FYR, up to 50% percent yield declines for wheat in Mediterranean and Continental parts of Macedonia25 Maize Very high concentrations of Up to 50% percent aflatoxin concentration in yield declines in maize as a result of 2012 Mediterranean and drought in Serbia29 Continental parts of Macedonia, FYR25 Up to 11% reduced yields Albania26 Cotton 0 to 6% decrease in Up to 19% reduced Uzbekistan26,30 yields in Uzbekistan26 211 212 Table 5.7: Continued. OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Wheat 12% average wheat yield increase31 Kazakhstan a major Up to 28% reduced Up to 57% reduced hotspot of heat yields in spring wheat yields in spring stress affecting in Uzbekistan33 wheat and up to wheat production 32 43% reduced yields in winter wheat in Uzbekistan33 Up to 24% yield increase in Albania33 Livestock 14–33% increase of Bacillis anthracis habitat34 Human Health Most of the Balkans Increased vulnerability Increase of heat- become suitable of the Balkans related mortality rates for Aedes aegypti to dengue and to 1,000 per million; dengue-transmitting chikungunya35 very slight decline mosquito35 in cold-related mortality36 Tenfold increase in risk of mudflow occurrences in Kazakhstan37 Energy 2.58% increase in capacity for hydropower generation in Central Asia38 Capacity of nuclear and fossil-fueled power plants could decrease due to changes in river water temperature and in river flows39 Mean number of days during which electricity production is possible drops due to the increase in incidence of droughts and extreme river low flow39 OBSERVED AROUND 4°C VULNERABILITY AROUND 1°C AROUND 1.5°C AROUND 2.0°C AROUND 3.0°C AND ABOVE RISK/IMPACT OR CHANGE (≈2010s) (≈2030s) (≈2040s) (≈2060s) (≈2080s) Boreal Forests Tree-line expansion in the Increase in timber Moderate change in Russian forest harvests are only north40 harvest for larch and vegetation44 profitable until 2060 under 4°C Productivity decline within pine42 10 days increase warming42 interior boreal forests related Decrease in timber in fire risk (50 to 60 Dramatic changes in to warming-induced drought harvest for spruce days)44 vegetation44 stress41 and fir42 Large decrease in timber Methane emissions harvest, especially for spruce from permafrost and fir; larch forest might thawing in Russia increase productivity42 could increase by 25t ha-1 in tree carbon (150t 20–30%43 ha-1 vs. 125 t ha-1) more and ~60 t ha-1 less in soils and deadwood (33–35 t ha-1 vs. 98–99t ha-1) under climate change; harvest increased by 15% under climate change; productivity increase nullified by higher fire damage45 20–30 days increase in fire risk (60 to 80 days)44 Increasing vulnerability to diseases, fire, and decreased reproduction rates that might lead to large-scale forest dieback46 Please note that years indicate the decade during which warming levels are exceeded with a 50 percent or greater change (generally at start of decade) in a business-as-usual scenario (RCP8.5 scenario) and not in mitigation scenarios limiting warming to these levels, or below (since, in that case, the year of exceeding would always be 2100 or not at all). Exceedance with a likely chance (>66 percent) generally occurs in the second half of the decade cited. Impacts are given for warming levels irrespective of the timeframe (i.e. if a study gives impacts for 2°C warming in 2100, then the impact is given in the 2°C column). Impacts given in the observations column do not necessarily form the baseline for future impacts. Impacts for different warming levels may originate from different studies and therefore may be based on different underlying assumptions, meaning that the impacts are not always fully comparable (e.g., crop yields may decrease more in 3°C than 4°C because underlying the impact at 3°C warming is a study that features very strong precipitation decreases. Moreover, this report does not systematically review observed impacts. It highlights important observed impacts for current warming but does not conduct any formal process to attribute impacts to climate change. 213 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Endnotes 1 Kharin et al. (2013) 2 Sillmann et al. (2013b) 3 Prudhomme et al. (2013) 4 Giesen and Oerlemans (2013), Hijioka et al. (2014) 5 Marzeion et al. (2012) 6 Glaciers in Albanian Alps and Montenegrin Durmitor (Grunewald and Scheithauer 2010) 7 Siegfried et al. (2012) 8 Bliss et al. (2014) 9 Giesen and Oerlemans (2013), Marzeion et al. 2012, Radic ´ et al. (2013) 10 Radic ´ et al (2013), Marzeion et al. (2012) 11 Olsson et al. (2010) 12 Aus der Beek et al. (2011) 13 7 out of 8 rivers, Dimkic and Despotovic (2012) 14 Davletkeldiev et al. (2009) 15 Schewe et al. (2013) 16 Main Administation of Hydrometeorology (2009) 17 Dakova (2005) 18 Döll (2009) 19 World Bank 2013v 20 Thurmann (2011) 21 Maslac (2012) 22 World Bank (2013f) 23 World Bank (2013a) 24 UNDP (2014) 25 Sutton et al. (2013a) 26 Without CO2 fertilization; Sutton et al. (2013a,c) 27 World Bank (2013x) 28 World Bank (2013m) 29 Kos et al. (2013) 30 Sutton et al. (2013b) 31 Without changes in irrigation water availability and with CO2 fertilization effect (Sommer et al. 2013) 32 Without CO2 fertilization (Teixeira et al. 2013) 33 Without CO2 fertilization (Sutton et al. 2013b,c) 34 Joyner et al. (2010) 35 Caminade et al. (2012) 36 Ballester et al. (2011) 37 National communication of Kazakhstan (BMU and WHO-Europe 2009). 38 Killingtveit (2012) 39 van Vliet et al. (2012) 40 Berner et al. (2013); Devi et al. (2008); MacDonald et al. (2008) 41 McDowell (2011); Zhang et al. (2009b) 42 Lutz et al. (2013b) 43 Anisimov (2007) 44 Tchebakova et al. (2009) 45 Forest in Kostroma region, 450 km northeast of Moscow (Shanin et al. 2011) 46 Lenton et al. (2012) 214 Appendix Appendix A.1 Methods for Temperature, de-trended signal, monthly standard deviations were calculated, Precipitation, Heat Wave, and Aridity then averaged seasonally (i.e., the seasonally averaged monthly- standard deviations). For this analysis, we employed the standard Projections deviation calculated for the last half of 20th century (1951–2010); we found, however, that this estimate was robust with respect to A.1.1 ISI-MIP Bias Correction different time periods. Following Coumou and Robinson (2013) The temperature, precipitation, and heat wave projections were and Hansen et al. (2012), we used the 1951–1980 reference period, based on the ISI-MIP global climate database, using the historical which has the advantage of being a period of relatively stable (20th century) period and future scenarios RCP2.6 and RCP8.5. global mean temperature prior to rapid global warming. The ISI-MIP database consists of five CMIP5 global climate models We defined two different extreme thresholds: one at three (gfdl-esm2m, hadgem2-es, ipsl-cm5a-lr, miroc-esm-chem, and standard deviations warmer than the mean temperature (3-sigma noresm1-m) which were bias-corrected such that the models events) and one at five standard deviations warmer than the reproduce historically observed mean temperature and precipita- mean temperature (5 sigma events). During the reference period tion and their year-to-year variability. The statistical bias correc- (1951–1980), exceeding the 3-sigma threshold is extremely unlikely. tion algorithm as used by WaterMIP/WATCH has been applied Over most land regions, the monthly temperature distributions to correct temperature and precipitation values. The correction are close to a normal distribution for which 3-sigma events have factors were derived over a construction period of 40 years, where a return time of 740 years. Monthly temperature will not be nor- the GCM outputs are compared to the observation-based WATCH mally distributed everywhere and hence return times can differ. forcing data. A regression is performed monthly on the ranked Nevertheless, for the reference period, return times for 3-sigma datasets. Subsequently, the derived monthly correction factors are events will on average be at least 100 years, implying that in each interpolated toward daily ones. The correction factors are then year the land area expected to experience temperatures beyond applied to the projected GCM data (Warszawski 2013). 3-sigma will be one percent or less. The land area experiencing 3-sigma heat is affected by natural variability, with El Niño years A.1.2 Heat Extreme Analysis seeing a larger area exceeding this threshold (Coumou and Robin- son 2013). Irrespective of that, 3-sigma heat extremes are unlikely For each of the ISI-MIP bias-corrected CMIP5 simulation runs, events during the reference period. Furthermore, 5-sigma events we determined the local standard deviation due to natural vari- have a return time of several million years in normally distributed ability over the 20th century for each individual month (Coumou data. They can thus be considered to be essentially absent during and Robinson 2013). To do so, we first used a singular spectrum the reference period. analysis to extract the long-term non-linear warming trend (i.e., The effect of global warming is to shift the mean temperature the climatological warming signal). Next we de-trended the 20th over almost all land regions toward warmer values. Even in the century monthly time series by subtracting the long-term trend, absence of a change in variability (i.e., a broadening or narrow- which provided the monthly year-to-year variability. From this ing of the width of the distribution), this shift in the mean will 217 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL cause an increase in the likelihood that the extreme thresholds Table A.1: Climatic classification of regions according to are exceeded. Therefore, the observed warming since the 1980s Aridity Index (AI). has already strongly increased the land area experiencing 3-sigma heat—it is now about five percent (Coumou and Robinson 2013). MINIMUM AI VALUE MAXIMUM AI VALUE Currently, at least on a global scale, 5-sigma heat extremes are Hyper-Arid 0 0.05 not yet detectable. As shown in this report, future warming will Arid 0.05 0.2 strongly increase the likelihood of exceeding—and therefore the Semi-Arid 0.2 0.5 land area experiencing—the 3- and 5-sigma threshold extremes. As Sub-Humid 0.5 0.65 discussed in Coumou and Robinson (2013), this increased likeli- hood is on the global scale is due primarily to the projected shift in the mean toward warmer values. Regionally, changes in the variability on top of the shift in the mean might play a role as well. With ET0 in mm day–1, Rn the net radiation at the crop surface Throughout this report, we analyzed the occurrence of threshold- [MJ m–2 day–1], G the soil heat flux density [MJ m–2 day–1], T the exceeding extremes both spatially and temporally (e.g., the mean air temperature at 2 m height [°C], u2 the wind speed at 2071–2100 period) aggregated. Temporal averaging is performed 2 m height [m s–1], es the saturation vapor pressure [kPa], ea the on an individual grid cell basis and the results are presented in actual vapor pressure [kPa], Δ the slope of the vapor pressure probability maps globally and for the report’s three focus regions. curve [kPa °C–1], and γ the psychrometric constant [kPa °C–1]. Additional spatial averaging is achieved by area-weighted averag- We calculated monthly ET0 values for each grid point using ing over the individual grid cells in the region of interest. This climatological input from the ISI-MIP database for both the his- way, a spatial-temporal averaged number is derived that can be torical period and future scenarios. interpreted either spatially (e.g., 80 percent of land area covered) or temporally (e.g., 80 percent of summer months) for a given A.1.4 Spatial Averaging period and region of interest. The time series for precipitation, temperature, heat extremes, and A.1.3 Aridity Index and Potential Evaporation aridity index, as provided in this report, have been obtained by area-weighted averaging of grid-cells located in the countries of The Aridity Index (AI) quantifies the precipitation supply over interest, as provided by the World Bank. atmospheric water demand as: A.2 Sea-Level Rise Projections: Pr AI = Methods for This Report ET0 A.2.1 Individual Contributions Where Pr is precipitation and ET 0 is the reference crop evapotranspiration, sometimes also called potential evaporation. We followed a process-based approach similar to the IPCC AR5 It estimates the amount of evapotranspiration from a hypotheti- report; this is in contrast to the previously employed semi-empirical cal grass reference crop with specific characteristics at a surface approach (Rahmstorf 2007; Schaeffer et al. 2012), which was the which is not short of water. It thus captures the water demand of method employed in Turn Down the Heat II (in a risk-assessment the atmosphere from meteorological conditions. We used the FAO perspective) and presented as the upper bound in Turn Down Penman-Monteith method (Allen et al. 2006) to estimate potential the Heat I. evapotranspiration rate, which is a simple representation of the The projection methods are a synthesis between a recent paper physical and physiological factors governing the evapotranspira- by Hinkel et al. (2014) and the references therein (we refer the tion process and generally recommended for the definition and reader to this publication for details) and the Chapter 13 of the computation of the reference evapotranspiration. It requires radia- IPCC AR5 WGI Report (Church et al. 2013). In particular: tion, near-surface air temperature, humidity, pressure, and wind speed data, and is given by: • Similar methods were used for the thermal, glacier, and Green- land contributions in the two publications and in this report. 900 u (e – e ) • Compared to Hinkel et al. (2014), we (1) accounted for a nega- 0.408Δ(Rn – G) + γ 2 s a ET0 = T + 273 tive surface mass balance (SMB) contribution from Antarctica Δ + γ(1 + 0.34u2) in addition to its (positive) dynamic contribution; (2) accounted for a dynamic contribution on top of Greenland surface mass balance; (3) used a larger number of CMIP5 GCMs (Taylor 218 A P P END IX et al. 2012) in our analysis—CNRM-CM5, CSIRO-Mk3-6-0, oceanic melt in Antarctica’s mass balance, however, we believe HadGEM2-ES, IPSL-CM5A-MR, MIROC-ESM, MPI-ESM-MR, that scenario dependency has to be accounted for in the sea-level MRI-CGCM3, and NorESM1-M; (4) considered the uncertainties rise projections for Antarctica. While our approach yields mean independent when summing the various contributions; and estimates for Antarctic sea-level rise similar to the IPCC, our (5) we took the cross-ensemble uncertainties as the full GCM estimated upper bounds are scenario-dependent. Rejecting this range (minimum to maximum) to yield a range consistent would undermine this report’s objective to inform on the risks of with the IPCC AR5 ensemble (except for Antarctica, where different levels of global warming. the calculations are probabilistic) (Levermann et al. 2014). After adding surface mass balance (Church et al. 2013), we obtained a contribution of 0.04 m (–0.03 m to 0.30 m) for the RCP8.5 • The main difference with the IPCC AR5 is the contribution scenario and 0.04 m (–0.01 to 0.18 m) for the RCP2.6 scenario in from Antarctica’s ice discharge, which we discuss below. Addi- 2081–2100 compared to the reference period 1986–2005. While tionally, and consistent with the previous Turn Down the Heat the median projection is similar for both scenarios (and similar reports, we did not include a contribution from groundwater to the IPCC AR5), the risk of higher sea-level rise is reflected by mining estimated at +0.04 m (–0.01, 0.08) (Church et al. the higher upper bound in the RCP8.5 scenario as compared to 2013, SM Table 13.5) since it is not related to climate warm- RCP2.6. Nevertheless, our upper bound only reflects model uncer- ing. The same applies to the post-glacial rebound, which is tainty, and can be seen as covering the likely (67 percent) range not included here. only. In particular, it does not include self-amplifying feedbacks A novel approach is employed for Antarctica’s dynamic dis- in unstable marine ice in Antarctica. Recent literature indicates charge (Levermann et al. 2014), where model results from the that the Thwaites Glacier may already be in a state of irreversible fixed-melting SeaRISE experiments (Bindschadler et al. 2013) are retreat (Joughin et al. 2014; Rignot et al. 2014), which was not “transformed” to account for scenario-dependent ocean warming clear and thus downplayed at the time IPCC AR5 was written and resulting melt rates. It uses the ensemble of CMIP5 models for (Parizek et al. 2013). the translation of global mean to subsurface ocean temperatures and a well-constrained parameter for the translation of warming A.2.2 Comparison with Previous Reports to ice sheet melting. This new method was also used by Hinkel et al. and Expert-Elicitation Studies (2014). We found this physically-based approach better suited for this report than the approach of Little et al. (2013) on which the A detailed comparison with previous Turn Down the Heat reports, IPCC assessment is largely based. In particular: IPCC AR4 (Meehl et al. 2007), and IPCC AR5 (Church et al. 2013), • The Little et al. (2013) estimate comes with a questionable after removal of land-water contribution for comparison purposes), assumption of linear growth rate in ice discharge, and must is shown in Figure A.1 for RCP8.5 and other 4°C warming sce- use subjective, rather arbitrary prior distributions of the growth narios. While our median estimates are similar to AR5, our upper rate in various basins. estimates of total global sea-level rise in 2081–2100 are significantly higher. This is due to the novel method for projecting Antarctica’s • The Little et al. (2013) estimate does not include processes that contribution, as explained above. In a 4°C world, the 90 percent are yet to start but expected as Antarctic subsurface waters model-range of our new projections tightly encompasses the median warm. As such, it may underestimate the risk for enhanced “low” and “SEM” cases investigated in the previous Turn Down ice discharge as a response to strong ocean warming toward the Heat reports, indicating a similar level of risk. In a 2°C world the end of the 21st century. (Figure A.2), the new process-based projections are more optimistic • While CMIP5 projections show a clear scenario dependency regarding the benefit of cutting emissions to limit sea-level rise. for subsurface temperature in the proximity of the Antarctic The two other projections (light blue on the figures) present ice sheet, the IPCC AR5 assumes a scenario-independent results from a very different approach based on expect elicitation. contribution to sea-level rise from Antarctica. Bamber and Aspinall (2013) interviewed 14 experts and esti- • Studies subsequent to the IPCC AR5 report add new evidence mated the sea-level contribution from the large ice sheets, taking that oceanic melt plays an important role for mass loss in into account both mass balance and fast ice flow processes. The general (Rignot et al. 2013), for West Antarctica (Dutrieux median result is a 29 cm increase from ice sheets by 2100; from a et al. 2014), and potentially for East Antarctica (Mengel and risk perspective, the 95th percentile of the estimates is also highly Levermann 2014). relevant being 0.84 m by 2100.68 Adding our estimates for RCP85 It is understandable that the “scenario-independence” null- hypothesis is agreed on in a consensus-driven report like the 68 Bamber and Aspinall (2013) “find an overwhelming lack of certainty about the IPCC AR5. In light of increasing evidence for the importance of crucial issue of the origin of recent accelerated mass loss from the ice sheets.” 219 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL thermal expansion and glaciers, the maximum sea-level rise is close Our results include the direct effect of Southern Ocean warm- to 1.4 m in 2100. A broader expert elicitation based on 90 experts ing on ice-shelf basal melting and related ice stream acceleration concluded an upper estimate (95 percent confidence) of 0.7 m sea- in Antarctica; as in the IPCC AR5, however, they do not include level rise for the low RCP2.6 scenario in contrast to 1.5 m for the self-amplifying feedbacks responsible for marine ice sheet instabil- high RCP8.5 scenario from 2000 to 2100 (Horton et al. 2014). An ity. It is still unclear at this point whether this mechanism would assessment by the U.S. National Research Council (2012) yielded a significantly change the picture presented in this report. We note similar range of 0.5–1.4 m. These studies implicitly account for the that despite significant methodological update, our process-based risk of destabilization of the potentially unstable Antarctic marine upper bound of 1 m by 2080–2100 above the 1986–2005 baseline is ice, and are consistent with the latest IPCC assessment that any comparable to the median projection of the semi-empirical-based additional contribution from the Antarctic ice sheet would remain “high” or “SEM” scenario investigated in the previous Turn Down within a couple of tens of centimeters above the likely upper bound. the Heat reports (Figure A.1). Figure A.1: Comparison of sea level projections by 2081–2100 above the present day, for the current report, previous Turn Down the Heat reports, IPCC reports, and recent subjective expert judgment assessments for a 4°C world (“Experts”: Bamber and Aspinall 2013; Horton et al. 2014). The uncertainty ranges reflect upper and lower bounds as originally reported, without distinction between likely and very likely ranges. Note that expert judgment assessments were originally reported for 2100 and subsequently adjusted by us to match the common projection horizon (2081–2100), assuming a linear increase in the rate of rise from present day to the projection horizon. Present-day baseline climate refers to 1986–2005 in the current report and may vary in other assessments (e.g. 1980–1999 for AR4, 2010 for BA2013), but no adjustment was made because results are less sensitive to the chosen baseline. Emissions scenarios vary slightly across the reports, but were selected to make the comparison at least qualitatively meaningful (e.g., A1FI for AR4, RCP8.5 for AR5 and Turn Down the Heat). Note that land-water storage is not included in this comparison and was removed from the IPCC AR5 estimate for consistency (estimated at 0.04 m (–0.01 m, 0.09 m) between 1986–2005 and 2081–2100, Table 13.5 in Church et al. 2013). 220 A P P END IX Figure A.2: Same as Figure A.1 but for a 1.5°C world. A.3 Meta-analysis of Crop Yield sample (i.e., having therefore a disproportionate impact on the meta-analysis). To minimize bias toward those studies, we aver- Changes with Climate Change aged their yield results for the whole region in order to obtain a A meta-analysis of crop yield data was conducted separately sample with a good representation of all the available studies. for the three regions in this report. In addition to the regional Moreover, whenever a study showed a range of GCMs models meta-analyses presented in the main report, we present here the for a specific crop, we interpreted averages as being the expected meta-analysis of the aggregate crop yield data for all three regions. response of aggregate production. The data from numerous studies were analyzed with the One quality control consisted of examining the datasets for goal of summarizing the range of projected outcomes for each outliers. We followed the same procedure described in Challinor et al. of the regions and of assessing consensus. We addressed three (2014) and examined in detail site-scaled studies that produced main questions in this analysis: (1) what are the likely impacts of changes of greater than 50 percent in either direction. This led incremental degrees of warming on yields?; (2) what is the qualita- to the exclusion of four data points. The focus was on crop data tive impact of considering adaptation measures and the effects of only; livestock data was not included in this analysis. CO2 fertilization on changes in crop yields?; and (3) what is the Due to the small sample size for the three regions and the ability of adaptation measures and CO2 fertilization to counteract large variety of crop types analyzed in the different studies, an the negative effects of increased temperature? analysis per crop type was not possible. Patterns were analyzed for all crops jointly, and particular features for individual crops A.3.1 Data Processing are described only qualitatively and with direct reference to the individual study from which they were obtained. Due to the lack Single studies may provide numbers for multiple countries within of data, we were also unable to test the effect of increased pre- the region (or even multiple sites within one country) and for cipitation on crop yield change. A small but significant correlation numerous crops, and could potentially be over-represented in the was found between an increase in temperature and an increase in 221 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL precipitation (Challinor et al. 2014), showing that the coefficient for significance the relationship between crop yield and tempera- found for effect of increase in temperature on yields captures some ture increase below 2°C and over the whole range of temperature of the effects increased precipitation may have on crop yields. increases. As mentioned above, we investigated the impact of incremental degrees of global warming on crop yields. Not all studies provided A.4 Warming Level Attribution all data needed and occasionally the increase in temperature and Classification associated with a projected change in yield is not explicit in the studies. Because climate change impact studies usually specify Differentiating between impacts at different warming levels is one baseline conditions from which future (e.g., under one of the of the key objectives of the Turn Down the Heat report series. Most SRES scenario family) impacts are projected, we were able to infer impact studies present their results with regard to time slices and the increase in temperature. This was done using the Warming scenarios—while not specifying global warming levels that might Attribution Calculator (see below). We did not retain studies that differ substantially due to large variations in climate sensitivity did not provide this information. between different GCMs. Additionally, not all policy-relevant warm- The baseline periods are highly variable between studies and ing levels are covered by SRES (CMIP3) or RCP (CMIP5) scenarios; seem to have been selected to coincide with the baselines assumed often the results are based on a single transient scenario only. in the climate model used in each study. In our datasets, baseline Wherever possible, we derived warming levels for the impact periods differed by up to 50 years between different studies; this studies analyzed in the report that are based on models and sce- certainly has an important impact on the projected yield (White narios from the CMIP3 or CMIP5 database. Generally, impact studies et al. 2011). It is important to bear in mind that the different lev- analyze changes with respect to a base period (e.g., 1986–2005). els of temperature increase examined here are not the same in Consistent with the IPCC AR5 WGI report, the warming level for absolute terms in relation to one baseline as we focused on the the period of interest was derived as the sum of the projected model crop response associated with incremental levels of warming. ensemble warming in global mean temperature (GMT) relative to This process led to a variable number of studies per region, the base period plus warming level of the base period relative to and the meta-analysis summarized the results from a total of pre-industrial levels (1850–1900) based on the HadCrut4 dataset 10 independent studies for Latin America and the Caribbean (e.g., 0.6°C for 1986–2005). (63 data points), 16 studies for the Middle East and North Africa If the impact study was based on multi-model analysis, a (167 data points), and three studies for Europe and Central Asia mean GMT time series was derived for all ensemble members (51 data points). Due to very low statistical power, we were unable equally weighted. Please note that this approach actually assumes to provide confident results for the ECA region. linear scaling of the impacts with temperature in the vicinity of the warming levels, which is an approximation that might not A.3.2 Statistical Analysis always be appropriate. Many impact studies do not differentiate, however, among the different ensemble members underlying the Statistical analysis was conducted using Matlab 2013a. We fitted projections, but give numbers only for ensemble averages. generalized linear models to the data to investigate the relation- For studies based on regional climate models, the GMT time ship between changes in crop yields and temperature increases, series of the corresponding GCM is used. If no climate model was and to address the effect of adaptation measures and of CO2. A specified in the impact study, the corresponding ensemble aver- t-test was conducted to test the relationships for significance. For age (either CMIP3 or CMIP5) was used. In addition, the CMIP3 the significant relationships, we present plots with best-fit lines data base contains not all scenarios that are analyzed in impact obtained using a polynomial fit. Five hundred bootstrap replicates studies (specifically, SRES A1F1 and SRES B2 are missing). For were carried out to derive a 95 percent confidence interval; these these scenarios, GMT scenarios were emulated using MAGICC6 are presented in patches. To evaluate the relationships, we looked (Rogelj et al. 2012). into the values of r-square (the amount of variability share between Based on this methodology, the impacts of climate change as the two variables in question) and the slope of the regression line, apparent from different studies can be classified for different warming which told us whether temperature increase influences crop yield levels that comprise the following ranges of global mean warming: change positively or negatively (all results are provided in tables). Previous analysis (Easterling et al. 2007; World Bank 2013) WARMING showed that the influence of temperature increase on crop yield LEVEL OBSERVED 1°C 1.5°C 2°C 3°C 4°C can be considerably stronger under high levels of temperature Range [°C] <0.8 0.8–1.25 1.25–1.75 1.75–2.25 2.25–3.5 >3.5 increase (higher than 2°C). Where data was available, we assessed 222 A P P END IX A.5 Summary of Evidence Concerning much research into the extent of relationship between climate change and conflict or social tensions, there is no consensus in Social Vulnerability the literature. This may reflect discrepancies between definitions The tables below are based on extensive literature searching on or approaches—it does not, for example, reflect a qualitative/ social vulnerability in the three focal regions and beyond. While quantitative divide, as both types of studies have found evidence extensive, this was not a systematic review process and the tables for and against strong relationship between climate change and make no claim to cover all relevant literature. The tables are orga- violence. Though some studies have raised concerns that both nized by themes and primarily focus on vulnerability. They do not extreme events and slow-onset disasters could lead to increased comprehensively review the very large literature on adaptation. gender-based violence, evidence remains anecdotal. Most evidence of social vulnerability to the effects of climate There is also some strong evidence of likely negative impacts change is based on and extrapolates from contemporary and his- on agriculture- and fisheries-based livelihoods, and more limited torical examples. Because of the inherent difficulties of modelling evidence concerning pastoral livelihoods—this may reflect research changes in social systems at different degrees of global warming, priorities, or may reflect limitations of our search. The evidence given the multiple other development trends which already interact concerning impacts of climate change on urban livelihoods with climate change to affect social vulnerability, and will do in emphasizes the destructive effects of extreme events on household the future, most evidence concerns the short to medium term. assets, community infrastructure and the knock-on effects on both Both large-scale quantitative research and small-scale contextual employed and self-employed workers. There is moderately strong research that probes social dynamics in particular locations are evidence that negative effects on livelihoods and wellbeing of both represented in research on social vulnerability to climate change. extreme events and slow-onset climate change may lead to both Most quantitative sources discuss the likely scale of vulnerability short-term displacement and longer-term migration. if no adaptation takes place. Evidence on the poverty implications of climate change is Of the three focus regions, there is considerably more evidence relatively limited, and largely based on projections of impacts of current and potential future social impacts of climate change on agriculture, food prices and consumption. These typically in Latin America and the Caribbean and the Middle East and suggest a likely increase in poverty among (already poor) small- North Africa than there is for the parts of Europe and Central scale producers in rural areas, and among low-income urban Asia studied in this report.69 This may reflect language barriers, households. Few studies discuss the implications of climate change for chronic poverty, or on social aspects of poverty, such or the disciplinary emphases of both natural and social scientists as on social cohesion. Relatively few studies probe the linkages working in Europe and Central Asia. Much of the existing evi- between climate change and coping strategies that may under- dence on social vulnerability has been funded by international mine social wellbeing, such as child labour or forced marriage, organizations. with a particular gap in evidence from the three focal regions. This said, compared with other regions for which there is Systematic treatment of the gender dimensions of climate change substantially more socially disaggregated evidence (such as Sub- is also weak for the three focal regions, with a particular absence Saharan Africa and South Asia), evidence for the Middle East and in the literature on ECA. North Africa, and to a lesser extent Latin America and the Carib- Although evidence from other regions (and to a lesser extent bean is still relatively thin on social analysis. These tables draw from Latin America) highlights the importance of responsive on the considerably larger literature from Sub-Saharan Africa and institutions and opportunities for voice in building resilience South, South-East and East Asia and the Pacific to highlight issues to climate change, there is relatively little evidence from either that may also be of relevance in LAC, MENA and ECA. MENA or ECA. Arguments for the importance of greater attention The aspect of social vulnerability where evidence is clearest to voice, rights and institutional development are largely based is health: there is strong evidence on vulnerability to heat waves on experience in other regions. and on access to clean, safe water, mixed but suggestive evidence The following tables set forth the evidence on current social on the likelihood of vector- and water-borne diseases spreading, vulnerability, as studies examining potential impacts in scenarios and on the likely implications of climate change for malnutrition. of greater climate change extrapolate from contemporary evidence. The evidence is least strong for impacts on mental health, and on As far as possible, assessments of the strength of evidence are all forms of interpersonal violence. Thus, although there has been based on discussions in the relevant chapters of the IPCC Working Group II 5th Assessment Report, Climate Change 2014: Impacts, 69 Note that this region excludes Western Europe, for which there is a large body Adaptation, and Vulnerability (IPCC 2014b). Where the strength of evidence. of evidence is not specifically signaled in the relevant IPCC 5th 223 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Assessment Report chapter, we made our own assessment and • A limited evidence base indicates inconclusive findings, or categorized the strength of evidence as follows: fewer than three studies with similar findings. • A strong evidence base denotes a consensus among studies, These relatively small numbers reflect the limited evidence and/or eight or more studies with similar findings. base on many aspects of social vulnerability. • A moderate evidence base denotes mixed findings, or 4–7 studies with similar findings. 224 Table A.2: Summary of Evidence: Food Security and Nutrition. INTERACTIONS WITH ADAPTATION AND TYPE OF BIOPHYSICAL INTERACTION WITH WHO IS PARTICULARLY MITIGATION EVIDENCE RISK/IMPACT CHANGE OTHER TRENDS AFFECTED AND WHERE PROCESSES BASE KEY REFERENCES Reduction of Water scarcity Soil degradation Regions: Adaptation: Moderate Swedish Government (2007) land available/ Sea-level rise Global pressures on land, MENA, especially Israel, Changes in land use, such as Hazell and Wood (2008) suitable for crops and ecosystems Salt water intrusion including for the production Lebanon, Syria, Iraq and the adjusting the location of crop Ortiz et al. (2008) (variable by region) River flow changes of biofuels Islamic Republic of Iran production in higher latitudes Gornall et al. (2010) Desertification Crop diversification potential Groups: Development of markets that Schroth et al. (2009) Deforestation Small-scale farmers and reward sustainable land-use Bellon et al. (2011) Land enclosure marginalized groups most practices Tacoli et al. (2013) likely to be displaced by Changes in land allocation competition for land Forest conservation Indigenous communities and small-scale farmers who lack land entitlement Potential knock-on effects on food prices related to squeeze on land availability Reduction in crop Temperature increases Trade flows Regions: Adaptation: High Schmidhuber and Tubiello productivity (e.g., Sea-level rise Crop diversification potential Tropical and subtropical Altered cultivation techniques confidence (2007) cereal production) especially for Changes in precipitation regions and sowing times Parry (2007) wheat and maize/ Increased frequency and Rain-fed agriculture in LAC Improved sowing techniques Lobell et al. (2008) negative yields severity of extreme weather Western Balkans (e.g., dry sowing, seedling Battisti and Naylor (2009) impacts for nut events such as flooding and transplanting, seed Van Dingenen et al. (2009) and fruit trees warm spells Groups: priming, double cropping or Wassmann et al. (2009) Melting glaciers and changes Rural food producers intercropping) and breeding Welch et al. (2010) in river hydrology Low-income urban drought-tolerant crop varieties Gornall et al. (2010) Decline of winter chill consumers Improved climate forecasts to Thornton et al. (2011) accumulation Groups reliant on glacial melt inform crop risk management Avnery et al. (2011) water irrigation in Central Asia Crop insurance programs that Okada et al. (2011) and LAC are accessible to smallholders Lobell et al. (2011) Irrigation optimization Zwiers et al. (2011) Semenov et al. (2012) Teixeira et al. (2013) Porter et al. (2014) 225 A P P END IX 226 Table A.2: Continued. INTERACTIONS WITH ADAPTATION AND TYPE OF BIOPHYSICAL INTERACTION WITH WHO IS PARTICULARLY MITIGATION EVIDENCE RISK/IMPACT CHANGE OTHER TRENDS AFFECTED AND WHERE PROCESSES BASE KEY REFERENCES Disruption to Temperature increases Governance Hazard prone countries and Moderate Swedish Government (2007) the production, Sea-level rise Political instability areas Tacoli et al. (2013) storage, and transport of staple Changes in precipitation and Conflict Carty (2013) food supplies flooding Global staple food price Nelson, Rosegrant, and Increased frequency and increases Palazzo et al. (2010) severity of extreme weather Trade flows Lobell et al. (2011) events such as landslides, Stock levels Hertel et al. (2010) mudslides, hailstorms, and Ziervogel and Ericksen (2010) erosion damage Douglas (2009) Reduction in Related to changes to Increased links between Low-income countries and Adaptation: Medium Ivanic and Martin (2008) affordability of productivity and disruption of energy and agricultural food-importing countries On-farm agronomic confidence Battisti and Naylor (2009) food/variability of food prices transport markets and oil price Regions: adaptation (evidence Lobell et al. (2011) Increased frequency and fluctuations Africa, Central America, Adaptation of food systems depend on Roberts and Schlenker (2010) severity of extreme weather Finance speculation northeast Brazil, parts of the Marketing arrangements assumptions Wright (2011) events Increased crop demand Andean region Diversification of activities of models) World Bank (2012) (e.g., biofuels) Central Asia Tacoli et al. (2013) Increased global staple food MENA FAO (2013) prices Groups: OECD/FAO (2013) Trade flows Low-income people in rural Skoufias et al. (2012) Conflict and urban areas Porter et al. (2014) Changing diets Children at risk of malnutrition Increased Temperature increases Population pressure Regions: Adaptation: Medium for Schmidhuber and Tubiello livestock Increased frequency and Land ownership patterns Arid and semi-arid regions Use of more suitable livestock mortality (2007) vulnerability and mortality severity of extreme weather Europe and North America breeds or species Limited for Swedish Government (2007) events such as drought Groups: Migratory pastoralist activities livestock UK Government (2011) and flooding, affecting Agro-pastoralists and Adjusted livestock and water Craine et al. (2010) productivity/availability of pastoralists management to forage Izaurralde et al. (2011) grazing land and production production Hatfield et al. (2010) of forage and feed Use of diet supplements Guis et al. (2012) Rapid spread of livestock Enhanced climate forecasts Porter et al. (2014) diseases and virus and information systems INTERACTIONS WITH ADAPTATION AND TYPE OF BIOPHYSICAL INTERACTION WITH WHO IS PARTICULARLY MITIGATION EVIDENCE RISK/IMPACT CHANGE OTHER TRENDS AFFECTED AND WHERE PROCESSES BASE KEY REFERENCES Increase in crop Drought Poverty Groups: Medium Schmidhuber and Tubiello diseases or pests, Temperature increases Population pressures Food producers (especially confidence for (2007) which may reduce or increase crop Heavy rainfalls small-scale farmers). weeds Tubiello et al. (2007) yields (depending Medium Luck et al. (2011) on crop type) confidence for Porter et al. (2014) insect pests Disruption to Temperature increases Population pressures and Regions: Adaptation: Strong Roessig et al. (2005) fishery and shell- Ocean warming increased demand for Tropical developing countries Changes in water and land Perry et al. (2005) fishery production, including fish Ocean acidification seafood Caribbean coasts, the use (e.g., increased offtake of Limited Swedish Government (2007) migrations Increased frequency and Fish stock overexploitation Amazon estuaries, and the water to irrigate new land) evidence Last et al. (2011) severity of extreme weather Pollution Rio de la Plata Increased dam building but high Doney et al. (2012) events Increase in fishery production International regulations to agreement Schmidhuber and Tubiello in some higher-latitude areas limit overfishing on the (2007) Groups: Integrated water use planning socioeconomic FAO (2013) Artisanal fishermen impacts Cheung et al. (2013) People engaged in fish of ocean Porter et al. (2014) processing and trading acidification Small coastal communities Declines in coral Temperature increases Overfishing Regions: High Wilson et al. (2006) reefs resulting Ocean warming Caribbean, Western Indian confidence Burke et al. (2004) in decline of fish stocks Ocean acidification Ocean Hoegh-Guldberg et al. Groups: (2014) Small coastal communities relying on coral ecosystems People engaged in fish processing and trading 227 228 Table A.3: Summary of Evidence: Poverty Impacts. INTERACTION WITH ADAPTATION TYPE OF BIOPHYSICAL INTERACTION WITH WHO IS PARTICULARLY AND MITIGATION EVIDENCE RISK/IMPACT CHANGE OTHER TRENDS AFFECTED AND WHERE PROCESSES BASE KEY REFERENCES Increase in poverty Effects on food prices related Structure of employment Regions: Adaptation: Medium Ruth and Ibarran (2009) headcount rate to effects on productivity Patterns of food production Sub-Saharan Africa (Malawi, Social protection confidence Jacoby et al. (2011) and risk of chronic poverty in different Effects on employment and Increasing food commodity Mozambique, Tanzania, (mixed Hertel et al. (2012) warming scenarios wages related to agricultural prices Zambia), Asia (Bangladesh), findings, Olsson et al. (2014) productivity Reductions in the availability and Latin America (Mexico) depending on of and access to natural Groups: assumptions resources Urban poor groups and urban of models) Corruption wage laborers Residents of informal settlements Dwellers in rural hotspots where hunger is expected to become prevalent Increase in Increased frequency and Extent of social protection Regions: Adaptation: Limited Carter et al. (2007) disaster-related severity of extreme weather and effective adaptation People in exposed areas Social protection evidence Shepherd et al. (2013) impoverishment and destruction events such as flooding, strategies (e.g., low-lying coastal areas, Improved access to basic (some Rossing and Rubin (2011) of assets. Risk of drought, landslides, and Urbanization flood-prone land, mountain services variation Leichenko and Silva (2014) chronic poverty cyclones Corruption slopes) Development of insurance in findings Olsson et al. (2014) varies, but is Implementation of disaster Groups: systems, particularly among by type of compounded by often-limited risk reduction measures Low income rural and urban vulnerable groups disaster and access by the Increasing proportion of groups Disaster risk reduction context) poorest to disaster assets and people located Children and adolescents strategies relief in areas of high exposure to Self-employed urban groups hazards Coping strategies Slow onset changes Inadequate investment in Regions: Adaptation: Limited Brown (2012) with negative (e.g., drought) poverty reduction Exposed areas in all regions Social protection ILO (2012) social impacts Extreme events (e.g., flooding, Increasing proportion of Groups: Improved access to basic North (2010) cyclones) assets and people located in Low-income groups services Swarup et al. (2011) high exposure areas Children (e.g., child labors, Development of insurance Broader economic shocks those removal from school) systems, particularly among Girls/young women (e.g., vulnerable groups forced marriages) Disaster risk reduction strategies INTERACTION WITH ADAPTATION TYPE OF BIOPHYSICAL INTERACTION WITH WHO IS PARTICULARLY AND MITIGATION EVIDENCE RISK/IMPACT CHANGE OTHER TRENDS AFFECTED AND WHERE PROCESSES BASE KEY REFERENCES Strained social Slow onset changes (e.g., Changing social norms and Regions: Adaptation: Limited Afifi et al. (2012; 2011) cohesion and drought) expectations of reciprocity Exposed areas in all regions Social protection Some Berman et al. (2014) decline in reciprocity Extreme events (e.g., flooding, Inadequate investment in Groups: Improved access to basic counter- Buechler (2014; 2009) cyclones) poverty reduction Low-income groups services evidence Slettebak (2012) Conflict Groups experiencing Development of insurance suggesting Broader economic shocks sudden impoverishment and systems, particularly among increase competition for resources vulnerable groups in social Disaster risk reduction cohesion in strategies immediate aftermath of disasters Worsening – Commercial pressures on Regions: Mitigation: Medium Olsson et al. (2014) poverty for some land Tropical forests and farmland Land alienation for forest confidence Biddulph (2012) groups as a result of mitigation Growing demand for energy Groups: protection (REDD+) or Chia et al. (2013) strategies Groups with limited land rights biofuels Couto Pereira (2010) (including indigenous groups, Hodbod and Tomei (2013) women, smallholders without Vanwey (2009) formal tenure) Socially excluded groups 229 230 Table A.4: Summary of Evidence: Migration. INTERACTIONS TYPE OF CLIMATE WITH ADAPTATION CHANGE OR SECOND- INTERACTION WITH WHO IS PARTICULARLY AND MITIGATION EVIDENCE RISK/IMPACT ORDER EFFECT OTHER TRENDS AFFECTED PROCESSES BASE KEY REFERENCES Migration as Gradual changes in Poverty Regions: Adaptation: Medium Black et al. (2008) a means for temperature and rainfall Socioeconomic inequalities Coastal cities and fertile deltas Temporary migration as a risk Tacoli (2009) securing livelihoods in the patterns Environmental degradation which are likely to experience management strategy or last Barnett and Webber (2010) face of slow-onset Drought Slow employment growth in a large population increase resort option Warner (2010) climatic stress Coastal erosion relation to population pressure Small islands and coastal Livelihood and sustainable Gemenne (2012) Increased sedimentation Landholding access plains at higher levels of sea development programmers Verner (2012) Rising sea levels Communications and rise under 4°C warming (e.g., have key role in reducing Grant et al. (2014) Thawing permafrost infrastructure Caribbean and Mediterranean vulnerability Crate (2013) Ability to adapt (particularly Coast countries) Adger et al. (2014) important for dry land regions) Maghreb countries that Social networks serve as receiving and transit Employment availability in countries for Sahelian and urban centers other Sub-Saharan African migrants Russian Arctic likely to experience flooding, subsidence and emigration, but also some in-migration to exploit emerging farming and extractive opportunities Groups: People with few or no land holdings Men are more likely to migrate but this depends on local social norms and labour market opportunities. Women left behind face additional work burdens. INTERACTIONS TYPE OF CLIMATE WITH ADAPTATION CHANGE OR SECOND- INTERACTION WITH WHO IS PARTICULARLY AND MITIGATION EVIDENCE RISK/IMPACT ORDER EFFECT OTHER TRENDS AFFECTED PROCESSES BASE KEY REFERENCES Displacement (as Increased frequency and Poverty Regions: Adaptation: Medium Black et al. (2011) a result of extreme severity of extreme weather Inequality Areas prone and vulnerable to Temporary migration as a risk Cutter (2009) events) events (e.g., flooding) Increased movement of hazards management strategy or last Barnett and Webber (2010) people and assets to areas of Groups: resort option Tacoli (2009) high exposure Elderly and poorest are less Disaster risk reduction Kartiki (2011) likely to leave. However, strategies Grant et al. (2014) when they leave, poor are Adger et al. (2014) (at greater risk of permanent displacement) Displaced women sometimes find it more difficult to generate a livelihood in discriminatory labour markets. 231 232 Table A.5: Summary of Evidence: Health. INTERACTION WITH ADAPTATION TYPE OF BIOPHYSICAL INTERACTION WITH WHO IS PARTICULARLY AND MITIGATION EVIDENCE RISK/IMPACT CHANGE OTHER TRENDS AFFECTED AND WHERE PROCESSES BASE KEY REFERENCES Increase in malaria Temperature Increase (Land Globalization Regions: Adaptation: Medium Kjellstrom and McMichael and sea surface temperature) Population displacement Highland areas Improved access to basic (Smith et al., (2013) Humidity change Local patterns of behavior MENA regions could be more services and health care 2014) Gage, Burkot, Eisen and Water Scarcity and settlement exposed to risks of malaria Substantial Hayes (2008) Flooding Existing health structures Groups: evidence with Mantilla et al. (2009) El Niño effects Land use changes People who lack immunity. diverging Huynen et al. (2013) Poor children at greatest risk conclusions Smith et al. (2014) Migrants However, Low income groups most stronger exposed evidence for highland areas Increase in dengue Temperature Increase (land Globalization Regions: Adaptation: Medium Patz et al. (2005) fever and sea surface) Population Displacement Tropical cities Development of pre-seasonal (Smith et al., BMU/WHO (2009) Increased precipitation Local patterns of behavior Groups: treatments (spreading of 2014) Costello et al. (2009) Increased frequency and (e.g., water storage) Low income groups most insecticides) Smith et al. (2014) severity of extreme weather exposed Improved access to basic events such as flooding and services and health care tropical cyclone activity Increase in water- Increased precipitation Local patterns of behavior Regions: Adaptation: Medium Few et al. (2004) borne diseases Changing rainfall patterns Population Pressure Tropical cities Improved access to basic (Smith et al., Azad et al. (2014) (diarrheal disease Flooding Inadequate basic services Coastal populations services and health care 2014) Khan et al. (2011) and cholera) Water Scarcity such as sanitation Groups: BMU/WHO (2009) Increased salinization Children in poverty affected Smith et al. (2014) El Niño effects areas Elderly populations Lower socioeconomic groups INTERACTION WITH ADAPTATION TYPE OF BIOPHYSICAL INTERACTION WITH WHO IS PARTICULARLY AND MITIGATION EVIDENCE RISK/IMPACT CHANGE OTHER TRENDS AFFECTED AND WHERE PROCESSES BASE KEY REFERENCES Increase in Temperature increase Air pollution Groups: Mitigation: ‘Very high Haines et al. (2006) respiratory Spread of pollen and other Industrial development Older people and children Political regulations to control confidence’ in Shea et al. (2008) diseases allergens Use of solid biomass fuels biologically at greater risk and reduce air pollution medium term D’Amato and Cecci (2008) Poor combustion of solid Women at risk from indoor air CO2 and ozone emission (Smith et al., Smith et al. (2014) cooking fuels pollution controls 2014) Urbanization Poorer socioeconomic groups Adaptation: Improved health services and access to health care Increase in food Temperature increase Energy costs Groups: Adaptation: ‘High Kovats et al. (2004) borne infectious Changes in precipitation Food storage practices Poorer socio-economic Improved access to basic confidence’ Patz et al. (2005) diseases groups services and health care Smith et al. Swedish Government (2007) Older people (2014) Smith et al. (2014) Children Reduction in Flooding Population pressure Regions: Adaptation: Strong Kjellstrom and McMichael availability of clean Drought Displacement Coastal Zones and low lying Improved access to basic (2013) water supply and Salt water intrusion Urbanization populations (e.g. Bangladesh) services and health care Black et al. (2013) sanitation Increased frequency and Cities reliant on highland or Barnett and Webber (2010) severity of extreme weather on declining ground water Moser et al. (2010) events sources Groups: Low income groups in rural & urban areas Women and children— longer distances to obtain water, affecting health. Also increased risk of violence Poor children esp. vulnerable to disease 233 234 Table A.5: Continued. INTERACTION WITH ADAPTATION TYPE OF BIOPHYSICAL INTERACTION WITH WHO IS PARTICULARLY AND MITIGATION EVIDENCE RISK/IMPACT CHANGE OTHER TRENDS AFFECTED AND WHERE PROCESSES BASE KEY REFERENCES Increase in heat Temperature increase Population ageing Regions: Adaptation: ‘Very high Kjellstrom et al. (2009) related illnesses Increased incidence of heat Urbanization Global Improved access to basic confidence’ Hanna et al. (2011) and reduced waves Particularly in densely services and health care for illnesses Berry et al. (2010) labour productivity Ozone ambient pollution populated large cities and ‘high Kjellstrom and McMichael Evidence from MENA, such confidence’ (2013) as the Arabian peninsula for reduced Costello et al. (2009) Groups: labour Smith et al. (2014) Elderly productivity Manual laborers and those (Smith et al. working outdoors are more 2014). exposed to heat stress Overweight people Displaced/people living in shelters People who are not able to cool housing in the night, lost sleep Residents of urban heat islands Increased Flooding Lack of adequate shelter, Regions: Adaptation: ‘Very high Neumayer and Plumper mortality rates Tropical cyclones water and sanitation services. ECA esp. Western Balkans. Improved access to basic confidence’ (2007) from extreme Lack of access to health Groups: services and health care (Smith et al. Pradhan et al. (2007) weather events and disasters services Women and girls at increased Investment in disaster 2014) Bartlett (2008) Concentration of poor onto risk if social norms prevent management and early Bradshaw and Fordham vulnerable land them acquiring survival skills warning systems (2013) Men/older boys—if expected Disaster Risk Reduction Skinner (2011) to risk their lives to rescue strategies Smith et al. (2014) others Children, older people more biologically vulnerable Poorer households People in low elevation costal zones and on land prone to flooding and landslides INTERACTION WITH ADAPTATION TYPE OF BIOPHYSICAL INTERACTION WITH WHO IS PARTICULARLY AND MITIGATION EVIDENCE RISK/IMPACT CHANGE OTHER TRENDS AFFECTED AND WHERE PROCESSES BASE KEY REFERENCES Increase in mental Temperature increase Access to health services Urban areas where mental Limited Berry, Bowen and Kjellstrom illnesses Increased frequency and Social support systems disorders are more common evidence (2010) severity of extreme weather Forced displacement from Poorer groups Alston (2010) events familiar environments Displaced people Smith et al. (2014) Stress of livelihood failure with Gender differences related to changing climate local social norms Increasing Temperature increase Population Pressure Regions: Adaptation: ‘Medium Lloyd et al. (2011) Malnutrition Changing precipitation Decline in agricultural Sub-Saharan Africa, South Use and development of confidence’ Parry et al. (2009) Increased frequency and production Asia, Central America and irrigation facilities (Smith et al. Skinner (2011) severity of extreme weather Fluctuations of food prices MENA Improved access to markets 2014) Nelson et al. (2009) events Lack of income and access Groups: Improved social protection Smith et al. (2014) to food Children (esp. infants) Discriminatory social norms Subsistence farmers in low Existing undernutrition rainfall areas patterns Urban poor Governmental instability Women (particularly in South Asia) Potential for Increased frequency and Existing gender-based Groups: Adaptation: Limited Swarup et al. (2011) increased risk severity of extreme weather inequalities Women and children across Improved targeting of evidence Ahmad (2012) of domestic and events Lack of supporting justice cultures and countries are awareness campaigns, Azad et al. (2014) sexual violence Longer-term climatic stress system primarily affected by gender- particularly where illiteracy Bradshaw and Fordham Increased stress from based violence rates are high. (2013) displacement and poverty Lack of security within shelters/in post-disaster environment 235 236 Table A.6: Summary of Evidence: Conflict and Security. INTERACTIONS WITH ADAPTATION TYPES OF INTERACTION WITH WHO IS PARTICULARLY AND MITIGATION EVIDENCE RISK/IMPACT BIOPHYSICAL CHANGE OTHER TRENDS AFFECTED AND WHERE PROCESSES BASE KEY REFERENCES Risk of land and Drought Governance instability Regions: Adaptation: Medium Rayleigh and Urdal (2007) water scarcity (or Land degradation Population pressures and Countries already affected by Diversification of income- Lack of Hendrix and Saleyhan (2012) excess of water) contributing to Changes to precipitation high population density conflict Countries where there generating activities in conclusive Harris et al. (2013) conflict/tensions Glacial melting High demand for natural are tensions between the agriculture and fishing findings on the Kallis and Zografos (2012) Sea-level rise resources mining industry and farmers/ Temporary migration as a risk relationship Adano et al. (2012) Lack of entitlements and indigenous groups (e.g. management strategy or last between Theisen (2012) access to key resources Peruvian Andes) resort option climatic Kloos et al. (2013) Overexploitation of Low-lying areas exposed to Transboundary water changes and Benjaminsen et al. (2012) groundwater reserves sea-level rise cooperation and flood conflict Adger et al. (2014) Storage capacities Groups: management Interboundary relations Land holders Preexisting tensions Farmers/subsistence farmers High projected urbanization Herders rates Indigenous groups Extreme weather Floods Preexisting tensions Regions: Adaptation: Limited Nel and Righarts (2008) events or sudden Tropical cyclones Grievances (e.g., related to Countries where governance Disaster risk reduction (inconclusive Harris et al. (2013) disasters leading to conflict/social distribution of relief supplies is weak or visibly inequitable strategies evidence) Bergholt and Lujala (2012) unrest and other resources) Groups: Adger et al. (2014) Food price fluctuations Poor people Weak governance Children Protests related to Poverty and inequality Regions: Limited (much Ortiz et al. (2013) increased food or Financial speculation over More common where speculation Hoffman and Jamal (2012) fuel prices food stocks and energy governance is weak or visibly but few prices inequitable studies) Carbon subsidies Groups: Low income urban groups INTERACTIONS WITH ADAPTATION TYPES OF INTERACTION WITH WHO IS PARTICULARLY AND MITIGATION EVIDENCE RISK/IMPACT BIOPHYSICAL CHANGE OTHER TRENDS AFFECTED AND WHERE PROCESSES BASE KEY REFERENCES Increased risk of Increased frequency and Population pressures Regions: Adaptation: Risk of Buhaug et al. (2010) conflict through severity of extreme weather Reduction of and competition Countries where resources Diversification of income- increased Stern (2013) climate/extreme event-induced events over natural resources are scarce/or physically generating activities in conflict cannot Bernauer et al. (2012) displacement Visible inequalities vulnerable to climate change agriculture and fishing be ruled out Koubi et al. (2012) with inequalities on ethnic/ Temporary migration as a risk Scheffran et al. (2011) Cultural clashes regional lines management strategy Theisen et al. (2012) Groups: Transboundary water Adger et al. (2014) Political instability Low-income groups cooperation and flood People who lack political management recognition 237 Bibliography Bibliography Abdulla, F., Eshtawi, T., and Assaf, H. (2008). “Assessment of the Impact Agarwal, B. (1994). “Gender and Command over Property: A Critical Gap of Potential Climate Change on the Water Balance of a Semi-arid in Economic Analysis and Policy in South Asia.” World Development, Watershed,” Water Resources Management, 23(10), 2051–68. 22(10), 1455–78. Abiodun, B.J., Pal, J.S., Afiesimama, E.A., Gutowski, W.J., and Adedoyin Aguirre-Salado, C.A., Treviño-Garza, E.J., Aguirre-Calderón, O.A., Jiménez- A. (2007). “Simulation of West African Monsoon using RegCM3 Part Pérez, J., González-Tagle, M.A., and Valdez-Lazalde, J.R. (2012). II: Impacts of Deforestation and Desertification.” Theoretical and “Desertification-Climate Change Interactions: México’s Battle Against Applied Climatology, 93(3–4), 245–61. Desertification.” Diversity of Ecosystems, edited by A. Mahamane, Abramovitz, J. (2001a). “Unnatural Disasters.” Worldwatch Paper 158, InTech: Rijeka, Croatia. The Worldwatch Institute: Washington DC. Ahizoun, M., Maire, E., Haida, S., Darrozes, J., and Probst, J. (2009). Abramovitz, J. (2001). “Averting Unnatural Disasters,” in State of the “Estimation des Changements de la Ligne de Rivage de la Zone Côtière World, 123–243. Sablonneuse de Kénitra au Maroc Résumé Abstract,” 05(2), 184–98. Ackerman, F., and Stanton, E.A. (2013). “Climate Impacts on Agriculture: Ahmad N. (2012). “Gender and Climate Change in Bangladesh: The Role of A Challenge to Complacency?” Global Development and Environment Institutions in Reducing Gender Gaps in Adaptation Program.” Social Institute, Tufts University. Development Paper 126, World Bank, Washington DC. Adamo, S.B. (2013). “Migration, Cities, and Climate Change in Latin Ahmed, S.A., Diffenbaugh, N.S., and Hertel, T.W. (2009). “Climate America.” The Hamburg Conference: Actions for Climate-Induced Volatility Deepens Poverty Vulnerability in Developing Countries.” Migration Hamburg, 16–18 July 2013, Center for International Earth Environmental Research Letters, 4(3), 034004. Science Information Network. Ainsworth, E.A., Leakey, A.D.B., Ort, Donald R., and Long, S.P. (2008). Adano, W.R., Dietz, T., Witsenburg, K., and Zaal, F. (2012). “Climate “FACE-ing the Facts: Inconsistencies and Interdependence among Change, Violent Conflict, and Local Institutions in Kenya’s Drylands.” Field, Chamber, and Modeling Studies of Elevated CO2 Impacts on Journal of Peace Research, 49(1), 65–80. Crop Yield and Food Supply.” New Phytologist, 179, 5–9. Adeel, Z., Safriel, U., Niemeijer, D., White, R., de Kalbermatten, G., Aiyyer, A., and Thorncroft, C. (2011). “Interannual-to-Multidecadal Vari- Glantz, M., Salem, B., Scholes, B., Niamir-Fuller, M., Ehui, S., and Yapi- ability of Vertical Shear and Tropical Cyclone Activity.” Journal of Gnaore, V. (2005). “Ecosystems and Human Well-Being: Desertification Climate, 24(12), 2949–62. Synthesis.” The Millennium Ecosystem Assessment 2005, edited by Akbari, S. (2011). “Dust Storms, Sources in the Middle East, and Economic J. Sarukhán and A. Whyte, World Resources Institute: Washington D.C. Model for Survey its Impacts.” Australian Journal of Basic and Applied Adger, W.N., Pulhin, J., Barnett, J., Dabelko, G.D., Hovelsrud, G.K., Sciences, 5(12), 227–33. Levy, M., Spring, Ú.O., Vogel, C., Adams, H., Hodbod, J., Kent, S. Akpan, U., Essien, M., and Isihak, S. (2013). “The Impact of Rural Elec- and Tarazona, M. (2014). “Human Security.” Climate Change 2014: trification on Rural Micro-enterprises in Niger Delta, Nigeria.” Energy Impacts, Adaptation, and Vulnerability. Contribution of Working Group for Sustainable Development, 17(5), 504–9. II to the Fifth Assessment Report of the Intergovernmental Panel on Aksakal, A., and Rehman, S. (1999). “Global Solar Radiation in Northeast- Climate Change. Chapter 12. Final Draft. Edited by C. Field, V. Barros, ern Saudi Arabia.” Renewable Energy, 17(4), 461–72. K. Mach, and M. Mastrandrea, IPCC AR5 WGII, Cambridge University Al-Bakri, J., Suleiman, A., Abdulla, F., and Ayad, J. (2011). “Potential Press: Cambridge, UK. Impact of Climate Change on Rain-fed Agriculture of a Semi-arid Aerts, J.C.J.H., Botzen, W.J.W., Emanuel, K., Lin, N., de Moel, H., Michel- Basin in Jordan.” Physics and Chemistry of the Earth, Parts A/B/C, Kerjan, E.O. (2014). “Evaluating Flood Resilience Strategies for Coastal 36(5-6), 125–34. Megacities.” Science, 344(6183), 473–475. Allan, J.A. (2001). The Middle East Water Question: Hydropolitics and the Afifi T. (2009). “EACH-FOR Environmental Change and Forced Migration Global Economy. I.B.Tauris: London, UK. Scenarios. Egypt Case Study Report. Deliverable D2.5.2.1.” UNU- Allen, R.G., Perreira, Raes, and Smith (2006). “FAO Irrigation and Drain- EHS: Bonn. age Paper (No 56).” Food and Agriculture Organization: Rome, Italy. Afifi, T., Govil, R., Sakdapolrak, P., and Warner, K. (2012). “Climate Change, Allison, E.H., Perry, A.L., Badjeck, M.C., Neil Adger, W., Brown, K., Con- Vulnerability, and Human Mobility: Perspectives of Refugees from the way, D., Halls, A.S., Pilling, G.M., Reynolds, J.D., Andrew, N.L., and East and Horn of Africa.” UNHCR, UNU-Institute for Environment and Dulvy N.K. (2009). “Vulnerability of National Economies to the Impacts Human Security, Report No. 1. of Climate Change on Fisheries,” Fish and Fisheries, 10(2), 173–96. 239 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Alston, M. (2010). “Gender and Climate Change in Australia,” Journal of Ranges.” International Centre for Integrated Mountain Development Sociology, 47(1), 53–70. (ICIMOD): Kathmandu. de Alves M. C., de Carvalho, L.G., Pozza, E.A., Sanches, L., and Maia J.C. Arndt, D.S., Baringer, M.O., Johnson, M.R., Diamond, H.J., Fogt, R.L., Levy, de S. (2011). “Ecological Zoning of Soybean Rust, Coffee Rust, and J.M., Thorne, P.W., Vincent, L.A., Watkins, A.B., and Willett, K.M. Banana Black Sigatoka based on Brazilian Climate Changes.” Procedia (2010). “State of the Climate in 2009.” 91(7). Environmental Sciences, 6, 35–49. Arnell, N.W., and Gosling, S.N. (2013),. “The Impacts of Climate Change Amengual, A., Homar, V., Romero, R., Ramis, C., and Alonso, S. (2014). on River Flow Regimes at the Global Scale.” Journal of Hydrology, “Projections for the 21st Century of the Climate Potential for Beach- 486, 351–64. based Tourism in the Mediterranean.” International Journal of Cli- Arnold, M., Mearns, R., Oshima, K., and Prasad, V. (2014). “Climate and matology, n/a–n/a. Disaster Resilience. The Role of Community-Driven Development.” Anciães, M., and Peterson, A. (2006). “Climate Change Effects on Neo- World Bank: Washington DC. tropical Manakin Diversity based on Ecological Niche Modeling.” The Artale, V., Calmanti, S., Carillo, A., Dell’Aquila, A., Herrmann, M., Condor, 108(4), 778–91. Pisacane, G., Ruti, P.M., Sannino, G., Struglia, M.V., Giorgi, F., Bi, X., Andersen, L., Lotte, L., and Verner, D. (2010). “Migration and Climate Pal, J.S., and Rauscher S. (2010). “An Atmosphere–Ocean Regional Change.” Reducing Poverty, Protecting Livelihoods, and Building Climate Model for the Mediterranean Area: Assessment of a Present Assets in a Changing Climate Social Implications of Climate Change Climate Simulation.” Climate Dynamics, 35(5), 721–40. for Latin America and the Caribbean, edited by D. Verner, World Arumí, J.L., Riviera, D., Holzapfel, E., and Muñoz, E. (2013). “Effect of Bank: Washington DC. Drought on Groundwater in a Chilean Irrigated Valley.” Water Man- Anfuso, G., Nachite, D., Jones, A., and Phillips M. (2011). “Climate Change agement, 166(WM5), 231–41. and the Mediterranean Southern Coasts.” Disappearing Destinations: Asian Development Bank (2011). “Climate Change and Migration in Climate Change and Future Challenges for Coastal Tourism, edited by Asia and the Pacific.” Asian Development Bank: Mandaluyong City, A. Jones and M. Phillips, CABI: Preston, United Kingdom. Manilla, Philippines. Anisimov O. (2007). “Potential Feedback of Thawing Permafrost to the Asner, G.P., Knapp, D.E., Broadbent, E.N., Oliveira, P.J.C., Keller, M., Global Climate System through Methane Emission.” Environmental and Silva, J. N. (2005). “Selective Logging in the Brazilian Amazon.” Research Letters, 2(4), 045016. Science 310 (5747), 480-482. Anisimov, O., and Reneva, S. (2006), “Permafrost and Changing Climate: Assad, E., Pinto, H., Nassar, A., Harfuch, L., Freitas, S., Farinelli, B., Lun- The Russian Perspective.” Ambio, 35(4), 169–75. dell, M., Fernandes, E.C.M., and LCSAR (2013). “Impacts of Climate Anzidei, M., Antonioli, F., Lambeck, K., Benini, A., Soussi, M., and Change on Brazilian Agriculture.” World Bank: Washington DC. Lakhdar R. (2011), “New Insights on the Relative Sea-level Change Asseng, S., Ewert, F., and Rosenzweig, C. (2013). “Uncertainty in Simulat- during Holocene along the Coasts of Tunisia and Western Libya from ing Wheat Yields under Climate Change.” Nature Climate Change, Archaeological and Geomorphological Markers.” Quaternary Inter- 3(9), 627–32. national, 232(1–2), 5–12. Assunção, J., and Feres, F. (2009). “Climate Change, Agricultural Produc- Arab Monetary Fund (2010). “Annual Report 2010,” AMF: Abu Dhabi, tivity, and Poverty.” World Bank: Rio de Janeiro, Brazil. United Arab Emirates. Aus der Beek, T., Voß, F., and Flörke, M. (2011). “Modelling the Impact Aragão, L.E.O.C., Malhi, Y., Roman-Cuesta, R.M., Saatchi, S., Ander- of Global Change on the Hydrological System of the Aral Sea Basin.” son, L.O., and Shimabukuro, Y.E. (2007). “Spatial Patterns and Fire Physics and Chemistry of the Earth, Parts A/B/C, 36(13), 684–95. Response of Recent Amazonian Droughts.” Geophysical Research Avnery, S., Mauzerall, D.L., Liu, J., and Horowitz, L.W. (2011). “Global Crop Letters, 34(7), L07701. Yield Reductions due to Surface Ozone Exposure: 2. Year 2030 Potential Araújo, J.C., and Knight, D.W. (2005). “Assessment of Sediment Yield Crop Production Losses and Economic Damage under Two Scenarios of Watersheds by Reservoir Survey and Simulation Modelling in of O3 Pollution.” Atmospheric Environment, 45(13), 2297–2309. Brazilian Semi-arid Environment.” Geomorphological Processes and Azad, A.K., Hossain, K.M. and Nasreen, M. (2014). “Flood-induced Vulner- Human Impact in River Basins, edited by R. Batalla, IAHS Press: abilities and Problems Encountered by Women in Northern Bangla- Wallingford, UK. desh.” International Journal of Disaster Risk Science, 4(4), 190–99. Arendt, A., Bolch, T., Cogley, J.G., Gardner, A., Hagen, J.-O., Hock, Bagley, J.E., Desai, A.R., Harding, K.J., Snyder, P.K., and Foley, J.A. (2014). ´ L., Andre- R.G., Kaser, W.T., Pfeffer, G., Moholdt, F., Paul, V., Radic, “Drought and Deforestation: Has Land Cover Change Influenced Recent assen, S., Bajracharya, M., Beedle, E., Berthier, R., Bhambri, A., Precipitation Extremes in the Amazon?” Journal of Climate, 27, 345–61. Bliss, I., Brown, E., Burgess, D., Burgess, F., Cawkwell, T., Chinn, L., Baker, J.L. (2012). “Climate Change, Disaster Risk, and the Urban Poor: Copland, B., Davies, H., De Angelis, E., Dolgova, K., Filbert, R.F., and Cities Building Resilience for a Changing World.” Urban Development Zheltyhina, N. (2012). “Randolph Glacier Inventory [v2.0]: A Dataset Series, World Bank: Washington, DC. of Global Glacier Outlines.” Global Land Ice Measurements from Space Bakun, A., (1990). “Global Climate Change and Intensification of Coastal (GLIMS), Digital Media: Boulder, CO. Ocean Upwelling.” Science (New York, N.Y.), 247(4939), 198–201. Arent D., Tol, R.S.J., Faust, E., Hella, J.P., Kumar, S., Strzepek, K.M., Bakun, A., Field, D.B., Redondo-Rodriguez, A., and Weeks, S.J. (2010). Toth, F.L., and Yan, D. (2014). “Key Economic Sectors and Services.” “Greenhouse Gas, Upwelling-favorable Winds, and the Future of Coastal Climate Change 2014: Impacts, Adaptation, and Vulnerability. Con- Ocean Upwelling Ecosystems.” Global Change Biology, 16(4), 1213–28. tribution of Working Group II to the Fifth Assessment Report of the Bale, J.S., Masters, G.J., Hodkinson, I.D., Awmack, C., Bezemer, T.M., Intergovernmental Panel on Climate Change. Chapter 10, edited by Brown, V.K., Butterfield, J., Buse, A., Coulson, J.C., Farrar, J., A. Abdulla, H. Kheshgi, and H. Xu, IPCC AR5 WGII, Cambridge Good, J.E.G., Harrington, R., Hartley, S., Jones, T.H., Lindroth, R.L., University Press: Cambridge, UK and New York, NY. Press, M.C., Symrnioudis, I., Watt, A.D., and Whittaker, J.B. (2002). Arevalo-Herrera, M., Quinones, M.L., Guerra, C., Cespedes, N., Giron, S., “Herbivory in Global Climate Change Research: Direct Effects of Rising Ahumada, M., Pineros, J.G., Padilla, N., Terrientes, Z., Rosas, A., Temperature on Insect Herbivores.” Global Change Biology, 8(1), 1–16. Padilla, J.C., Escalante, A.A., Beier, J.C., and Herrera, S. (2012). Ballester, J., Robine, J.-M., Herrmann, F.R., and Rodo, X. (2011). “Long- “Malaria in Selected non-Amazonian Countries of Latin America,” term Projections and Acclimatization Scenarios of Temperature-related Acta Tropica, 121, 303–14. Mortality in Europe,” Nature Communications, 2(358). Armstrong R.L. (2010). “The Glaciers of the Hindu Kush-Himalayan Region: Balmaseda, M.A., Trenberth, K.E., and Källén, E. (2013). “Distinctive A Summary of the Science Regarding Glacier Melt/Retreat in the Climate Signals in Reanalysis of Global Ocean Heat Content.” Geo- Himalayan, Hindu Kush, Karakoram, Pamir, and Tien Shan Mountain physical Research Letters, 40(February), 1754–59. 240 B I BLIOGRA P HY Balshi, M.S., McGuire, A.D., Zhuang, Q., Melillo, J., Kicklighter, D.W., Productivity Across Alaska Consistent with Biome Shift.” Ecology Kasischke, E., Wirth, C., Flannigan, M., Harden, J., Clein, J.S., Burn- letters, 14(4), 373–79. side, T.J., McAllister, J., Kurz, W.A., Apps, M., and Shvidenko, A. Beck, U. (2010). “Remapping Social Inequalities in an Age of Climate Change: (2007). “The Role of Historical Fire Disturbance in the Carbon Dynam- For a Cosmopolitan Renewal of Sociology.” Global Networks, 2, 165–81. ics of the Pan-boreal Region: A Process-based Analysis.” Journal of Becker, A., Inoue, S., Fischer, M., and Schwegler, B. (2012), “Climate Geophysical Research, 112(G2), G02029. Change Impacts on International Seaports: Knowledge, Perceptions, Bamber, J., van den Broeke, M., Ettema, J., Lenaerts, J., and Rignot, E. and Planning Efforts among Port Administrators.” Climatic Change, (2012). “Recent Large Increases in Freshwater Fluxes from Greenland 110(1–2), 5–29. into the North Atlantic.” Geophysical Research Letters, 39(19), 8–11. Becker, A.H., Acciaro, M., Asariotis, R., Cabrera, E., Cretegny, L., Crist, P., Bamber, J.L., and Aspinall, W.P. (2013). “An Expert Judgment Assess- Esteban, M., Mather, A., Messner, S., Naruse, S., Ng, A.K.Y., Rahm- ment of Future Sea-level Rise from the Ice Sheets.” Nature Climate storf, S., Savonis, M., Song, D.-W., Stenek, V., and Velegrakis, A.F. Change, 2(12), 1–4. (2013). “A Note on Climate Change Adaptation for Seaports: A Bamber, J.L., Riva, R.E.M., Vermeersen, B.L.A., and LeBrocq, A.M. (2009). Challenge for Global Ports, a Challenge for Global Society.” Climatic “Reassessment of the Potential Sea-Level Rise from a Collapse of the Change, 120(4), 683–95. West Antarctic Ice Sheet.” Science, 324(5929), 901–3. Béguin, A., Hales, S., Rocklöv, J., Åström, C., Louis, V.R., and Sauerborn, R. Bankoff, G. (2001). “Rendering the World Unsafe: ‘Vulnerability’ as Western (2011). “The Opposing Effects of Climate Change and Socioeconomic Discourse.” Disasters, 25(1), 19–35. Development on the Global Distribution of Malaria.” Global Environ- Baraer, M., Mark, B., McKenzie, J., Condom, T., Bury, J., Huh, K.-I., Por- mental Change, 21(4), 1209–14. tocarrero, C., Gomez, J. and Rathay, S. (2012). “Glacier Recession and Behrenfeld, M.J., O’Malley, R.T., Siegel, D. A, McClain, C.R., Sarmiento, J.L., Water Resources in Peru’s Cordillera Blanca.” Journal of Glaciology, Feldman, G.C., Milligan, A.J., Falkowski, P.G., Letelier, R.M., and 58(207), 134–50. Boss, E.S. (2006). “Climate-driven Trends in Contemporary Ocean Barber, V.A., Juday, G.P., and Finney, B.P. (2000). “Reduced Growth of Productivity.” Nature, 444(7120), 752–55. Alaskan White Spruce in the Twentieth Century from Temperature- Belaroui, K., Djediai, H., and Megdad, H. (2013). “The Influence of Soil, induced Drought Stress,” Nature, 405(6787), 668–73. Hydrology, Vegetation and Climate on Desertification in El-Bayadh Barbieri, S.A.F., Domingues, E., Queiroz, B.L., Ruiz, R.M., Carvalho, J.A., Region (Algeria).” Desalination and Water Treatment, 1–7. and Resende, M. (2010). “Climate Change and Population Migration Bellard, C., Bertelsmeier, C., Leadley, P., Thuiller, W., and Courchamp, F. in Brazil’s Northeast: Scenarios for 2025–2050.” Population and (2012). “Impacts of Climate Change on the Future of Biodiversity.” Environment, 31(5), 344–70. Ecology letters, 365–77. Barclay. E. (2008). “Is Climate Change Affecting Dengue in the Americas?” Bellard, C., Leclerc, C., and Courchamp, F. (2014). “Impact of Sea-level The Lancet, 371, 973–74. Rise on the 10 Insular Biodiversity Hotspots.” Global Ecology and Bardsley, D.K., and Hugo, G.J. (2010). “Migration and Climate Change: Biogeography, 23(2), 203–12. Examining Thresholds of Change to Guide Effective Adaptation Deci- Bellon, M.R., Hodson, D., and Hellin, J. (2011). “Assessing the Vulnerabil- sion Making.” Population and Environment, 32(2–3), 238–62. ity of Traditional Maize Seed Systems in Mexico to Climate Change.” Barnett, J., and Webber, M. (2010). “Accommodating Migration to Promote Proceedings of the National Academy of Sciences of the United States Adaptation to Climate Change.” World Bank: Washington DC. of America, 108(33), 13432–37. Barnett, T.P., Adam, J.C., and Lettenmaier, D.P. (2005), “Potential Impacts Ben-Ahmed, K., Aoun, K., Jeddi, F., Ghrab, J., El-Aroui, M.-A., and Bourat- of a Warming Climate on Water Availability in Snow-dominated bine, A. (2009). “Visceral Leishmaniasis in Tunisia: Spatial Distribution Regions.” Nature, 438(7066), 303–9. and Association with Climatic Factors.” American Journal of Tropical Barros, V. (2005). “Global Climate Change and the Coastal Areas of the Medicine and Hygiene, 81(1), 40–45. Río de la Plata.” Washington, DC. Bender, M.A., Knutson, T.R., Tuleya, R.E., Sirutis, J.J., Vecchi, G. a, Gar- Barros, V., Menéndez, A., Natenzon, C., Kokot, R., Codignotto, J., Re, M., ner, S.T., and Held, I.M. (2010). “Modeled Impact of Anthropogenic Bronstein, P., Camilloni, I., Ludueña, S., Riós, D., and González, S. (2008). Warming on the Frequency of Intense Atlantic Hurricanes.” Science “Storm Surges, Rising Seas and Flood Risks in Metropolitan Buenos (New York, N.Y.), 327(5964), 454–58. Aires.” Climate Change and Vulnerability, edited by N. Leary, C. Conde, Benjaminsen, T. A., Alinon, K., Buhaug, H., and Buseth, J.T. (2012). “Does J. Kulkarni, A. Nyong, and J. Pulhin, Earthscan: London, Sterling, VA. Climate Change Drive Land-use Conflicts in the Sahel?” Journal of Bartlett S. (2008). “Climate Change and Urban Children: Impacts and Peace Research, 49(1), 97–111. Implications for Adaptation in Low and Middle Income Countries.” Berg, A., De Noblet-Ducoudré, N., Sultan, B., Lengaigne, M., and Guimber- Human Settlements Discussion Paper, Climate Change 2, IIED, London. teau, M. (2013). “Projections of Climate Change Impacts on Potential Battisti D., and Naylor, R. (2009). “Historical Warnings of Future Food C4 Crop Productivity over Tropical Regions.” Agricultural and Forest Insecurity with Unprecedented Seasonal Heat.” Science, 323(Janu- Meteorology, 170, 89–102. ary), 240–44. Bergholt, D., and Lujala, P. (2012). “Climate-related Natural Disasters, Bayram, M.H., and Öztürk, A. (2014). “Global Climate Change, Desertifi- Economic Growth, and Armed Civil Conflict.” Journal of Peace cation, and Its Consequences in Turkey and the Middle East.” Global Research, 49(1), 147–62. Climate Change and Public Health. Chapter 17., edited by K. E. Berman, R.J., Quinn, C.H., Paavola, J., and No, W.P. (2014). “The Impact Pinkerton and W. N. Rom, Springer: New York. of Climatic Hazards on Social Network Structure : Insights from Beaumont, L.J., Pitman, A., Perkins, S., Zimmermann, N.E., and Yoc- Community Support Networks in Western Uganda Centre for Climate coz, N.G. (2010). “Impacts of Climate Change on the World’s Most Change Economics and Policy.” Centre forWorking paper No. 179/ 65, Exceptional Ecoregions.” PNAS, 108(6), 2306–2311 Climate Change Economics and Policy/ Sustainabiliity Research Bebber, D.P., Holmes, T., and Gurr, S.J. (2014). “The Global Spread of Crop Institute, London/Leeds. Pests and Pathogens.” Global Ecology and Biogeography. Bernauer, T., Böhmelt, T., and Koubi, V. (2012). “Environmental Changes Beck, P., and Goetz S.J. (2012). “Corrigendum: Satellite Observations of and Violent Conflict.” Environmental Research Letters, 7(1), 015601. High Northern Latitude Vegetation Productivity Changes between Bernauer, T., and Siegfried, T. (2012). “Climate Change and International 1982 and 2008: Ecological Variability and Regional Differences.” Water Conflict in Central Asia.” Journal of Peace Research, 49(1), 227–39. Environmental Research Letters, 7(2), 029501. Berner, L.T., Beck, P.S.A., Bunn, A.G., and Goetz, S.J. (2013). “Plant Beck, P., Juday, G.P., Alix, C., Barber, V.A., Winslow, S.E., Sousa, E.E., Response to Climate Change along the Forest-Tundra Ecotone in Heiser, P., Herriges, J.D., and Goetz, S.J. (2011). “Changes in Forest Northeastern Siberia.” Global change biology, 19(11), 3449–62. 241 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Berner, L.T., Beck, P.S.A., Bunn, A.G., Lloyd, A.H., and Goetz, S.J. (2011). Large Marine Ecosystems.” Philosophical transactions of the Royal “High-latitude Tree Growth and satellite Vegetation Indices: Correla- Society of London. Series B, Biological sciences, 367(1605), 2979–89. tions and Trends in Russia and Canada (1982–2008).” Journal of Blanford, G.J., Kriegler, E., and Tavoni, M. (2014). “Harmonization vs. Geophysical Research: Biogeosciences, 116(G1). Fragmentation: Overview of Climate Policy Scenarios in EMF27,” Bernie, D., Lowe, J., Tyrell, T., and Legge, O. (2010). “Influence of Climatic Change, 123(3-4), 383–96. mitigation policy on ocean acidification”. GRL Vol 37, L15704, Bliss, A., Hock, R., and Radic, ´ V. (2014). “Global Response of Glacier Run- doi:10.1029/2010GL043181. off to Twenty-first Century Climate Change.” Journal of Geophysical Berry, H.L., Bowen, K., and Kjellstrom, T. (2010). “Climate Change and Research: Earth Surface, accepted J. Mental Health: a Causal Pathways Framework.” International journal Bloom, A.J., Burger, M., Rubio Asensio, J.S., and Cousins, A.B. (2010), of public health, 55(2), 123–32. “Carbon Dioxide Enrichment Inhibits Nitrate Assimilation in Wheat Von Bertrab, É., and Wester, P. (2005). “Gobernabilidad del agua en México and Arabidopsis.” Science (New York, N.Y.), 328(5980), 899–903. . La Crisis de Agua de Guadalajara y el Destino del Lago de Chapala.” BMU and WHO-Europe (2009). “Protecting Health from Climate Change.” Medio Ambiente y Urbanización, 62(1–2), 143–60. German Federal Ministry for the Environment, Nature Conservation Betts, R. (2000). “Offset of the Potential Carbon Sink from Boreal Foresta- and Nuclear Safety, The World Health Organisation Europe. tion by Decreases in Surface Albedo.” Nature, 408(6809), 187–90. Boberg, F., and Christensen, J.H. (2012). “Overestimation of Mediterranean Betts R.A., Cox P.M., Collins M., Harris P.P., Huntingford C. and Jones Summer Temperature Projections Due to Model Deficiencies.” Nature C.D. (2004), “The role of Ecosystem-Atmosphere Interactions in Climate Change, 2(6), 433–36. Simulated Amazonian Precipitation Decrease and Forest Dieback Bogardi, J.J. (2008). “Impact of Gradual Environmental Change on Migration: under Global Climate Warming.” Theoretical and Applied Climatol- A Global Perspective.” Expert Seminar: Migration and Environment, ogy, 78(1–3), 157–75. International Organization for Migration. Biddulph R. (2012). “REDD and Poverty in Cambodia.” FOCALI: Gothenburg. Bohannon, J. (2010). “The Nile Delta’s Sinking Future.” Science, 327(5972), Bigano, A., Bosello, F., Roson, R., and Tol, R. (2008). “Economy-wide 1444–47. Estimates of the Implications of Climate Change: A Joint Analysis for Boko, M., Niang, I., Nyong, A., Vogel, C., Githeko, A., Medany, M., Osman- Sea-level Rise and Tourism.” Mitigation and Adaptation Strategies for Elasha, B., Tabo, R., and Yanda, P. (2007). “Africa. Climate Change Global Change, 13(8), 765–91. 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Bindoff, N.L., Stott, P.A., AchutaRao, K.M., Allen, M.R., Gillett, N., Gutzler, Group II to the Fourth Assessment Report of the Intergovernmental D., Hansingo, K., Hegerl, G., Hu, Y., Jain, S., Mokhov, I., Overland, Panel on Climate Change.” Edited by M. L. Parry, O. F. Canziani, J., Perlwitz, J., Sebbari, R. and Zhang, X. (2013). “Detection and J. P. Palutikof, P. J. van der Linden, and C. E. Hanson, IPCC AR4 WGII, Attribution of Climate Change: from Global to Regional.” in Climate Cambridge University Press: Cambridge UK, 433–467. Change 2013: The Physical Science Basis. Contribution of Working Bolch, T., Peters, J., Yegorov, A., Pradhan, B., Buchroithner, M., and Group I to the Fifth Assessment Report of the Intergovernmental Panel Blagoveshchensky, V. (2011). “Identification of Potentially Dangerous on Climate Change. Chapter 10, edited by T. F. Stocker, D. Qin, G.-K. Glacial Lakes in the Northern Tien Shan.” Natural Hazards, 59(3), Plattner, M. Tignor, S. K. Allen, J. Boschung, A. Nauels, Y. Xia, V. 1691–1714. Bex, and P. M. Midgley, IPCC AR5 WGI, Cambridge University Press: Bonan, G.B. (2008). “Forests and Climate Change: Forcings, Feedbacks, Cambridge, United Kingdom and New York, NY. and the Climate Benefits of Forests.” Science (New York, N.Y.), Bindschadler, R.A., Nowicki, S., Abe-OUCHI, A., Aschwanden, A., Choi, H., 320(5882), 1444–49. Fastook, J., Granzow, G., Greve, R., Gutowski, G., Herzfeld, U., Jack- Bond, W.J., and Midgley, G.F. (2012). “Carbon Dioxide and the Uneasy son, C., Johnson, J., Khroulev, C., Levermann, A., Lipscomb, W.H., Interactions of Trees and Savannah Grasses.” Philosophical transac- Martin, M. A., Morlighem, M., Parizek, B.R., Pollard, D., Price, S.F., tions of the Royal Society of London. Series B, Biological sciences, Ren, D., Saito, F., Sato, T., Seddik, H., Seroussi, H., Takahashi, K., 367(1588), 601–12. Walker, R., and Wang, W.L. (2013). “Ice-sheet Model Sensitivities to Bouchaou, L., Michelot, J.L., Vengosh, A., Hsissou, Y., Qurtobi, M., Gaye, Environmental Forcing and Their Use in Projecting Future Sea Level C.B., Bullen, T.D., and Zuppi, G.M. (2008). “Application of Multiple (the SeaRISE Project).” Journal of Glaciology, 59(214), 195–224. Isotopic and Geochemical Tracers for Investigation of Recharge, Sali- Bishop, M.L., and Payne, A. (2012). “Climate Change and the Future of nization, and Residence Time of Water in the Souss–Massa Aquifer, Caribbean Development.” Journal of Development Studies, 48(10), Southwest of Morocco.” Journal of Hydrology, 352(3), 267–87. 1536–53. Bozkurt, D. and Sen, O.L. (2013). “Climate Change Impacts in the Euphrates– Bizzarri M. (2012). “Gender-based Violence and Natural Disasters in Tigris Basin based on Different Model and Scenario Simulations.” Latin America and the Caribbean.” United Nations Population Fund Journal of Hydrology, 480, 149–61. (UNFPA): New York. Bradshaw, S., and Fordham, M. (2013). “Women, Girls, and Disasters: A Black, R., Adger, N., Arnell, N., Dercon, S., Geddes, A. and Thomas, Review for DFID.” Department for International Development: London. D. (2011). “Migration and Global Environmental Change: Future Brando, P.M., Balch, J.K., Nepstad, D.C., Morton, D.C., Putz, F.E., Coe, M.T., Challenges and Opportunities.” UK Government Office for Science: Silvério, D., Macedo, M.N., Davidson, E. A, Nóbrega, C.C., Alencar, A., Foresight Project. and Soares-Filho, B.S. (2014). “Abrupt Increases in Amazonian Tree Black, R., Arnell, N., Adger, W.N., Thomas, D. and Geddes, A. (2013). Mortality due to Drought-Fire Interactions.” Proceedings of the National “Migration, Immobility, and Displacement Outcomes Following Extreme Academy of Sciences of the United States of America, 111(17), 6347–52. Events.” Environmental Science & Policy, 27, Supple, S27–S43. Brando, P.M., Goetz, S.J., Baccini, A., Nepstad, D.C., Beck, P.S.A., and Black, R., Bennett, S.R.G., Thomas, S.M., and Beddington, J.R. (2011). Christman, M.C. (2010). “Seasonal and inter-annual Variability of “Climate Change: Migration as Adaptation.” Nature, 478(7370), Climate and Vegetation Indices across the Amazon.” Proceedings of 447–49. the National Academy of Sciences of the United States of America, Black, R., Kniveton, D., Skeldon, R., Coppard, D., Murata, A., and Schmidt- 107(33), 14685–90. Verkerk, K. (2008). “Demographics and Climate Change : Future Trends Brando, P.M., Nepstad, D.C., Balch, J.K., Bolker, B., Christman, M.C., And their Policy Implications for Migration.” Development Research Coe, M., and Putz, F.E. (2012). “Fire-induced Tree Mortality in a Centre on Migration, Globalisation and Poverty: Brighton, UK. Neotropical Forest: The Roles of Bark Traits, Tree Size, Wood Density Blanchard. J.L., Jennings. S., Holmes. R., Harle. J., Merino. G., Allen. J.I., and Fire Behavior.” Global Change Biology, 18(2), 630–41. Holt. J., Dulvy, N.K., and Barange, M. (2012). “Potential Consequences Brando, P.M., Nepstad, D.C., Davidson, E.A., Trumbore, S.E., Ray, D., and of Climate Change for Primary Production and Fish Production in Camargo, P. (2008). “Drought Effects on Litterfall, Wood Production 242 B I BLIOGRA P HY and Belowground Carbon Cycling in an Amazon Forest: Results of Buhaug, H. (2010). “Climate Not to Blame for African Civil Wars.” Pro- a Throughfall Reduction Experiment.” Philosophical Transactions of ceedings of the National Academy of Sciences of the United States of the Royal Society of London. Series B, Biological sciences, 363(1498), America, 107(38), 16477–82. 1839–48. Buhaug, H. (2014). “Concealing Agreements over climate-Conflict Results.” Bravi, A., and Solbrandt, N. (2012). “Food Security Challenges Facing Proceedings of the National Academy of Sciences , 111 (6 ), E636–E636. Central Asia.” UNDP Regional Bureau for Europe and the CIS Buhaug, H., Gleditsch, N.P., and Theisen, O.M. (2010), “Implications of Bravo, J.M., Collischonn, W., Paz, A.R., Allasia, D. and Domecq, F. (2013). Climate Change for Armed Conflict.” Social Dimensions of Climate “Impact of Projected Climate Change on Hydrologic Regime of the Change. Equity and Vulnerability in a Warming World, edited by Upper Paraguay River Basin.” Climatic Change. R. Mearns and A. Norton, World Bank: Washington, DC. Brecht, H., Dasgupta, S., Laplante, B., Murray, S., and Wheeler, D. (2012). Bunn, A.G., and Goetz, S.J. (2006). “Trends in Satellite-Observed Circum- “Sea-level Rise and Storm Surges: High Stakes for a Small Number of polar Photosynthetic Activity from 1982 to 2003: The Influence of Developing Countries.” The Journal of Environment & Development, Seasonality, Cover Type, and Vegetation Density.” Earth Interactions, 21(1), 120–38. 10(12), 1–19. Brinkmann K., Dickhoefer U., Schlecht E. and Buerkert A. (2011). “Quan- Burke, L., Maidens, J., Spalding, M., Kramer, P., Green, E., Greenhalgh, tification of Aboveground Rangeland Productivity and Anthropogenic S., Nobles, H., and Kool, J. (2004). Reefs at Risk in the Caribbean. Degradation on the Arabian Peninsula using Landsat Imagery and Field World Resources Institute: Washington, DC. Inventory Data.” Remote Sensing of Environment, 115(2), 465–74. Burke, M.B., Miguel, E., Satyanath, S., Dykema, J.A., and Lobell, D.B. Brody, A., Demetriades, J., Esplen, E., and Britain, G. (2008). “Gender (2009). “Warming Increases the Risk of Civil War in Africa.” Pro- and Climate Change: Mapping the Linkages: A Scoping Study on ceedings of the National Academy of Sciences of the United States of Knowledge and Gaps.” 44(June). America, 106(49), 20670–74. Bromirski, P.D., Miller, A.J., Flick, R.E., and Auad, G. (2011). “Dynamical Bury, J.T., Mark, B.G., McKenzie, J.M., French, A., Baraer, M., Huh, K.I., Suppression of Sea-level Rise along the Pacific Coast of North America: Zapata Luyo, M.A., and Gómez López, R.J. (2011). “Glacier Recession Indications for Imminent Acceleration.” Journal of Geophysical and Human Vulnerability in the Yanamarey Watershed of the Cordillera Research, 116(C7), C07005. Blanca, Peru.” Climatic Change, 105(1–2), 179–206. Brown, D., Chanakira, R.R., Chatiza, K., Dhliwayo, M., Dodman, D., and Buxton, N., Escobar, M., Purkey, D., and Lima, N. (2013). “Water Scarcity, Masiiwa, M. (2012). “Climate Change Impacts , Vulnerability, and Adap- Climate Change and Bolivia: Planning for Climate Uncertainties.” tation in Zimbabwe.” IIED, Climate Change WORKING PAPER No.3. Stockholm Environment Institute: Davis, U.S. Brown, D.P., Beven, J.L., Franklin, J.L., and Blake, E.S. (2010). “Atlantic Hur- Buytaert, W., and De Bièvre, B. (2012). “Water for Cities: The Impact of ricane Season of 2008.” Monthly Weather Review, 138(5), 1975–2001. Climate Change and Demographic Growth in the Tropical Andes.” Brown, G. (2012), “Out of Eedlock, Into School. Combating Child Marriage Water Resources Research, 48(8), 1–13. through Education.” Office of Gordon and Sarah Brown: London. Buytaert, W., Célleri, R., De Bièvre, B., Cisneros, F., Wyseure, G., Deckers, Brown, O., and Crawford, A. (2009). “Rising Temperatures, Rising Ten- J., and Hofstede, R. (2006). “Human impact on the Hydrology of the sions. Climate Change and the Risk of Violent Conflict in the Middle Andean Páramos.” Earth-Science Reviews, 79(1–2), 53–72. East.” International Institute for Sustainable Development: Winnipeg. Buytaert, W., Célleri, R., and Timbe, L. (2009), “Predicting Climate Change Brown, S., Kebede, A.S., and Nicholls, R.J. (2009). “Sea-Level Rise and Impacts on Water Resources in the Tropical Andes: Effects of GCM Impacts in Africa, 2000 to 2100.” Unpublished Report to Stockholm Uncertainty.” Geophysical Research Letters, 36(7), 1–5. Environment Institute, Oxford. Cai, W., Borlace, S., Lengaigne, M., van Rensch, P., Collins, M., Vecchi, G., Brown, S., Kebede, A.S. and Nicholls, R.J. (2011). “Sea-Level Rise and Timmermann, A., Santoso, A., McPhaden, M.J., Wu, L., England, M.H., Impacts in Africa, 2000 to 2100.” Southampton, UK. Wang, G., Guilyardi, E., and Jin, F.-F. (2014). “Increasing Frequency Brücker, P., Bougnoux, N., and Wodon, Q. (2012). “Policy Brief en Afrique of Extreme El Niño Events due to Greenhouse Warming.” Nature du Nord.” Climat, (13), 1–4. Climate Change, 5(1), 1–6. Bruckner, T., Bashmakov, I.A., and Mulugetta, Y. (2014). “Energy Calafat, F.M., Chambers, D.P., and Tsimplis, M.N. (2012). “Mechanisms Systems.” Climate Change 2014: Mitigation of Climate Change. of Decadal Sea Level Variability in the Eastern North Atlantic and the Contribution of Working Group III to the Fifth Assessment Report of Mediterranean Sea.” Journal of Geophysical Research: Oceans, 117(C9). the Intergovernmental Panel on Climate Change. Chapter 7, edited Caldeira, K. (2013). “Coral Bleaching: Coral ‘Refugia’ Amid Heating Seas.” by K. Parikh and J. Skea, IPCC AR5 WGIII, Cambridge University Nature Climate Change, 3(5), 444–45. Press: Cambridge. Caldeira, K., and Wickett, M.E. (2003). “Anthropogenic Carbon and Ocean Brun, C.J.A. (2007). “Adapting to Impacts of Climate Change on Water Ph.” Nature, 425(September), 365. Supply in Mexico City.” Human Development Report 2007/2008, Callaghan, T. V., Johansson, M., Brown, R.D., Groisman, P.Y., Labba, N., No. 2007/42, Human Development Report Office, UNDP. Radionov, V., Barry, R.G., Bulygina, O.N., Essery, R.L.H., Frolov, D.M., Bucknall, J. (2007). Making the Most of Scarcity: Accountability for Bet- Golubev, V.N., Grenfell, T.C., Petrushina, M.N., Razuvaev, V.N., Rob- ter Water Management Results in the Middle East and North Africa. inson, D.A., Romanov, P., Shindell, D., Shmakin, A.B., Sokratov, S.A., World Bank Publications: Washington DC. Warren, S., and Yang, D. (2011). “The Changing Face of Arctic Snow Buddemeier, R.W., Lane, D.R., and Martinich, J. A. (2011). “Modeling Cover: A Synthesis of Observed and Projected Changes,” Ambio, Regional Coral Reef Responses to Global Warming and Changes in 40(S1), 17–31. Ocean Chemistry: Caribbean Case Study.” Climatic Change, 109(3–4), Calvo, J.J. (2013). “Análisis de las Condiciones Sociales en América Latina 375–97. en el Contexto del Cambio Climático: Diagnóstico, Perspectivas y Budyko M.I. (1958), “The Heat Balance of the Earth’s Surface (in Rus- Lineamientos de una Política Social Verde,” Santiago, Chile. sian) Gidrometeorologicheskoe, Leningrad, 1956.” English translation Camargo, M.B.P. De (2010). “The Impact of Climatic Variability and Climate by N.A. Stepanova, Office of Technical Services, U.S. Department of Change on Arabic Coffee Crop in Brazil.” Bragantia, 69(1), 239–47. Commerce: Washington, DC. Camargo, S.J., Ting, M., and Kushnir, Y. (2012). “Influence of Local and Buechler, S. (2009). “Gender, Water and Climate Change in Sonora, Mexico: Remote SST on North Atlantic Tropical Cyclone Potential Intensity.” implications for Policies and Programmes on Agricultural Income Climate Dynamics, 40(5-6). 1515–29. Generation.” Gender and Development, 17(1), 51–66. Camilloni, I.A., Saurral, R.I., and Montroull, N.B. (2013). “Hydrological Projec- Bueno, R., Herzfeld, C., Stanton, E.A., and Ackerman, F. (2008). “The tions of Fluvial Floods in the Uruguay and Paraná Basins under Different Caribbean and Climate Change: The Costs of Inaction.” Boston, MA. Climate Change Scenarios.” Intl. J. River Basin Management, 1–11. 243 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Caminade, C., Kovats, S., Rocklov, J., Tompkins, A.M., Morse, A.P., Colon- CEPAL (2002). La Sostenibilidad del Desarrollo en América Latina y el Gonzalez, F.J., Stenlund, H., Martens, P., and Lloyd, S.J. (2014). Caribe : Desafíos y Oportunidades. United Nations Press: Santiago “Impact of Climate Change on Global Malaria Distribution” PNAS, de Chile. 111(9), 3286–91. Chaherli, N., and Nash, J. (2013). “Agricultural Exports from Latin America Caminade, C., Medlock, J.M., Ducheyne, E., McIntyre, K.M., Leach, S., and the Caribbean : Harnessing Trade to Feed the World and Promote Baylis, M., and Morse, A.P. (2012). “Suitability of European Climate Development.” for the Asian Tiger Mosquito Aedes albopictus: Recent Trends and Chakraborty, S., and Newton, A.C. (2011). “Climate Change, Plant Dis- Future Scenarios.” J. R. Soc Interface, 9, 2708–17. eases, and Food Security: An Overview.” Plant Pathology, 60(1), 2–14. Camp Alatoo (2013). “Climate Risk Assessment Guide—Central Asia.” Challinor, A.J., Watson, J., Lobell. D.B., Howden, S.M., Smith. D.R., and Almaty. Chhetri, N. (2014). “A Meta-analysis of Crop Yield under Climate Campbell, J.D., Taylor, M. A., Stephenson, T.S., Watson, R. A., and Change and Adaptation.” (March), 1–5. Whyte, F.S. (2011). “Future Climate of the Caribbean from a Regional Chambers, R., and Conway, G. (1991). “Sustainable Rural Livelihoods: Climate Model.” International Journal of Climatology, 31(12), 1866–78. Practical Concepts for the 21st Century.” Insitute of Development Canadell, J.G., Le Quéré, C., Raupach, M.R., Field, C.B., Buitenhuis, E.T., Studies: Brighton. Ciais, P., Conway, T.J., Gillett, N.P., Houghton, R.A., and Marland, G. Chapin, F.S., Sturm, M., Serreze, M.C., McFadden, J.P., Key, J.R., Lloyd, A.H., (2007). “Contributions to Accelerating Atmospheric CO2 Growth from McGuire, A.D., Rupp, T.S., Lynch, A.H., Schimel, J.P., Beringer, J., Economic Activity, Carbon Intensity, and Efficiency of Natural Sinks.” Chapman, W.L., Epstein, H.E., Euskirchen, E.S., Hinzman, L.D., Proceedings of the National Academy of Sciences of the United States Jia, G., Ping, C.L., Tape, K.D., Thompson, C.D.C., Walker, D.A., and of America, 104(47), 18866–70. Welker, J.M. (2005). “Role of Land-surface Changes in Arctic Summer Cane, M.A. (2005). “The Evolution of El Niño, Past and Future.” Earth Warming.” Science, 310(5748), 657–60. and Planetary Science Letters, 230(3–4), 227–40. Chavez F.P., and Messié, M. (2009). “A Comparison of Eastern Boundary Cannon, T. (2000). “Vulnerability Analysis and Disasters.” Floods, edited Upwelling Ecosystems.” Progress in Oceanography, 83(1–4), 80–96. by D. J. Parker, Routledge: London. Checkley, W., Epstein, L.D., Gilman, R.H., Figueroa, D., Cama, R.I., Patz, J., Cao, C., Shao, X., and Uprety, S. (2013). “Detecting Light Outages After and Black, R.E. (2000). “Effects of El Niño and Ambient Temperature Severe Storms Using the S-NPP/VIIRS Day/Night Band Radiances.” on Hospital Admissions for Diarrhoeal Diseases in Peruvian Children.” Geoscience and Remote Sensing Letters, IEEE. The Lancet, 355. Carbajo, A.E., Cardo, M. V., and Vezzani, D. (2012). “Is Temperature Chenoweth, J., Hadjinicolaou, P., Bruggeman, A., Lelieveld, J., Levin, Z., the Main Cause of Dengue in Non-endemic Countries? The Case Lange, M.A., Xoplaki, E., and Hadjikakou, M. (2011). “Impact of Cli- of Argentina.” International Journal of Health Geographics, 11(26). mate Change on the Water Resources of the Eastern Mediterranean Cardenas, R., Sandoval, M., Rodriguez-Morales, A.J., and Franco-Paredes, and Middle East Region: Modeled 21st Century Changes and Implica- C. (2006). “Impact of Climate Variability in the Occurrence of Leish- tions.” Water Resources Research, 47(6), W06506. maniasis in Northeastern Colombia.” American Journal of Tropical Chertov, O.G., Komarov, A.S., Gryazkin, A. V., Smirnov, A.P., and Bhatti, Medicine and Hygiene, 75(2), 273–77. D.S. (2013). “Simulation Modeling of the Impact of Forest Fire on the CARE (2013). “Climate Vulnerability and Capacity of Ethnic Minorities in Carbon Pool in Coniferous Forests of European Russia and Central the Northern, Mountainous Region of Vietnam.” CARE: Atlanta, GA. Canada.” Contemporary Problems of Ecology, 6(7), 727–33. Carey, M. (2005). “Living and Dying with Glaciers: People’s Historical Cheung, W.W.L., Dunne, J., Sarmiento, J.L., and Pauly, D. (2011). “Integrat- Vulnerability to Avalanches and Outburst Floods in Peru.” Global and ing Ecophysiology and Plankton Dynamics into Projected Maximum Planetary Change, 47(2–4), 122–34. Fisheries Catch Potential under Climate Change in the Northeast Carey, M., Huggel, C., Bury, J., Portocarrero, C., and Haeberli, W. (2012). Atlantic.” ICES Journal of Marine Science, 68(6), 1008–18. “An Integrated Socio-environmental Framework for Glacier Hazard Cheung, W.W.L., Lam, V.W.Y., Sarmiento, J.L., Kearney, K., Watson, R., Management and Climate Change Adaptation: Lessons from Lake 513, Zeller, D., and Pauly, D. (2010). “Large-scale Redistribution of Maxi- Cordillera Blanca, Peru.” Climatic Change, 112, 733–67. mum Fisheries Catch Potential in the Global Ocean under Climate Carius, A., and Maas, A. (2009). “Climate Change and Security Three Change.” Global Change Biology, 16(1), 24–35. Scenarios for Middle America.” Cheung, W.W.L., Watson, R., and Pauly, D. (2013). “Signature of Ocean Carneiro. J.F., Boughriba. M., Correia. A., Zarhloule. Y., Rimi. A.. and El Warming in Global Fisheries Catch.” Nature, 497(7449), 365–68. Houadi. B. (2009). “Evaluation of Climate Change Effects in a Coastal Chevallier, P., Pouyaud, B., Suarez, W., and Condom, T. (2011). “Climate Aquifer in Morocco using a Density-dependent Numerical Model.” Change Threats to Environment in the Tropical Andes: Glaciers and Environmental Earth Sciences, 61(2), 241–52. Water Resources.” Regional Environmental Change, 11(1), 179–87. Carrigan, A.D., and Puotinen, M. (2014). “Tropical Cyclone Cooling Combats Chhetri, P., Co, J., Gekara, V., Corbitt, B., Wickramasinghe, N., Jayat- Region-wide Coral Bleaching.” Global Change Biology. illeke, G., Basic, F., Scott, H., Manzoni, A., and Maddox, C. (2013). Carter, M.R., Little, P.D., Mogues, T., and Negatu, W. (2007). “Poverty “Functional Resilience of Port Environs in a Changing Climate—Assets Traps and Natural Disasters in Ethiopia and Honduras.” World Devel- and Operations.” National Climate Change Adaptation Research opment, 35(5), 835–56. Facility: Melbourne. Carty, T., and Magrath, J. (2013). “Growing Disruption. Climate Change, Chia, E.L., Somorin, O.A., Sonwa, D.J., and Tiani, A.M. (2013). “Local Food, and the Fight Against Hunger.” Oxfam International: Oxford. Vulnerability, Forest Communities, and Forest-Carbon Conservation: Cashman, A. (2013). “Water Security and Services in the Caribbean Water Case of Southern Cameroon.” International Journal of Biodiversity Security and Services in the Caribbean.” Technical Note IDB-TN-514, and Conservation, 5(August), 498–507. Inter-American Development Bank. Chou, C., Chiang, J.C.H., Lan, C.W., Chung, C.H., Liao, Y.C. and Lee, C.J. Cashman, A., Nurse, L., and Charlery, J. (2010). “Climate Change in the (2013). “Increase in the Range between Wet and Dry Season Precipita- Caribbean: The Water Management Implications,” The Journal of tion.” Nature Geoscience, 6(4), 263–67. Environment & Development, 19(1), 42–67. Christensen, J.H., Hewitson, B., Busuioc, A., Chen, A., Gao, X., Held, I., CCRIF (2010). “Enhancing the Climate Risk and Adaptation Fact Base for Jones, R., Kolli, R.K., Kwon, W.T., Laprise, R., Rueda, V.M., Mearns, L., the Caribbean.” Caribbean Catastrophe Risk Insurance Facility: Grand Menéndez, C.G., Räisänen, J., Rinke, A., Sarr, A., and Whetton, P. Cayman, Cayman Islands. (2007). “Regional Climate Projections.” Climate Change 2007: The Center for Naval Analysis (2007). “National Security and the Threat of Physical Science Basis. Contribution of Working Group I to the Fourth Climate Change.” CNA Corporation: Alexandria, Virginia. Assessment Report of the Intergovernmental Panel on Climate Change, 244 B I BLIOGRA P HY edited by S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, Cook, K.H., and Vizy, E.K. (2008). “Effects of Twenty-First-Century Climate K. B. Averyt, M. Tignor, and H. L. Miller, IPCC AR4 WGI, Cambridge Change on the Amazon Rain Forest.” Journal of Climate, 21(3), 542–60. University Press: Cambridge, UK and New York, NY. Cortés, G., Vargas, X., and McPhee, J. (2011). “Climatic Sensitivity of Christensen, J.H., Kumar, K.K., Aldrian, E., An, S.I., Cavalcanti, I.F.A., Cas- Streamflow Timing in the Extratropical Western Andes Cordillera.” tro, M. de, Dong, W., Goswami, P., Hall, A., Kanyanga, J.K., Kitoh, A., Journal of Hydrology, 405(1–2), 93–109. Kossin, J., Lau, N.C., Renwick, J., Stephenson, D.B., Xie, S.P., and Costa, L.C., Justino, F., Oliveira, L.J.C., Sediyama, G.C., Ferreira, W.P.M., and Zhou, T. (2013). “Climate Phenomena and their Relevance for Future Lemos, C.F. (2009). “Potential Forcing of CO2, Technology and Climate Regional Climate Change.” Climate Change 2013: The Physical Sci- Changes in Maize ( Zea mays ) and Bean ( Phaseolus vulgaris ) Yield ence Basis. Contribution of Working Group I to the Fifth Assessment in Southeast Brazil.” Environmental Research Letters, 4(1), 014013. Report of the Intergovernmental Panel on Climate Change, edited by Costa, M.H. and Pires, G.F. (2010). “Effects of Amazon and Central Brazil P. M. M. Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, Deforestation Scenarios on the Duration of the Dry Season in the Arc of J. Boschung, A. Nauels, Y. Xia, V. Bex, Cambridge University Press: Deforestation.” International Journal of Climatology, 30(13), 1970–79. Cambridge UK and New York, NY. Costello, A., Abbas, M., Allen, A., Ball, S., Bell, S., Bellamy, R., Friel, S., Christmann, S., Martius, C., Bedoshvili, D., Bobojonov, I., Carli, C., Groce, N., Johnson, A., Kett, M., Lee, M., Levy, C., Maslin, M., Devkota, K., Ibragimov, Z., Khalikulov, Z., Kienzler, K., Man- Mccoy, D., Mcguire, B., Montgomery, H., Napier, D., Pagel, C., Patel, J., thrithilake, H., Mavlyanova, R., Mirzabaev, A., Nishanov, N., and Antonio, J. (2009). “Managing the Health Effects of Climate Sharma, R.C., Tashpulatova, B., Toderich, K., and Turdieva, M. (2009). Change,” The Lancet, 373, 1693–1733. “Food Security and Climate Change in Central Asia and the Caucasus.” Coumou, D., and Rahmstorf, S. (2012). “A Decade of Weather Extremes.” CGIAR-PFU, ICARDA: Tashkent, Uzbekistan. Nature Climate Change, 2(7), 491–96. Church, J.A., Clark, P.U., Cazenave, A., Gregory, J.M., Jevrejeva, S., Lev- Coumou, D., and Robinson, A. (2013). “Historic and Future Increase in ermann, A., Merrifield, M.A., Milne, G.A., Nerem, R.S., Nunn, P.D., the Global Land Area affected by Monthly Heat Extremes.” Environ. Payne, A.J., Pfeffer, W.T., Stammer, D., and Unnikrishnan, A.S. (2013). Res. Lett., 8, 0–6. “Sea Level Change.” Climate Change 2013: The Physical Science Basis. Coumou, D., Robinson, A., and Rahmstorf, S. (2013). “Global Increase Contribution of Working Group I to the Fifth Assessment Report of the in Record-breaking Monthly-mean Temperatures.” Climatic Change, Intergovernmental Panel on Climate, edited by T. F. Stocker, D. Qin, 118(3-4), 771–82. G.-K. Plattner, M. Tignor, S. K. Allen, J. Boschung, A. Nauels, Y. Xia, Couto Pereira, S.N. (2010). “Payment for Environmental Services in the V. Bex, and P. M. Midgley, Cambridge University Press: Cambridge, Amazon Forest: How Can Conservation and Development Be Recon- UK, and New York, NY. ciled?” The Journal of Environment & Development, 19(2), 171–90. Church, J.A., and White, N.J. (2011). “Sea-level Rise from the Late 19th Cowie, S.M., Knippertz, P., and Marsham, J.H. (2013). “Are Vegetation- to the Early 21st Century.” Surveys in Geophysics, 32(4–5), 585–602. related Roughness Changes the Cause of the Recent Decrease in Dust CIESIN (2011). “Population, Landscape, And Climate Estimates (PLACE), Emission from the Sahel?” Geophysical Research Letters, 40(9), 1868–72. v3 (1990, 2000, 2010),” Center for International Earth Science Infor- Cowtan, K., and Way, R.G. (2014). “Coverage Bias in the HadCRUT4 mation Network (CIESIN). Temperature Series and its Impact on Recent Temperature Trends.” Colbert, A.J., Soden, B.J., Vecchi, G.A., and Kirtman, B.P. (2013). “The Quarterly Journal of the Royal Meteorological Society, Accepted. Impact of Anthropogenic Climate Change on North Atlantic Tropical Cox. P.M., Betts. R.A., Collins, M., Harris, P.P., Huntingford, C. and Jones, Cyclone Tracks.” Journal of Climate, 26(12), 4088–95. C.D. (2004). “Amazonian Forest Dieback under Climate-Carbon Cycle Collins, M., An, S.I., Cai, W., Ganachaud, A., Guilyardi, E., Jin, F.F., Projections for the 21st Century.” Theoretical and Applied Climatol- Jochum, M., Lengaigne, M., Power, S., Timmermann, A., Vecchi, G., ogy, 78(1–3), 137–56. and Wittenberg, A. (2010). “The Impact of Global Warming on the Cox, P.M., Harris, P.P., Huntingford, C., Betts, R.A., Collins, M., Jones, Tropical Pacific Ocean and El Niño.” Nature Geoscience, 3(6), 391–97. C.D., Jupp, T.E., Marengo, J.A., and Nobre, C.A. (2008). “Increasing Collins, M., Knutti, R., Arblaster, J., Dufresne, J.L., Fichefet, T., Friedlingstein, Risk of Amazonian Drought due to Decreasing Aerosol Pollution.” P., Gao, X., Gutowski, W.J., Johns, T., Krinner, G., Shongwe, M., Tebaldi, Nature, 453(7192), 212–15. C., Weaver, A.J., and Wehner, M. (2013). “Long-term Climate Change: Cox, P.M., Pearson, D., Booth, B.B., Friedlingstein, P., Huntingford, C., Projections, Commitments, and Irreversibility.” Climate Change 2013: Jones, C.D., and Luke, C.M. (2013). “Sensitivity of Tropical Carbon to The Physical Science Basis. Contribution of Working Group I to the Fifth Climate Change constrained by Carbon Dioxide Variability.” Nature, Assessment Report of the Intergovernmental Panel on Climate Change. 494(7437), 341–44. Chapter 12, edited by T. F. Stocker, D. Qin, G.-K. Plattner, M. Tignor, Cozzi, L. and Gül, T. (2013). “Redrawing the Energy-Climate Map. World S. K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex, and P. M. Midgley, Energy Outlook Special Report,” International Energy Agency: Paris, IPCC AR5 WGI, Cambridge University Press: Cambridge and New York. France. Colon-Gonzales, F.J., Fezzi, C., Lake, I.R., and Hunter, P.R. (2013). “The CPC/NCEP/NWS/IRICS (2010). “El Niño/Southern Oscillation (ENSO) Effects of Weather and Climate Change on Dengue.” PLoS Negl Trop Diagnostic Discussion.” CLIMATE PREDICTION CENTER/NCEP/ Dis, 7(11), e2503. NWS and the International Research Institute for Climate and Society, Conca, W., Al-Nuaimi, K., and Nagelkerke, N. (2010). “The Complex- (accessed June 18, 2014). Available at http://www.cpc.ncep.noaa ity of Regional Warming in the United Arab Emirates in the Period .gov/products/analysis_monitoring/enso_advisory/. 1982–2009.” Journal of Global Warming, 2(4), 225–33. Craine, J.M., Elmore, A.J., Olson, K.C., and Tolleson, D. (2010). “Climate Confalonieri, U., Menne, B., Akhtar, R., Ebi, K.L., Hauengue, M., Kovats, Change and Cattle Nutritional Stress.” Global Change Biology, 16(10), R.S., Revich, B., and Woodward, A. (2007). “Human Health.” Climate 2901–11. Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Crate, S.A. (2013). “Climate Change and Human Mobility in Indigenous Com- Working Group II to the Fourth Assessment Report of the Intergovern- munities of the Russian North.” Brookings Institution: Washington, DC. mental Panel on Climate Change, edited by M. L. Parry, O. F. Canziani, Craufurd, P.Q., and Wheeler, T.R. (2009). “Climate Change and the J. P. Palutikof, P. J. van der Linden, and C. E. Hanson, Cambridge Flowering Time of Annual Crops.” Journal of Experimental Botany, University Press: Cambridge. 60(9), 2529–39. Cook, B., Zeng, N., and Yoon, J.-H. (2012). “Will Amazonia Dry Out? Creamean, J.M., Suski, K.J., Rosenfeld, D., Cazorla, A., DeMott, P.J., Magnitude and Causes of Change from IPCC Climate Model Projec- Sullivan, R.C., White, A.B., Ralph, F.M., Minnis, P., Comstock, J.M., tions.” Earth Interactions, 16(3), 1–27. Tomlinson, J.M., and Prather K.A. (2013). “Dust and Biological Aerosols 245 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL from the Sahara and Asia Influence Precipitation in the Western U.S.” Davidson, E.A., and Janssens, I.A. (2006). “Temperature Sensitivity of Soil Science (New York, N.Y.), 339(6127), 1572–78. Carbon Decomposition and Feedbacks to Climate Change.” Nature, Cretaux, J.F., Letolle, R., and Bergé-Nguyen, M. (2013). “History of Aral 440(7081), 165–73. Sea-level Variability and Current Scientific Debates.” Global and Davie, J.C.S., Falloon, P.D., Kahana, R., Dankers, R., Betts, R., Port- Planetary Change, 110, 99–113. mann, F.T., Wisser, D., Clark, D.B., Ito, A., Masaki, Y., Nishina, K., Cruz, A., and Krausmann, E. (2013). “Vulnerability of the Oil and Gas Fekete, B., Tessler, Z., Wada, Y., Liu, X., Tang, Q., Hagemann, S., Sector to Climate Change and Extreme Weather Events.” Climatic Stacke, T., Pavlick, R., Schaphoff, S., Gosling, S.N., Franssen, W., Change, 121(1), 41–53. and Arnell, N. (2013). “Comparing Projections of Future Changes in Curry, J., Jelinek, M., Foskey, B., Suzuki, A., and Webster, P. (2009). Runoff from Hydrological and Biome Models in ISI-MIP.” Earth System “Potential Economic Impacts of Hurricanes in Mexico, Central America, Dynamics, 4(2), 359–74. and the Caribbean ca. 2020–2025.” LCR Sustainable Development Davletkeldiev, A., Takenov, Z., Iliasov, S., and Yakimov, V. (2009). “The Working Paper, Washington, DC. Kyrgyz Republic’s Second National Communication to the United Cushing L. and Kopas J. (2011), “Principal Human Rights Impacts of Nations Framework Convention on Climate Change.” United Nations Climate Change in Latin America.” Development Program (UNDP), The State Agency for the Environ- Cutter, S.L., Emrich, C.T., Webb, J.J., and Morath, D. (2009). “Social Vul- ment Protection and Forestry under the Government of the Kyrgyz nerability to Climate Variability Hazards: A Review of the Literature.” Republic: Bishkek. University of South Carolina: Columbia. De’ath, G., Fabricius, K.E., Sweatman, H. and Puotinen, M. (2012). “The D’Amato, G., and Cecchi, L. (2008). “Effects of Climate Change on 27-year Decline of Coral Cover on the Great Barrier Reef and its Environmental Factors in Respiratory Allergic Diseases.” Clinical & Causes.” Proceedings of the National Academy of Sciences of the Experimental Allergy, 38, 1264–74. United States of America, 1–5. D’Odorico, P., and Bhattachan, A. (2012). “Hydrologic Variability in Dry- Deichmann, U., Meisner, C., Murray, S., and Wheeler, D. (2011). “The land Regions: Impacts on Ecosystem Dynamics and Food Security.” Economics of Renewable Energy Expansion in Rural Sub-Saharan Philosophical transactions of the Royal Society of London. Series B, Africa.” Energy Policy, 39(1), 215–27. Biological sciences, 367(1606), 3145–57. Dell, J. and Pasteris, P. (2010). “Adaptation in the Oil and Gas Industry to D’Odorico, P., Bhattachan, A., Davis, K.F., Ravi, S., and Runyan, C.W. Projected Impacts of Climate Change.” Society of Petroleum Engineers, (2013). “Global Desertification: Drivers and Feedbacks.” Advances Society of Petroleum Engineers: Rio de Janeiro. in Water Resources, 51, 326–44. Demarcq, H. (2009). “Trends in Primary Production, Sea Surface Tem- Dadush, U., and Stancil, B. (2010). “The World Order in 2050.” Carnegie perature and Wind in Upwelling Systems (1998–2007).” Progress in Endowment for International Peace: Washington, DC. Oceanography, 83(1-4), 376–85. Dahlberg, M., and Gustavsson, M. (2005). “Inequality and Crime: Sepa- Demaria, E.M.C., Maurer, E.P., Thrasher, B., Vicuña, S., and Meza, F.J. rating the Effects of Permanent and Transitory Income.” Stockholm. (2013). “Climate Change Impacts on an Alpine Watershed in Chile: Dai, A. (2012). “Increasing Drought under Global Warming in Observations Do New Model Projections Change the Story?” Journal of Hydrology, and Models.” Nature Climate Change, 3(1), 52–58. 502, 128–38. Dakova, S. (2005). “Climate Change Effect on River Flow in Basins under Demetriades, J., and Esplen, E. (2009). “The Gender Dimensions of Poverty Different Climate Conditions—An Example from Rivers on the Balkan and Climate Change Adaptation.” IDS Bulletin, 39(4), 24–31. Peninsula.” Regional Hydrological Impacts of Climate Change—Impact Denton, F. (2002). “Climate change Vulnerability, Impacts, and Adaptation: Assessment and Decision Making, IAHS. Why Does Gender Matter?” Gender and Development, 10(2–Climate Dankelman, I. (2010). Gender and Climate Change. An Introduction. Change), 10–20. Earthscan: London. Deprez, A. (2010). “Climate Migration in Latin America: A future ‘Flood Dankers, R., and Feyen, L. (2009). “Flood Hazard in Europe in an Ensemble of Refugees’ to the North?” Council on Hemispheric Affairs: Wash- of Regional Climate Scenarios.” Journal of Geophysical Research, ington, DC. 114(D16108). Deutsch, C.A., Tewksbury, J.J., Huey, R.B., Sheldon, K.S., Ghalambor, C.K., DARA (2012). “Labour Productivity Indicator.” Climate Vulnerability Haak, D.C., and Martin, P.R. (2008). “Impacts of Climate Warming Monitor: A Guide to the Cold Calculus of a Hot Planet, 2nd ed., on Terrestrial Ectotherms Across Latitude Thermal Safety Margin.” DARA: Madrid, Spain. Proceedings of the National Academy of Sciences of the United States Darwish, K., Safaa, M., Momou, A. and Saleh, S. (2013). “Egypt: Land of America, 105(18), 6668–72. Degradation Issues with Special Reference to the Impact of Climate Devi, N., Hagedorn, F., Moiseev, P., Bugmann, H., Shiyatov, S., Mazepa, Change.” Combating Desertification in Asia, Africa, and the Middle East, V. and Rigling, A. (2008). “Expanding Forests and Changing Growth edited by G. A. Heshmati and V. Squires, Springer Dordrecht: Dordrecht. Forms of Siberian Larch at the Polar Urals Treeline during the 20th Dasgupta, S., Laplante, B., Meisener, C., Wheeler, D. and Yan, J. (2007). Century.” Global Change Biology, 14(7), 1581–91. “The Impact Of Sea-level Rise On Developing Countries: A Compara- Dilley, M., Chen, R.S., Deichmann, U., Lerner-Lam, A.L., Arnold, M., tive Analysis: Policy Research Working Papers.” Washington, DC. Agwe, J., Buys, P., Kjekstad, O., Lyon, B., and Yetman, G. (2005), Dasgupta, S., Laplante, B., Meisner, C., Wheeler, D., and Yan, J. (2009). “Natural Disaster Hotspots: A Global Risk Analysis.” Disaster Risk “The Impact of Sea-level Rise on Developing Countries: A Compara- Management Series, Washington DC. tive Analysis.” Climatic Change, 93(3-4), 379–88. Dimkic, D., and Despotovic, J. (2012), “Analysis of the Changes of the Dasgupta, S., Laplante, B., Murray, S. and Wheeler, D. (2009). “Sea-level Streamflows in Serbia Due to Climate Changes.” Climate Change— Rise and Storm Surges: A Comparative Analysis of Impacts in Devel- Inferences from Paleoclimate and Regional Aspects, edited by A. Berger, oping Countries,” Washington, DC. F. Mesinger, and D. Sijacki, Springer Vienna: Austria. Dasgupta, S., Laplante, B., Murray, S. and Wheeler, D. (2011). “Exposure Van Dingenen, R., Dentener, F.J., Raes, F., Krol, M.C., Emberson, L. and of Developing Countries to Sea-level Rise and Storm Surges.” Climatic Cofala, J. (2009). “The Global Impact of Ozone on Agricultural Crop Change, 106(4), 567–79. Yields under Current and Future Air Quality Legislation.” Atmospheric Davidson, E.A., de Araújo, A.C., Artaxo, P., Balch, J.K., Brown, I.F., C Busta- Environment, 43(3), 604–18. mante, M.M., Coe, M.T., DeFries, R.S., Keller, M., Longo, M., Munger, Dixon, J.A., Gibbon, D.P., and Gulliver, A. (2001a). Farming Systems and J.W., Schroeder, W., Soares-Filho, B.S., Souza, C.M., and Wofsy, S.C. Poverty: Improving Farmers´ Livelihoods in a Changing World. Edited (2012). “The Amazon Basin in Transition.” Nature, 481(7381), 321–28. by M. Hall, FAO and Worldbank: Rome, Washington. 246 B I BLIOGRA P HY Dixon, J.A., Gibbon, D.P., and Gulliver, A. (2001b). Global Farming Sys- DREF (2009). “DREF Operation—Syria: Drought.” Disaster Relief Emer- tems Study: Challenges and Priorities to 2030. Synthesis and Global gency Fund of the International Federation of Red Cross and Red Overview. The Food and Agriculture Organization: Rome, Italy. Cresent Societies. Döll, P. (2009). “Vulnerability to the Impact of Climate Change on Renewable Drine, I. (2011). “Climate Change Compounding Risks in North Africa.” Groundwater Resources: A Global-scale Assessment.” Environmental Working paper, World Institute for Development Economics Research Research Letters, 4(3), 035006. (WIDER), No. 2011,32. Döll, P., and Schmied, H.M. (2012). “How is the Impact of Climate Change Dukhovny, V.A., and Schutter, J. de (2011). Water in Central Asia—Past, Pres- on River Flow Regimes related to the Impact on Mean Annual Runoff? ent, and Future. CRC Press/Balkema: Taylor & Francis Group, London. A Global-scale Analysis.” Environmental Research Letters, 7(1), 1–11. Dulamsuren, C., Wommelsdorf, T., Zhao, F., Xue, Y., Zhumadilov, B.Z., Dolman, A. J., Shvidenko, A., Schepaschenko, D., Ciais, P., Tchebakova, N., Leuschner, C., and Hauck, M. (2013). “Increased Summer Temperatures Chen, T., van der Molen, M.K., Belelli Marchesin,i L., Maximov, T.C., Reduce the Growth and Regeneration of Larix sibirica in Southern Maksyutov, S., and Schulze, E.-D. (2012). “An Estimate of the Terrestrial Boreal Forests of Eastern Kazakhstan.” Ecosystems, 16(8), 1536–49. Carbon Budget of Russia Using Inventory-based, Eddy Covariance and Durack, P.J., Wijffels, S.E., and Matear, R.J. (2012). “Ocean Salinities Inversion Methods.” Biogeosciences, 9(12), 5323–40. Reveal Strong Global Water Cycle Intensification During 1950 to 2000.” Donat, M.G., Alexander, L. V., Yang, H., Durre, I., Vose, R., Dunn, R.J.H., Science, 336(6080), 455–58. Willett, K.M., Aguilar, E., Brunet, M., Caesar, J., Hewitson, B., Dutrieux, P., De Rydt, J., Jenkins, A., Holland, P.R., Ha, H.K., Lee, S.H., Jack, C., Klein Tank, A.M.G., Kruger, A.C., Marengo, J., Peterson, T.C., Steig, E.J., Ding, Q., Abrahamsen, E.P., and Schröder, M. (2014). Renom, M., Oria Rojas, C., Rusticucci, M., Salinger, J., Elrayah, A.S., “Strong Sensitivity of Pine Island Ice-shelf Melting to Climatic Vari- Sekele, S.S., Srivastava, A.K., Trewin, B., Villarroel, C., Vincent, L.A., ability.” Science (New York, N.Y.), 343(6167), 174–78. Zhai, P., Zhang, X., and Kitching, S. (2013). “Updated Analyses of Eakin, C.M., Morgan, J. A, Heron, S.F., Smith, T.B., Liu, G., Alvarez- Temperature and Precipitation Extreme Indices Since the Beginning of Filip, L., Baca, B., Bartels, E., Bastidas, C., Bouchon, C., Brandt, M., the Twentieth Century: The HadEX2 Dataset.” Journal of Geophysical Bruckner, A.W., Bunkley-Williams, L., Cameron, A., Causey, B.D., Research: Atmospheres, 118(5), 2098–2118. Chiappone, M., Christensen, T.R.L., Crabbe, M.J.C., Day, O., de la Donat, M.G., Peterson, T.C., Brunet, M., KingA.D., Almazroui, M., Guardia, E., Díaz-Pulido, G., DiResta, D., Gil-Agudelo, D.L., Gil- Kolli, R.K., Boucherf, D., Al-Mulla, A.Y., Nour, A.Y., Aly, A.A., liam, D.S., Ginsburg, R.N., Gore, S., Guzmán, H.M., Hendee, J.C., Nada, T.A.A., Semawi, M.M., Dashti, H.A. Al, Salhab, T.G., El Hernández-Delgado, E.A., Husain, E., Jeffrey, C.F.G., Jones, R.J., Fadli, K.I., Muftah, M.K., Eida, S.D., Badi, W., Driouech, F., Rhaz, K. El, Jordán-Dahlgren, E., Kaufman, L.S., Kline, D.I., Kramer, P. A, Abubaker, M.J.Y., Ghulam, A.S., Erayah, A.S., Mansour, M. Ben, Lang, J.C., Lirman, D., Mallela, J., Manfrino, C., Maréchal, J.P., Alabdouli, W.O., Dhanhani, J.S. Al, and Al Shekaili, M.N. (2013). Marks, K., Mihaly, J., Miller, W.J., Mueller, E.M., Muller, E.M., Orozco “Changes in Extreme Temperature and Precipitation in the Arab Toro, C.A., Oxenford, H.A, Ponce-Taylor, D., Quinn, N., Ritchie, K.B., Region: Long-term Trends and Variability related to ENSO and NAO.” Rodríguez, S., Ramírez, A.R., Romano, S., Samhouri, J.F., Sánchez, J.A, International Journal of Climatology, 34(3), 581–92. Schmahl, G.P., Shank, B.V, Skirving, W.J., Steiner, S.C.C., Villamizar, E., Donat, M.G., Sillmann, J., Wild, S., Alexander, L. V., Lippmann, T. and Walsh, S.M., Walter, C., Weil, E., Williams, E.H., Roberson, K.W., and Zwiers, F.W. (2014). “Consistency of Temperature and Precipitation Yusuf, Y. (2010). “Caribbean Corals in Crisis: Record Thermal Stress, Extremes across Various Global Gridded In Situ and Reanalysis Data- Bleaching, and Mortality in 2005.” PloS one, 5(11), e13969. sets.” Journal of Climate, 27(13), 5019–35. Easdale, M.H., and Domptail, S.E. (2014). “Fate Can be Changed! Arid Doney, S.C., Ruckelshaus, M., Duffy, J.E., Barry, J.P., Chan, F., English, Rangelands in a Globalizing World—A Complementary Co-evolutionary C.A., Galindo, H.M., Grebmeier, J.M., Hollowed, A.B., Knowlton, N., Perspective on the Current ‘Desert Syndrome.’” Journal of Arid Envi- Polovina, J., Rabalais, N.N., Sydeman, W.J., and Talley, L.D. (2012). ronments, 100, 52–62. “Climate Change Impacts on Marine Ecosystems.” Annual Review of Easterling, W.E., Aggarwal, P.K., Batima, P., Brander, K.M., Erda, L., Marine Science, 4, 11–37. Howden, S.M., Kirilenko, A., Morton, J., Soussana, J.F., Schmidhu- Donnelly, J.P., and Woodruff, J.D. (2007). “Intense Hurricane Activity ber, J., and Tubiello, F.N. (2007), “Food, Fibre and Forest Products.” over the Past 5,000 Years Controlled by El Nino and the West African Climate Change 2007: Impacts, Adaptation and Vulnerability. Contri- Monsoon.” Nature, 447(7143), 465–68. bution of Working Group II to the Fourth Assessment Report of the Donohue, R.J., Roderick, M.L., McVicar, T.R., and Farquhar, G.D. (2013). Intergovernmental Panel on Climate Change, edited by M. L. Parry, “CO2 Fertilisation has Increased Maximum Foliage Cover across the O. F. Canziani, J. P. Palutikof, P. J. van der Linden, and C. E. Hanson, Globe’s Warm, Arid Environments.” Geophysical Research Letters, Cambridge University Press: Cambridge, UK. 40(12). Ebinger, J., and Vergara, W. (2011). “Climate Impacts on Energy Systems.” Dore, J.E., Lukas, R., Sadler, D.W., Church, M.J., and Karl, D.M. (2009). Washington, DC. “Physical and Biogeochemical Modulation of Ocean Acidification in ECDC (2012). “The Climatic Suitability for Dengue Transmission in the Central North Pacific.” Proceedings of the National Academy of Continental Europe.” European Centre for Disease Prevention and Sciences of the United States of America, 106(30), 12235–40. Control: Stockholm. Douglas, I. (2009). “Climate Change, Flooding, and Food Security in South ECDC (2013). “Annual Epidemiological Report: Reporting on 2010 Surveil- Asia.” Food Security, 1(2), 127–36. lance Data and 2011 Epidemic Intelligence Data.” European Centre Douglas, I., Alam, K., Maghenda, M., Mcdonnell, Y., Mclean, L., and for Disease Prevention and Control: Stockholm. Campbell, J. (2008). “Unjust Waters: Climate Change, Flooding and Echeverri, J.Á. (2009). “Pueblos Indígenas y Cambio Climático: El Caso the Urban Poor in Africa.” Environment and Urbanization, 20(1), de la Amazonía Colombiana.” Bulletin de l’Institut Français d’Études 187–205. Andines, 38(1), 13–28. Dove, S.G., Kline, D.I., Pantos, O., Angly, F.E., Tyson, G.W., and Hoegh- ECLAC (2001). “International Migration and Development in the Ameri- Guldberg, O. (2013). “Future Reef Decalcification under a Business-as- cas.” Symposium on International Migration in the Americas, United Usual CO2 Emission Scenario.” Proceedings of the National Academy Nations Economic Commission for Latin America and the Caribbean of Sciences of the United States of America, 110(38), 15342–47. (ECLAC): San Jose, Costa Rica. Drabo, A., and Mbaye, L.M. (2011). “Climate Change, Natural Disasters and ECLAC (2010). “Economics of Climate Change in Latin America and the Migration: An Empirical Analysis in Developing Countries,” Discussion Caribbean. Summary 2010.” United Nations Economic Commission paper No. 5927, Forschungsinstitut zur Zukunft der Arbeit, Bonn. for Latin America and the Caribbean (ECLAC): Santiago, Chile. 247 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL ECLAC (2014). “Economic and Social Panorama of the Community of Latin of the National Academy of Sciences of the United States of America, American and Caribbean States, 2013.” United Nations Economic Com- 110(30), 12219–24. mission for Latin America and the Caribbean (ECLAC): Santiago, Chile. Emch, M., Feldacker, C., Sirajul Islam, M., and Ali, M. (2008). “Season- ECLAC, FAO, and IICA (2012). “The Outlook for Agriculture and Rural ality of Cholera from 1974 to 2005: A Review of Global Patterns.” Development in the Americas: A Perspective on Latin America and International Journal of Health Geographics, 7(31). the Caribbean.” Economic Commission for Latin America and the Enarson, E. (2003). “Working With Women At Risk. Practical Guidelines Caribbean (ECLAC), Food and Agriculture Organization of the UN for Assessing Local Disaster Risk.” International Hurricane Center, (FAO), Inter-American Institute for Cooperation on Agriculture (IICA): Florida International University. Santiago, Chile. Engelstaedter, S., Tegen, I., and Washington, R. (2006). “North African ECLAC and UNFPA (2009). 2009 Social Panorama of Latin America. United Dust Emissions and Transport.” Earth-Science Reviews, 79(1), 73–100. Nations Economic Commission for Latin America and the Caribbean England, M.H., Mcgregor, S., Spence, P., Meehl, G.A., Timmermann, A., (ECLAC) and the United Nations Population Fund (UNFPA): New Cai, W., Gupta, A. Sen, Mcphaden, M.J., Purich, A., and Santoso, A. York, NY. (2014). “Recent Intensification of Wind-driven Circulation in the Pacific Edward, P., and Sumner, A. (2013). “The Future of Global Poverty in a and the Ongoing Warming Hiatus.” Nature Climate Change, 4(February). Multi-Speed World. New Estimates of Scale and Location, 2010–2030.” Ennesser, Y., Cataliotti, D., and Terrier, M. (2010). “Changement Climatique Working Paper 327, Centre for Global Development: Washington DC. et Evaluation des Risques Côtiers en Afrique du Nord.” Recueil des Edwards O. (2011). “Method for Assessing the Potential Impact of Climate Actes, Impacts du Changement Climatique sue les Risques Cotiers,15–16 Change on Streamflow Regimes and Its Implications in Trinidad, Using November 2010, BRGM-Orleans, France. the Maracas/St. Joseph Catchment as a Case Study.” University of Van der Ent, R.J., Savenije, H.H.G., Schaefli, B., and Steele-Dunne, S.C. the West Indies, Barbados. (2010). “Origin and Fate of Atmospheric Moisture over Continents.” EEA (2012). “Climate Change, Impacts and Vulnerability in Europe 2012.” Water Resources Research, 46(9), W09525. European Environment Agency: Copenhagen. ENVSEC and UNEP (2011). “Environment and Security in the Amu Arya EEAA (2010). “Egypt: Second National Communication under the United Basin.” Environment and Security Initiative (ENVSEC), United Nations Nations Framework Convention on Climate Change.” Egyptian Envi- Environment Program (UNEP), GRID Arendal and Zoi Environment ronmental Affairs Agency: Cairo. Network: Nairobi, Kenya. EIA (2013). “International Energy Outlook 2013.” Washington, DC. ENVSEC and UNEP (2012). “Climate Change in the West Balkans.” The EIA (2014a). “Total Oil Supply (in thousands of barrels per day).” U.S. Environment and Security Initiative (ENVSEC), United Nations Environ- Energy Information Administration (EIA), International Energy Sta- ment Program (UNEP) and Zoï Environment Network: Nairobi, Kenya. tistics, (accessed May 19, 2014). Available at http://www.eia.gov/ Erbs, M., Manderscheid, R., Jansen, G., Seddig, S., Pacholski, A., and cfapps/ipdbproject/iedindex3.cfm?tid=5&pid=53&aid=1. Weigel, H.J. (2010). “Effects of Free-air CO2 Enrichment and Nitrogen EIA (2014b). “Gross Natural Gas Production (in Billion Cubic Feet).” Supply on Grain Quality Parameters and Elemental Composition of U.S. Energy Information Administration (EIA), International Energy Wheat and Barley Grown in a Crop Rotation.” Agriculture, Ecosystems Statistics, (accessed May 19, 2014). Available at http://www.eia.gov/ & Environment, 136(1–2), 59–68. cfapps/ipdbproject/IEDIndex3.cfm?tid=3&pid=3&aid=1. Espinoza Villar, J.C., Guyot, J.L., Ronchail, J., Cochonneau, G., Filizola, Eisenberg, M.C., Kujbida, G., Tuite, A.R., Fisman, D.N., and Tien, J.H. N., Fraizy, P., Labat, D., de Oliveira, E., Ordoñez, J.J., and Vauchel, (2013). “Examining Rainfall and Cholera Dynamics in Haiti using P. (2009). “Contrasting Regional Discharge Evolutions in the Amazon Statistical and Dynamic Modeling Approaches,” Epidemics, 5, 197–207. Basin (1974–2004).” Journal of Hydrology, 375(3–4), 297–311. Elasha B.O. (2010). “Mapping Climate Change Threats and Human Develop- Estay, S. A., Lima, M., and Labra, F. A. (2009). “Predicting Insect Pest ment Impacts in the Arab Region. Arab Human Development Report Status under Climate Change Scenarios: Combining Experimental Data Research Paper.” UNEP, Regional Bureau of Arab States: New York. and Population Dynamics Modelling.” Journal of Applied Entomol- Elewa, H.H., and El Nahry, A.H. (2008). “Hydro-environmental Status ogy, 133(7), 491–99. and Soil Management of the River Nile Delta, Egypt.” Environmental Esteban, M., Thao, N.D., and Takagi, H. (2012). “Increase in Port Downtime Geology, 57(4), 759–74. and Damage in Vietnam Due To a Potential Increase in Tropical Cyclone El-Fadel, M., Ghanimeh, S., Maroun, R., and Alameddine, I. (2012). Intensity.” Climate Change and the Sustainable Use of Water Resources, “Climate Change and Temperature Rise: Implications on Food- and edited by W. Leal Filho, Springer Berlin Heidelberg: Heidelberg. Water-borne Diseases.” Science of the Total Environment, 437, 15–21. Estrella Arellano, C., and Swartzman, G. (2010). “The Peruvian Artisanal Elminir, H.K., Ghitas, A.E., Hamid, R.H., El-Hussainy, F., Beheary, M.M., Fishery: Changes in Patterns and Distribution Over Time.” Fisheries and Abdel-Moneim, K.M. (2006). “Effect of Dust on the Transparent Research, 101(3), 133–45. Cover of Solar Collectors.” Energy Conversion and Management, Evans, J., and Geerken, R. (2004). “Discrimination between Climate and 47(18-19), 3192–3203. Human-induced Dryland Degradation.” Journal of Arid Environments, Elmore, A.J., Kaste, J.M., Okin, G.S., and Fantle, M.S. (2008). “Groundwater 57, 535–54. Influences on Atmospheric Dust Generation in Deserts.” Journal of Evans, J.P. (2008). “21st Century Climate Change in the Middle East.” Arid Environments, 72(10), 1753–65. Climatic Change, 92(3–4), 417–32. El-Nahry, A.H., and Doluschitz, R. (2010). “Climate Change and its Impacts Evans, J.P. (2009). “Global Warming Impact on the Dominant Precipitation on the Coastal Zone of the Nile Delta, Egypt.” Environmental Earth Processes in the Middle East.” Theoretical and Applied Climatology, Sciences, 59(7), 1497–1506. 99(3-4), 389–402. El-Zein, A., Tewtel-Salem, M., and Nehme, G. (2004). “A Time Series Exbrayat, J.F., Buytaert, W., Timbe, E., Windhorst, D., and Breuer, L. Analysis of Mortality and Air Temperature in Greater Beirut.” Science (2014). “Addressing Sources of Uncertainty in Runoff Projections for of the Total Environment, 330, 71–80. a Data Scarce Catchment in the Ecuadorian Andes.” Climatic Change, Emanuel, K. (2005). “Increasing Destructiveness of Tropical Cyclones over 125(2), 221–35. the Past 30 Tears.” Nature, 436(7051), 686–88. Fabrega, J., Nakaegawa, T., Pinzón, R., Nakayama, K., Arakawa, O. ,and Emanuel, K. (2008). “The Hurricane—Climate Connection.” Bulletin of Group, S.T.C.M. (2013). “Hydroclimate Projections for Panama in the American Meteorological Society, 89(5), ES10–ES20. the late 21st Century.” Hydrological Research Letters, 7(2), 23–29. Emanuel, K.A. (2013). “Downscaling CMIP5 Climate Models Shows Faist, T., and Schade, J. (2013). “The Climate-Migration Nexus: A Reorien- Increased Tropical Cyclone Activity over the 21st Century.” Proceedings tation.” Disentangling Migration and Climate Change, Methodologies, 248 B I BLIOGRA P HY Political Discourses, and Human Rights, Springer Science and Business Fereira, F., Messina, J., RIgolini, J., Lopez-Calva, L., Lugo, M.A., and Media: Dordrecht, pp 3–25. Vakis, R. (2013). Economic Mobility and the Rise of the Latin Ameri- Fajnzylber, P., Lederman, D., and Loayza, N. (2002). “Inequality and can Middle Class, Choice Reviews Online. IADB and the World Bank: Violent Crime.” Journal of Law and Economics, 45(1), 1–40. Washington DC. FAO (2003). “Review of World Water Resources by Country.” Water Reports Fernandes, E.C.M., Soliman, A., Confalonieri, R., Donatelli, M., and Tubi- No. 23, Food and Agriculture Organization: Rome. ello, F. (2012). “Climate Change and Agriculture in Latin America, FAO (2010). “Fisheries and Aquaculture Country Profiles: Brazil.” 2020-2050.” Washington DC. FAO Fisheries and Aquaculture Department, accessed Decem- Fernandes, J.A., Cheung, W.W.L., Jennings, S., Butenschön, M., de Mora, ber 13, 2013. Available at http://www.fao.org/fishery/facp/BRA/ L., Frölicher, T.L., Barange, M., and Grant, A. (2013). “Modelling the en#CountrySector-ProductionSector. Effects of Climate Change on the Distribution and Production of Marine FAO (2012a). “The State of Food Insecurity in the World 2012. The multiple Fishes: Accounting for Trophic Interactions in a Dynamic Bioclimate Dimensions of Food Security. Key Messages.” Food and Agriculture Envelope Model.” Global Change Biology, 19(8), 2596–2607. Organization, World Food Programme, and the International Fund Ferrise, R., Moriondo, M., Trombi, G., Miglietta, F., and Bindi, M. (2013). for Agricultural Development: Rome. “Climate Change Impacts on Typical Mediterranean Crops and Evalu- FAO (2012b). The Russian Federation Forest Sector Outlook Study to 2030. ation of Adaptation Strategies to Cope With.” Regional Assessment of Food and Agriculture Organization: Rome. Climate Change in the Mediterranean, Springer, 49–70. FAO (2013). “The State of Food Insecurity in the World 2013. The Multiple Few, R., Ahern, M., Matthies, F., and Kovats, S. (2004). “Floods, Health Dimensions of Food Security. Key Messages.” Food and Agriculture and Climate Change: A Strategic Review.” Working Paper 63, Tyndall Organiszation, World Food Programme, and the International Fund Centre for Climate Change Research. for Agricultural Development: Rome. Ficke, A., Myrick, C., and Hansen, L. (2007). “Potential Impacts of Global FAO (2014a), “Food Price Index,” Food and Agriculture Organization of Climate Change on Freshwater Fisheries.” Reviews in Fish Biology the United Nations, (accessed March 7, 2014), [available at http:// and Fisheries, 17(4), 581–613. www.fao.org/worldfoodsituation/foodpricesindex/en/]. Field, C.B., Lobell, D.B., Peters, H.A., and Chiariello, N.R. (2007). “Feed- FAO (2014b). “Food Price Indices Data.” Food and Agriculture Organiza- backs of Terrestrial Ecosystems to Climate Change.” Annual Review tion: Rome. of Environment and Resources, 32(1), 1–29. FAO (2014c). “FAO’s Food Price Index Revisited.” Food and Agriculture Fish, M.R., Cote, I.M., Gill, J.A., Jones, A.P., Renshoff, S., and Watkinson, Organization: Rome. A.R. (2005). “Predicting the Impact of Sea-level Rise on Caribbean FAO (2014d). “Report of the 22nd Session of the Committee on Forestry: Sea Turtle Nesting Habitat.” Conservation Biology, 19(2), 482–91. Enhancing Work on Boreal Forests and Dryland Forests.” Food and Fisher, R., McDowell, N., Purves, D., Moorcroft, P., Sitch, S., Cox, P., Agriculture Organisation: Rome, COFO/2014/7.3a, pp 97-103. Huntingford, C., Meir, P., and Woodward, F.I. (2010). “Assessing Uncer- FAO-AQUASTAT (2012). “Irrigation in Central Asia in Figures.” AQUASTAT tainties in a Second-Generation Dynamic Vegetation Model caused by Survey, Food and Agriculture Organization: Rome. Ecological Scale Limitations.” The New Phytologist, 187(3), 666–81. FAOSTAT (2013). “Data: Agricultural Commodity Production Quantities.” Flannigan, M., Stocks, B., Turetsky, M., and Wotton, M. (2009). “Impacts FAOSTAT Agricultural Production Datasets, accessed November 25, of Climate Change on Fire Activity and Fire Management in the Cir- 2013. Available at http://faostat.fao.org/. cumboreal Forest.” Global Change Biology, 15(3), 549–60. Farber, E., Vengosh, A., Gavrieli, I., Marie, A., Bullen, T.D., Mayer, B., Foden, W.B., Butchart, S.H.M., Stuart, S.N., Vié, J.C., Akçakaya, H.R., Holtzman, R., Segal, M. and Shavit, U. (2004). “The Origin and Angulo, A., DeVantier, L.M., Gutsche, A., Turak, E., Cao, L., Donner, Mechanisms of Salinization of the Lower Jordan River.” Geochimica S.D., Katariya, V., Bernard, R., Holland, R.A., Hughes, A.F., O’Hanlon, et Cosmochimica Acta, 68(9), 1989–2006. S.E., Garnett, S.T., Sekercio lu, C.H., and Mace, G.M. (2013). “Identify- Favier, L., Durand, G., Cornford, S.L., Gudmundsson, G.H., Gagliardini, O., ing the World’s Most Climate Change Vulnerable Species: A Systematic Gillet-Chaulet, F., Zwinger, T., Payne, A. J., and Le Brocq, A. M. Trait-based Assessment of all Birds, Amphibians, and Corals.” PloS (2014). “Retreat of Pine Island Glacier Controlled by Marine Ice-sheet one, 8(6), e65427. Instability.” Nature Climate Change, 4(2), 117–21. Foley, J.A., Coe, M.T., Scheffer, M., and Wang, G. (2003). “Regime Shifts Fay, M., Block, R.I., and Ebinger, J. (2010). Adapting to Climate Change in the Sahara and Sahel: Interactions between Ecological and Climatic in Eastern Europe and Central Asia, World Bank: Washington DC. Systems in Northern Africa.” Ecosystems, 6(6), 524–32. Feeley, K.J., Malhi, Y., Zelazowski, P., and Silman, M.R. (2012). “The Relative Foley, J.A., Costa, M.H., Delire, C., Ramankutty, N., and Snyder, P. (2003). Importance of Deforestation, Precipitation Change, and Temperature “Green Surprise? How Terrestrial Ecosystems Could Affect Earth’s Sensitivity in Determining the Future Distributions and Diversity of Climate.” Frontiers in Ecology and the Environment, 1(1), 38–44. Amazonian Plant Species.” Global Change Biology, 18(8), 2636–47. Folke, C. (2003). “Freshwater for Resilience: A Shift in Thinking.” Philosophi- Feldt, H. (2011). “Fortalecimiento de Organizaciones Indígenas en América cal Transactions of the Royal Society of London, Series B, Biological Latina: Pueblos Indígenas y Cambio Climático. Relación entre Cambio Sciences, 358(1440), 2027–36. Climático y Pueblos Indígenas y sus Posiciones en el Contexto de las Fordham, M. (2012). “Gender, Sexuality, and Disaster.” The Routledge Negociaciones en la Convención Marco sobre el Cambio Climático.” Deut- Handbook of Hazards and Disaster Risk Reduction, edited by B. Wisner, sche Gesellschaft für Internationale Zusammenarbeit (GIZ): Frankfurt. J.C. Gaillard, and I. Kelman, Routledge: New York. Femia, F., and Werrell, C.E. (2013). “Climate Change Before and After the Foresight (2011). “Migration and Global Environmental Change. Challenges Arab Awakening: The Cases of Syria and Libya.” The Arab Spring and and Opportunities. Final Project Report.” The Government Office for Climate Change. A Climate and Security Correlation Series, edited by Science: London. C. E. Werrell and F. Femia, Center for American Progress, Stimson Foster, G. and Rahmstorf, S. (2011). “Global Temperature Evolution Center: Washington DC. 1979–2010.” Environmental Research Letters, 6(4), 044022. Feng, S., and Fu, Q. (2013). “Expansion of Global Drylands under a Warm- Franke, C.R., Ziller, M., Staubach, C., and Latif, M. (2002). “Impact of ing Climate.” Atmospheric Chemistry and Physics, 13(19), 10081–94. the El Niño/Southern Oscillation on Visceral Leishmaniasis, Brazil.” Feng, S., Krueger, A.B. and Oppenheimer, M. (2010). “Linkages among Emerging Infectious Diseases, 8(9). Climate Change, Crop Yields, and Mexico-U.S. Cross-border Migra- Frieler, K., Meinshausen, M., Golly, A., Mengel, M., Lebek, K., Donner, S.D., tion.” Proceedings of the National Academy of Sciences of the United and Hoegh-Guldberg, O. (2012). “Limiting Global Warming to 2°C is States of America, 107(32), 14257–62. Unlikely to Save Most Coral Reefs.” Nature Climate Change, 2(9), 1–6. 249 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Friend, A.D., Lucht, W., Rademacher, T.T., Keribin, R., Betts, R., Cadule, P., Bolch, T., Sharp, M.J., Hagen, J.O., van den Broeke, M.R., and Paul F. Ciais, P., Clark, D.B., Dankers, R., Falloon, P.D., Ito, A., Kahana, R., (2013). “A Reconciled Estimate of Glacier Contributions to Sea-level Kleidon, A., Lomas, M.R., Nishina, K., Ostberg, S., Pavlick, R., Peylin, Rise: 2003 to 2009.” Science (New York, N.Y.), 340(6134), 852–57. P., Schaphoff, S., Vuichard, N., Warszawski, L., Wiltshire, A., and De Garin, A., and Bejaran, R. (2003). “Mortality Rate and Relative Strain Woodward, F.I. (2014). “Carbon Residence Time Dominates Uncertainty Index in Buenos Aires City.” Int J Biometeorol, 48, 31–36. in Terrestrial Vegetation Responses to Future Climate and Atmospheric Garreaud, R.D., Vuille, M., Compagnucci, R., and Marengo, J. (2009). CO2.” Proceedings of the National Academy of Sciences of the United “Present-day South American Climate.” Palaeogeography, Palaeocli- States of America, 111(9), 3280–85. matology, Palaeoecology, 281(3-4), 1–16. Frihy, O.E., and El-Sayed, M.K. (2013). “Vulnerability Risk Assessment Gasparrini, A., and Armstrong, B. (2011). “The Impact of Heat Waves on and Adaptation to Climate Change Induced Sea-level Rise along the Mortality.” Epidemiology, 22(1), 68–73. Mediterranean Coast of Egypt.” Mitigation and Adaptation Strategies Gatti, L. V, Gloor, M., Miller, J.B., Doughty, C.E., Malhi, Y., Domingues, L.G., for Global Change, 18(8), 1215–37. Basso, L.S., Martinewski, A, Correia, C.S.C., Borges, V.F., Freitas, S., Fritz, C. (2010). “Climate Change and Migration: Sorting through Complex Braz, R., Anderson, L.O., Rocha, H., Grace, J., Phillips, O.L., and Issues without the Hype.” Migration Information Source, Migration Lloyd, J. (2014). “Drought Sensitivity of Amazonian Carbon Balance Policy Institute: Washington DC. revealed by Atmospheric Measurements.” Nature, 506(7486), 76–80. Fry, L.M., Watkins, D.W., Reents, N., Rowe, M.D., and Mihelcic, J.R. (2012). Geist, H., and Lambin, E. (2004). “Dynamic Causal Patterns of Desertifica- “Climate Change and Development Impacts on the Sustainability of tion.” BioScience, 54(9), 817–30. Spring-fed Water Supply Systems in the Alto Beni Region of Bolivia.” Gemenne, F. (2011). “Climate-induced Population Displacements in a Journal of Hydrology, 468-469, 120–29. 4°C+ World.” Philosophical Transactions. Series A, Mathematical, Fu, R., Yin, L., Li, W., Arias, P.A., Dickinson, R.E., Huang, L., Chakraborty, Physical, and Engineering Sciences, 369(1934), 182–95. S., Fernandes, K., Liebmann, B., Fisher, R., and Myneni, R.B. (2013). Gemenne, F. (2013). “Migration Doesn’t Have To Be a Failure to Adapt. “Increased Dry-season Length over Southern Amazonia in Recent An Escape from Environmental Determinism.” Climate Adaptation Decades and its Implication for Future Climate Projection.” Proceed- Futures, edited by J. Palutikof, S. L. Boulter, A. J. Ash, M. S. Smith, ings of the National Academy of Sciences of the United States of M. Parry, M. Waschka, and D. Guitart, John Wiley & Sons. America, 110(45), 18110–15. Gemenne, F., Barnett, J., Adger, W.N., and Dabelko, G.D. (2014). “Climate Füssel, H.M. (2012). “Vulnerability to Climate Change and Poverty.” and Security: Evidence, Emerging Risks, and a New Agenda.” Climatic Climate Change, Justice and Sustainability, edited by Edenhofer et Change, 123(1), 1–9. al., Springer: Dordrecht. Gencer, E.A. (2013). “An Overview of Urban Vulnerability to Natural Gage, K.L., Burkot, T.R., Eisen, R.J., and Hayes, E.B. (2008). “Climate Disasters and Climate Change in Central America and the Caribbean and Vectorborne Diseases.” American journal of preventive medicine, Region.” Fondazione Eni Enrico Mattei Working Papers, Paper no. 823. 35(5), 436–50. Gerten, D., Lucht, W., Ostberg, S., Heinke, J., Kowarsch, M., Kreft, H., Gaillard, J.C. (2010). “Vulnerability, Capacity, And Resilience: Perspec- Kundzewicz, Z.W., Rastgooy, J., Warren, R., and Schellnhuber, H.J. tives For Climate And Development Policy.” Journal of International (2013). “Asynchronous Exposure to Global Warming: Freshwater Development, 22, 218–32. Resources and Terrestrial Ecosystems.” Environmental Research Let- Gaillard, J.C. (2012). “Caste, Ethnicity, Religious Affiliation, and Disaster.” ters, 8(3), 034032. The Routledge Handbook of Hazards and Disaster Risk Reduction, Ghini, R., Bettiol, W., and Hamada, E. (2011). “Diseases in Tropical and edited by B. Wisner, J. Gaillard, and I. Kelman, Routledge: New York. Plantation Crops as affected by Climate Changes: Current Knowledge Gaillard, J.C., and Le Masson, V. (2007). “Traditional Societies’ Response and Perspectives.” Plant Pathology, 60(1), 122–32. to Volcanic Hazards in the Philippines.” Mountain Research and Ghoneim, E. (2009). “A Remote Sensing Study of Some Impacts of Global Development, 27(4), 313–17. Warming on the Arab Region.” Arab Environment: Climate Change. Galbraith, D., Levy, P.E., Sitch, S., Huntingford, C., Cox, P., Williams, M. Impact of Climate Change on Arab Countries, edited by M. K. Tolba and Meir, P. (2010). “Multiple Mechanisms of Amazonian Forest and N. W. Saab, Arab Forum for Environment and Development Biomass Losses in Three Dynamic Global Vegetation Models under (AFED): Beirut. Climate Change.” The New Phytologist, 187(3), 647–65. Giannakopoulos, C., Kostopoulou, E., Hadjinicolaou, P., Hatzaki, M., Karali, Galindo, L.M., Samaniego, J., Alatorre, J.E., and Ferrer, J.A. (2013). “Cambio A., Lelieveld, J., and Lange, M.A. (2013). “Impacts of Climate Change Climático : Agricultura y Pobreza en América Latina.” Santiago de Chile. Over the Eastern Mediterranean and Middle East Region Using the Gao, X., and Giorgi, F. (2008). “Increased Aridity in the Mediterranean Hadley Centre PRECIS RCM.” Advances in Meteorology, Climatology Region under Greenhouse Gas Forcing Estimated from High Resolution and Atmospheric Physics, Springer-Verlag Berlin Heidelberg. Simulations with a Regional Climate Model.” Global and Planetary Giannakopoulos, C., Le Sager, P., Bindi, M., Moriondo, M., Kostopoulou, Change, 62(3), 195–209. E., and Goodess, C.M. (2009a). “Climatic Changes and Associated Gao, X., Pal, J.S., and Giorgi, F. (2006). “Projected Changes in Mean and Impacts in the Mediterranean Resulting from a 2°C Global Warming.” Extreme Precipitation over the Mediterranean Region from a High Global and Planetary Change, 68(3), 209–24. Resolution Double Nested RCM Simulation.” Geophysical Research Giannakopoulos, C., Le Sager, P., Bindi, M., Moriondo, M., Kostopoulou, Letters, 33(3), L03706. E., and Goodess, C.M. (2009b). “Climatic Changes and Associated García, N.O., and Vargas, W.M. (1998). “The Temporal Climatic Variabil- Impacts in the Mediterranean Resulting from a 2°C Global Warming.” ity in the Río de la Plata Basin Displayed by the River Discharges.” Global and Planetary Change, 68(3), 209–24. Climatic Change, 38, 359–79. Giannini, A., Biasutti, M., Held, I.M., and Sobel, A.H. (2008). “A Global García-López, J.M., and Allué, C. (2013). “Modelling Future No-analogue Perspective on African Climate.” Climatic Change, 90(4), 359–83. Climate Distributions: A World-wide Phytoclimatic Niche-based Sur- Giesen, R.H., and Oerlemans, J. (2013). “Climate-model-induced Differ- vey.” Global and Planetary Change, 101, 1–11. ences in the 21st Century Global and Regional Glacier Contributions García-Ruiz, J.M., López-Moreno, J.I., Vicente-Serrano, S.M., Lasanta–Mar- to Sea-level Rise.” Climate Dynamics. tínez, T., and Beguería, S. (2011). “Mediterranean Water Resources in Gila, O.A., De Maturana Giéguez, V.L., and Zaratiegui, A.U. (2009). a Global Change Scenario.” Earth-Science Reviews, 105(3-4), 121–39. “EACH-FOR Environmental Change and Forced Migration Scenarios. Gardner, A.S., Moholdt, G., Cogley, J.G., Wouters, B., Arendt, A.A, Wahr, J., Western Sahara Case Study Report. Deliverable D2.5.2.2.” UPV-EHU: Berthier, E., Hock, R., Pfeffer, W.T., Kaser, G., Ligtenberg, S.R.M., Vitoria-Gasteiz, Spain. 250 B I BLIOGRA P HY Glantz, M., and Jamieson, D. (2000). “Societal Response to Hurricane Gregory, J.M., White, N.J., Church, J.A., Bierkens, M.F.P., Box, J.E., van Mitch and the Intra-versus Intergenerational Equity Issues: Whose den Broeke, M.R., Cogley, J.G., Fettweis, X., Hanna, E., Huybrechts, P., Norms Should Apply?” Risk Analysis, 20(6), 869–82. Konikow, L.F., Leclercq, P.W., Marzeion, B., Oerlemans, J., Tamisiea, Glasser, N.F., Harrison, S., Jansson, K.N., Anderson, K., and Cowley, A. M.E., Wada, Y., Wake, L.M., and van de Wal R.S.W. (2013). “Twentieth- (2011). “Global Sea-level Contribution from the Patagonian Icefields Century Global-Mean Sea-Level Rise: Is the Whole Greater than the since the Little Ice Age Maximum.” Nature Geoscience, 4(5), 303–7. Sum of the Parts?” Journal of Climate, 26(13), 4476–99. Gocic, M., and Trajkovic, S. (2013). “Analysis of Precipitation and Drought Data Greve, P., Orlowsky, B., and Mueller B. (2014). “Global Assessment of Trends in Serbia ver the Period 1980–2010.” Journal of Hydrology, 494, 32–42. in Wetting and Drying Over Land.” Nature Geoscience, 4(September). Gocic, M., and Trajkovic, S. (2014). “Spatiotemporal Characteristics of Griffin, D. (2007). “Atmospheric Movement of Microorganisms in Clouds Drought in Serbia.” Journal of Hydrology, 510, 110–23. of Desert Dust and Implications for Human Health.” Clinical Micro- Goetz, S.J., Mack, M.C., Gurney, K.R., Randerson, J.T., and Houghton, R.A. biology Reviews, 30(3), 459–77. (2007). “Ecosystem Responses to Recent Climate Change and Fire Dis- Grinsted, A., Moore, J.C., and Jevrejeva, S. (2012). “Homogeneous Record turbance at Northern High Latitudes: Observations and Model Results of Atlantic Hurricane Surge Threat since 1923.” 109(48). Contrasting Northern Eurasia and North America.” Environmental Grinsted, A., Moore, J.C., and Jevrejeva, S. (2013). “Projected Atlantic Research Letters, 2(4), 045031. Hurricane Surge Threat from Rising Temperatures.” Proceedings of Gomes, A.F., Nobre, A.A., and Cruz, O.G. (2012). “Temporal Analysis the National Academy of Sciences of the United States of America, of the Relationship between Dengue and Meteorological Variable in 110(14), 5369–73. the City of Rio de Janeiro, Brazil, 2001-2009.” Cad. Saúde Pública, Grunewald, K., and Scheithauer, J. (2010). “Europe’s Southernmost 28(11), 2189–97. Glaciers: Response and Adaptation to Climate Change.” Journal of Gomez, C., Rodriguez-Morales, A.J., and Franco-Paredes, C. (2006). Glaciology, 56(195), 129–42. “Impact of Climate Variability in the Occurrence of Leishmaniasis in Gualdi, S., Somot, S., Li, L., Artale, V., Adani, M., Bellucci, A., Braun, A., Bolivia.” American Journal of Tropical Medicine and Hygiene, 75(5). Calmanti, S., Carillo, A., Dell’Aquila, A., Déqué, M., Dubois, C., Gómez-Mendoza, L., and Arriaga, L. (2007). “Modeling the Effect of Cli- Elizalde, A., Harzallah, A., Jacob, D., L’Hévéder, B., May, W., Oddo, P., mate Change on the Distribution of Oak and Pine Species of Mexico.” Ruti, P., Sanna, A., Sannino, G., Scoccimarro, E., Sevault, F. and Conservation biology: the journal of the Society for Conservation Navarra, A. (2013). “The CIRCE Simulations: Regional Climate Change Biology, 21(6), 1545–55. Projections with Realistic Representation of the Mediterranean Sea.” Gomis, D., Tsimplis, M.N., Martín-Míguez, B., Ratsimandresy, A.W., Bulletin of the American Meteorological Society, 94(1), 65–81. García-Lafuente, J., and Josey, S.A. (2006). “Mediterranean Sea Level Gubert, F., and Nordman, C.J. (2010). “Migration Trends in North Africa and Barotropic Flow through the Strait of Gibraltar for the Period Focus on Morocco, Tunisia, and Algeria.” OECD Journal: General 1958–2001 and Reconstructed Since 1659.” Journal of Geophysical Papers, 4/2009. Research, 111(C11), C11005. Guilyardi, E., Bellenger, H., Collins, M., Ferrett, S., Cai, W., and Wit- Good, P., Jones, C., Lowe, J., Betts, R., and Gedney, N. (2013). “Comparing tenberg, A. (2012). “A First Look at ENSO in CMIP5.” 17(58), 29–32. Tropical Forest Projections from Two Generations of Hadley Centre Guimberteau, M., Ronchail, J., Espinoza, J.C., Lengaigne, M., Sultan, B., Earth System Models, HadGEM2-ES and HadCM3LC.” Journal of Polcher, J., Drapeau, G., Guyot, J.L., Ducharne, A., and Ciais, P. Climate, 26(2), 495–511. (2013). “Future Changes in Precipitation and Impacts on Extreme Gornall, J., Betts, R., Burke, E., Clark, R., Camp, J., Willett, K., and Streamflow over Amazonian Sub-basins,” Environmental Research Wiltshire, A. (2010). “Implications of Climate Change for Agricultural Letters, 8(1), 1–13. Productivity in the Early Twenty-first Century.” Philosophical transac- Guis, H., Caminade, C., Calvete, C., Morse, A.P., Tran, A., and Baylis, M. tions of the Royal Society of London. Series B, Biological sciences, (2012). “Modelling the Effects of Past and Future Climate on the Risk 365(1554), 2973–89. of Bluetongue Emergence in Europe.” Journal of the Royal Society, Gössling, S., Hall, C.M., and Scott, D. (2009). “The Challenges of Tour- Interface/the Royal Society, 9(67), 339–50. ism as a Development Strategy in an Era of Global Climate Change.” Gumpenberger, M., Vohland, K., Heyder, U., Poulter, B., Macey, K., Ram- Rethinking Development in a Carbon-constrained World. Development mig, A., Popp, A., and Cramer, W. (2010). “Predicting Pan-tropical Cooperation and Climate Change, edited by E. Palosuo, Ministry of Climate-change-induced Forest Stock Gains and Losses—Implications Foreign Affairs: Helsinki. for REDD.” Environmental Research Letters, 5(1), 014013. Gössling, S., Scott, D., Hall, C.M., Ceron, J.P., and Dubois, G. (2012). Gustafson, E.J., Shvidenko, A.Z., Sturtevant, B.R., and Scheller, R.M. (2010). “Consumer Behaviour and Demand Response of Tourists to Climate “Predicting Global Change Effects on Forest Biomass and Composition Change.” Annals of Tourism Research, 39(1), 36–58. in South-central Siberia.” Ecological Applications: A Publication of Goudie, A.S. (2009). “Dust Storms: Recent Developments.” Journal of the Ecological Society of America, 20(3), 700–715. Environmental Management, 90(1), 89–94. Habib, R.R., El Zein, K., and Ghanawi, J. (2010). “Climate Change and Government of the Republic of Tajikistan (2011). “Tajikistan: Strategic Health Research in the Eastern Mediterranean Region.” EcoHealth, 7, Programme for Climate Resilience Prepared under the Pilot Program 156–75. for Climate Resilience.” PPRC, Climate Investment Funds (CIF), Gov- Hagg, W., Hoelzle, M., Wagner, S., Mayr, E., and Klose, Z. (2013). “Glacier ernment of the Republic of Tajikistan. and Runoff Changes in the Rukhk Catchment, Upper Amu-Darya Basin Granit, J., Jägerskog, A., Löfgren, R., Bullock, A., de Gooijer, G., Pettigrew, until 2050.” Global and Planetary Change, 110, 62–73. S., and Lindstrom, A. (2010). “Regional Water Intelligence Report Hailo, D., Osório, R.G., and Tsukada, R. (2009). “Privatisation and Rena- Central Asia.” Regional Water Intelligence Reports No. 15, Water tionalisation: What Went Wrong in Bolivia’s Water Sector?” United Governance Facility, Stockholm. Nations Development Programme, International Policy Centre for Granit, J., and Löfgren, R. (2010). “Water and Energy Linkages in the Inclusive Growth (IPC-IG), Brasilia. Middle East,” SIWI: Stockholm. Hailo, D., and Portia, H. (2008). “Utility Provision: Contract Design in Grant, A., Burger, N. and Wodon, Q. (2014). “Climate-induced Migration the Interest of the Poor Latin America.” Policy Research Brief no. 10, in the MENA Region: Results from the Qualitative Fieldwork.” MPRA International Policy Centre for Inclusive Growth, Brasilia. Paper No. 56936, World Bank: Washington, DC. Haines, A., Kovats, R.S., Campbell-Lendrum, D., and Corvalan, C. (2006). Gregg, W.W. (2003). “Ocean Primary Production and Climate: Global “Climate Change and Human Health: Impacts, Vulnerability, and Decadal Changes.” Geophysical Research Letters, 30(15), 1809. Public Health.” Public health, 120(7), 585–96. 251 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Haldén, P. (2007). “The Geopolitics of Climate Change. Challenges to the Harper, C., Marcus, R., and Moore, K. (2003). “Enduring Poverty and the International System.” User Report FOI-R--2377--SE, FOI—Totalförs- Conditions of Childhood: Lifecourse and Intergenerational Poverty varets forskningsinstitut: Stockholm. Transmissions.” World Development, 31(3), 535–54. Hall, G., and Patrinos, H.A. (2005). “Indigenous Peoples, Poverty, and Harris, K., and Hawrylyshyn, K. (2012). “Climate Extremes and Child Rights Human Development in Latin America: 1994-2004.” Washington, DC. in South Asia: A Neglected Priority.” ODI Project Briefings, London. Hall, T.C., Sealy, A.M., Stephenson, T.S., Kusunoki, S., Taylor, M.A., Harris, K., Keen, D., and Mitchell, T. (2013). “When Disasters and Conflicts Chen A.A., and Kitoh, A. (2012). “Future Climate of the Caribbean Collide: Improving Links between Disaster Resilience and Conflict from a Super-high-resolution Atmospheric General Circulation Model.” Prevention.” Overseas Development Institute: London. Theoretical and Applied Climatology, 113(1-2), 271–87. Harvey, B.P., Gwynn-Jones, D., and Moore, P.J. (2013). “Meta-analysis Hallegatte, S. (2007). “The Use of Synthetic Hurricane Tracks in Risk Reveals Complex Marine Biological Responses to the Interactive Analysis and Climate Change Damage Assessment.” Journal of Applied Effects of Ocean Acidification and Warming.” Ecology and Evolution, Meteorology and Climatology, 46(11), 1956–66. 3(4), 1016–30. Hallegatte, S., Green, C., Nicholls, R.J., and Corfee-Morlot, J. (2013). Hassaan, M.A., and Abdrabo, M.A. (2013). “Vulnerability of the Nile “Future Flood Losses in Major Coastal Cities.” Nature Climate Change, Delta Coastal Areas to Inundation by sea-level Rise.” Environmental 3(9), 802–6. Monitoring and Assessment, 185(8), 6607–16. Hallegatte, S., Heal, G., Fay, M., and Treguer, D. (2011). “From Growth to Hatfield, J.L., Boote, K.J., Kimball, B.A., Ziska, L.H., Izaurralde, R.C., Green Growth: A Framework.” World Bank: Washington DC. Ort, D., Thomson, A.M., and Wolfe, D. (2011). “Climate Impacts on Hallegatte, S., Shah, A., Brown, C., Lempert, R., and Gill, S. (2012). Agriculture: Implications for Crop Production.” Agronomy Journal, “Investment Decision Making Under Deep Uncertainty: Application 103(2), 351–70. to Climate Change.” Policy Research Working Papers 6193, 41. Hayutin, A. (2006). “The Implications of Global Aging and Population Hamilton, J.M., Maddison, D.J., and Tol, R.S.J. (2005). “Effects of Climate Trends: Economic Prosperity, National Security, and Social Well- Change on International Tourism.” Climate Research, 29(3), 245–54. Being.” Stanford. Hamududu, B., and Killingtveit, A. (2012). “Assessing Climate Change Hazell, P., and Wood, S. (2008). “Drivers of Change in Global Agriculture.” Impacts on Global Hydropower.” Energies, 5(12), 305–22. Philosophical Transactions of the Royal Society of London. Series B, Hamza, M.A., El Faskaoui, B., and Fernmin, A. (2009). “EACH-FOR Envi- Biological Sciences, 363(1491), 495–515. ronmental Change and Forced Migration Scenarios. Morocco Case He, B., Lü, A., Wu, J., Zhao, L., and Liu, M. (2011). “Drought Hazard Study Report. Migration and Environmental Change in Morocco: The Assessment and Spatial Characteristics Analysis in China.” Journal Case of Rural Oasis Villages in the Middle Drâa Valley. Deliverable of Geographical Sciences, 21(2), 235–49. D2.5.2.3.” Erasmus University Rotterdam. Heinrichs, D., and Krellenberg, K. (2011). “Resilient Cities.” Resilient Cities: Han, S.R., Guikema, S.D., and Quiring, S.M. (2009). “Improving the Predic- Cities and Adaptation to Climate Change Proceedings of the Global tive Accuracy of Hurricane Power Outage Forecasts Using Generalized Forum 2010, edited by K. Otto-Zimmermann, Springer Netherlands: Additive Models.” Risk Analysis: An Official Publication of the Society Dordrecht. for Risk Analysis, 29(10), 1443–53. Hellmer, H.H., Kauker, F., Timmermann, R., Determann, J., and Rae, J. Han, W., Meehl, G.A., Rajagopalan, B., Fasullo, J.T., Hu, A., Lin, J., Large, (2012). “Twenty-first-century Warming of a Large Antarctic Ice-shelf W.G., Wang, J., Quan, X.W., Trenary, L.L., Wallcraft, A., Shinoda, T., Cavity by a Redirected Coastal Current.” Nature, 485(7397), 225–28. and Yeager, S. (2010). “Patterns of Indian Ocean Sea-level Change in HelpAge International (2012). “Strategies for an Ageing World under a a Warming Climate.” Nature Geoscience, 3(8), 546–50. Changing Climate.” HelpAge International: London. Handicap International (2008), “Mainstreaming Disability in Community Hemming, D., Buontempo, C., Burke, E., Collins, M., and Kaye, N. (2010). Based Disaster Risk Reduction. A Training Manual for Trainers and “How Uncertain are Climate Model Projections of Water Availability Field Practitioners,” Handicap International: New Delhi. Indicators across the Middle East?” Philosophical Transactions. Series A, Hanjra, M.A., and Qureshi, M.E. (2010). “Global Water Crisis and Future Mathematical, Physical, and Engineering Sciences, 368(1931), 5117–35. Food Security in an Era of Climate Change.” Food Policy, 35(5), 365–77. Hempel, S., Frieler, K., Warszawski, L., Schewe, J., and Piontek, F. (2013). Hanna, E.G., Kjellstrom, T., Bennett, C., and Dear, K. (2011), “Climate “A trend-preserving Bias Correction—the ISI-MIP Approach.” Earth Change and Rising Heat: Population Health Implications for Working System Dynamics, 4(2), 219–36. People in Australia.” Asia-Pacific Journal of Public Health/Asia-Pacific Hendrix, C.S., and Salehyan, I. (2012). “Climate Change, Rainfall, and Academic Consortium for Public Health, 23(2 Suppl), 14S–26. Social Conflict in Africa.” Journal of Peace Research, 49(1), 35–50. Hansen, J., Sato, M., and Ruedy, R. (2012). “Perception of Climate Change.” Hereher, M.E. (2010). “Vulnerability of the Nile Delta to Sea-level Rise: Proc. Nat. Ac. Sc., 109(37), E2415–E2423. An Assessment Using Remote Sensing.” Geomatics, Natural Hazards Hansen, M.C., Potapov, P.V, Moore, R., Hancher, M., Turubanova, S.A., and Risk, 1(4), 315–21. Tyukavina, A., Thau, D., Stehman, S.V, Goetz, S.J., Loveland, T.R., Herrmann, S.M., and Hutchinson, C.F. (2005). “The Changing Contexts Kommareddy, A., Egorov, A., Chini, L., Justice, C.O., and Townsh- of the Desertification Debate.” Journal of Arid Environments, 63(3), end, J.R.G. (2013). “High-resolution Global Maps of 21st-century 538–55. Forest Cover Change.” Science, 342(6160), 850–53. Hertel, T., Burke, M., and Lobell, D. (2010). “The Poverty Implications of Hanson, S., Nicholls, R., Ranger, N., Hallegatte, S., Corfee-Morlot, J., Her- Climate-Induced Crop Yield Changes by 2030.” Global Environmental weijer, C., and Chateau J. (2011). “A Global Ranking of Port Cities with Change, 20(4), 577–85. High Exposure to Climate Extremes.” Climatic Change, 104(1), 89–111. Hewitt, K. (1997). Regions of Risk: A Geographical Introduction to Disas- Hardoy, J., Mitlin, D., and Satterthwaite, D. (2001). Environmental Problems ters. Addison Wesley Longman: Essex. in an Urbanizing World. Earthscan Publications: London. Heyder, U., Schaphoff, S., Gerten, D., and Lucht, W. (2011). “Risk of Severe Hardoy, J., and Pandiella, G. (2009). “Urban Poverty and Vulnerability to Climate Change Impact on the Terrestrial Biosphere.” Environmental Climate Change in Latin America.” Environment and Urbanization, Research Letters, 6(3), 034036. 21, 203. Hidalgo, H.G., Amador, J.A., Alfaro, E.J., and Quesada, B. (2013). “Hydro- Hardoy, J., and Romero Lankao, P. (2011). “Latin American Cities and logical Climate Change Projections for Central America.” Journal of Climate Change: Challenges and Options to Mitigation and Adapta- Hydrology, 495, 94–112. tion Responses.” Current Opinion in Environmental Sustainability, Higgins, S., and Scheiter, S. (2012). “Atmospheric CO2 Forces Abrupt Veg- 3(3), 158–63. etation Shifts Locally, but Not Globally.” Nature, 488(7410), 209–12. 252 B I BLIOGRA P HY Hijioka, Y., Lin, E., Pereira, J.J., Corlett, R.T., Cui, X., Insarov, G., Lasco, R., Van Hooidonk, R., Maynard, J.A., and Planes, S. (2013). “Temporary Lindgren, E., and Surjan, A. (2014). “Regional Aspects. Chapter 24— Refugia for Coral Reefs in a Warming World.” Nature Climate Change, Asia.” Climate Change 2014: Impacts, Adaptation, and Vulnerability. 3(5), 508–11. Contribution of Working Group II to the Fifth Assessment Report of Horton, B.P., Rahmstorf, S., Engelhart, S.E., and Kemp, A.C. (2014). “Expert the Intergovernmental Panel on Climate Change. Final Draft, IPCC Assessment of Sea-level Rise by AD 2100 and AD 2300.” Quaternary AR5 WGII, Cambridge University Press: Cambridge and New York, NY. Science Reviews, 84, 1–6. Hinkel, J., Brown, S., Exner, L., Nicholls, R.J., Vafeidis, A.T., and Hoyos, I., Baquero-Bernal, A., and Hagemann, S. (2013). “How Accurately Kebede, A.S. (2012). “Sea-level Rise Impacts on Africa and the Effects Are Climatological Characteristics and Surface Water and Energy Bal- of Mitigation and Adaptation: An Application of DIVA.” Regional ances Represented for the Colombian Caribbean Catchment Basin?” Environmental Change, 12(1), 207–24. Climate Dynamics, 41(5-6), 1269–90. Hinkel, J., Lincke, D., Vafeidis, A.T., Perrette, M., Nicholls, R.J., Tol, R.S.J., Hsiang, S.M., Burke, M., and Miguel, E. (2013). “Quantifying the Influ- Marzeion, B., Fettweis, X., Ionescu, C., and Levermann, A. (2014). ence of Climate on Human Conflict.” Science, 341(6151), 1235367. “Coastal Flood Damage and Adaptation Costs under 21st Century Hsiang, S.M., and Meng, K.C. (2014). “Reconciling Disagreement Over Sea-level Rise.” Proceedings of the National Academy of Sciences. Climate-conflict Results in Africa.” Proceedings of the National Acad- Hirabayashi, Y., Mahendran, R., Koirala, S., Konoshima, L., Yamazaki, D., emy of Sciences, (1), 2–5. Watanabe, S., Kim, H., and Kanae, S. (2013). “Global Flood Risk Under Hsiang, S.M., Meng, K.C., and Cane, M.A. (2011). “Civil Conflicts are Climate Change.” Nature Climate Change, 3(9), 816–21. Associated with the Global Climate.” Nature, 476(7361), 438–41. Hirata, R., and Conicelli, B.P. (2012). “Groundwater Resources in Brazil: Huete, A.R., Didan, K., Shimabukuro, Y.E., Ratana, P., Saleska, S.R., A Review of Possible Impacts caused by Climate Change.” Anais da Hutyra, L.R., Yang, W., Nemani, R.R., and Myneni, R. (2006). “Ama- Academia Brasileira de Ciências, 84(2), 297–312. zon Rainforests Green-up with Sunlight in Dry Season.” Geophysical Hirota, M., Nobre, C., Oyama, M.D., and Bustamante, M.M.C. (2010). “The Research Letters, 33(6), L06405. Climatic Sensitivity of the Forest, Savanna, and Forest-Savanna Transi- Hughes, G. (2012). “Climate Change in the West Balkans.” Environment tion in Tropical South America.” The New phytologist, 187(3), 707–19. and Security Initivative (ENVSEC). Hodbod, J., and Tomei, J. (2013). “Demystifying the Social Impacts of Hugo, G. (2011). “Future Demographic Change and its Interactions with Migra- Biofuels at Local Levels: Where is the Evidence?” Geography Com- tion and Climate Change.” Global Environmental Change, 21, S21–S33. pass, 7(7), 478–88. Hunt, A., and Watkiss, P. (2011). “Climate Change Impacts and Adaptation Hoddinott, J., and Kinsey, B. (2001). “Child Growth in the Time Of Drought.” in Cities : A Review of the Literature.” Climatic Change, 104, 13–49. Oxford Bulletin of Economics and Statistics, 63, 409–36. Huntingford, C., Fisher, R.A., Mercado, L., Booth, B.B.B., Sitch, S., Harris, P.P., Hoegh-Guldberg, O. (1999). “Climate Change, Coral Bleaching, and the Cox, P.M., Jones, C.D., Betts, R.A., Malhi, Y., Harris, G.R., Collins, M., Future of the World’s Coral Reefs.” Marine and Freshwater Research, and Moorcroft, P. (2008). “Towards Quantifying Uncertainty in Predic- 50(8), 839. tions of Amazon ‘Dieback.’” Philosophical Transactions of the Royal Hoegh-Guldberg, O., Cai, R., Brewer, P.G., Fabry, V.J., Hilmi, K., Jung, S., Society of London. Series B, Biological Sciences, 363(1498), 1857–64. Poloczanska, E., and Sundby, S. (2014). “The Ocean.” Climate Change Huntingford, C., Zelazowski, P., Galbraith, D., Mercado, L.M., Sitch, S., 2014: Impacts, Adaptation, and Vulnerability. Contribution of Working Fisher, R., Lomas, M., Walker, A.P., Jones, C.D., Booth, B.B.B., Group II to the Fifth Assessment Report of the Intergovernmental Panel Malhi, Y., Hemming, D., Kay, G., Good, P., Lewis, S.L., Phillips, O.L., on Climate Change. Chapter 30. Final Draft, edited by L. Omar and Atkin, O.K., Lloyd, J., Gloor, E., Zaragoza-Castells, J., Meir, P., Betts, R., C. Turley, IPCC AR5 WGII, Cambridge University Press: Cambridge. Harris, P.P., Nobre, C., Marengo, J., and Cox, P.M. (2013). “Simulated Hoegh-Guldberg, O., Mumby, P.J., Hooten, A.J., Steneck, R.S., Green- Resilience of Tropical Rainforests to CO2-induced Climate Change.” field, P., Gomez, E., Harvell, C.D., Sale, P.F., Edwards, A.J., Cal- Nature Geoscience, 6(4), 268–73. deira, K., Knowlton, N., Eakin, C.M., Iglesias-Prieto, R., Muthiga, N., Hurtado-Diaz, M., Riojas-Rodriguez, H., Rothenburg, S.J., Gomez-Dantes, Bradbury, R.H., Dubi, A., and Hatziolos, M.E. (2007). “Coral Reefs H. and Cifuentes, E. (2007). “Impact of Climate Variability on the Under Rapid Climate Change and Ocean Acidification.” Science, Incidence of Dengue in Mexico.” Tropical Medicine and International 318(5857), 1737–42. Health, 12(11), 1327–37. Hoerling, M., Eischeid, J., Perlwitz, J., Quan, X., Zhang, T., and Pegion, P. Huynen, M.M.T.E., Martens, P., and Akin, S.M. (2013). “Climate Change: (2012). “On the Increased Frequency of Mediterranean Drought.” An Amplifier of Existing Health Risks in Developing Countries.” Envi- Journal of Climate, 25(6), 2146–61. ronment, Development and Sustainability, 15(6), 1425–42. Hof, C., Levinsky, I., Araújo, M.B., and Rahbek, C. (2011), “Rethinking Hyman, T.A. (2013). “Assessing the Vulnerability of Beach Tourism and Species’ Ability to Cope with Rapid Climate Change.” Global Change Non-beach Tourism to Climate Change: A Case Study from Jamaica.” Biology, 17(9), 2987–90. Journal of Sustainable Tourism, 0(0), 1–19. Hoffman, M., and Grigera, A.I. (2013). “Climate Change, Migration, and IAASTD (2009). Agriculture at a Crossroads. Latin America and the Carib- Conflict in the Amazon and the Andes. Rising Tensions and Policy bean. IAASTD: Washington, DC. Options in South America.” Center for American Progress: Wash- Ibarrarán, M.E. (2011). “Climate’s Long-term Impacts on Mexico’s City ington D.C. Urban Infrastructure,” UN-Habitat: Nairobi. Hoffman, M., and Jamal, A. (2012). “The Youth and the Arab Spring: Icduygu, A., and Sert, D. (2011). “Migration and Global Environmental Cohort Differences and Similarities.” Middle East Law and Gover- Change. PD13: International Migration in the Mashriq Countries: Past nance, 4(1), 168–88. Trends, Future Potentials, and Determinants—Possible Implications Hoffman, M., and Werz, M. (2013). “Climate Change, Migration and for Environmental Migration?” Foresight Project, Government Office Conflict.” ISN, Center for Security Studies, ETH Zurich. for Science: London. Högy, P., Brunnbauer, M., Koehler, P., Schwadorf, K., Breuer, J., Franza- Ichii, K., Kondo, M., Okabe, Y., Ueyama, M., Kobayashi, H., Lee, S.J., ring, J., Zhunusbayeva, D., and Fangmeier, A. (2013). “Grain Quality Saigusa, N., Zhu, Z., and Myneni, R. (2013). “Recent Changes in Characteristics of Spring Wheat (Triticum aestivum) as Affected by Terrestrial Gross Primary Productivity in Asia from 1982 to 2011.” Free-air CO2 Enrichment.” Environmental and Experimental Botany, Remote Sensing, 5(11), 6043–62. 88, 11–18. ICO (2013). “Report on the Outbreak of Coffee Leaf Rust in Central Holt, E. (2014). “Disease Outbreaks Predicted in Flood-ravaged Balkans.” America and Action Plan to Combat the Pest.” International Coffee The Lancet, 383, 1959. Organization: London. 253 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL ICSU-LAC (2010). Science for a Better Life: Developing Regional Scientific K. L. Ebi, Y. O. Estrada, R. C. Genova, B. Girma, E. S. Kissel, A. N. Programs in Priority Areas for Latin America and the Caribbean. Vol- Levy, S. MacCracken, P. R. Mastrandrea, and L. L. White, IPCC AR5 ume 1. Edited by M. K. T. Arroyo, R. Dirzo, J. C. Castillas, F. Cejas, WGII, Cambridge University Press: Cambridge, UK and New York, NY. and C. A. Joly, ICSU—LAC/CONACYT: Rio de Janeiro and Mexico City. IPCC (2014c). “Summary for Policymakers,” Climate Change 2014: Impacts, IFAD (2013). “IFAD in Latin America and the Caribbean.” International Adaptation, and Vulnerability. Contribution of Working Group II to the Fund for Agricultural Development [online], Rome, (accessed March 25, Fifth Assessment Report of the Intergovernmental Panel on Climate 2014). available at http://www.ifad.org/operations/projects/regions/ Change, edited by C. B. Field, V. R. Barros, D. J. Dokken, K. J. Mach, PL/index.htm. M. D. Mastrandrea, T. E. Bilir, M. Chatterjee, K. L. Ebi, Y. O. Estrada, IFPRI (2012). “Global Food Policy Report.” International Food Policy R. C. Genova, B. Girma, E. S. Kissel, A. N. Levy, S. MacCracken, P. R. Research Institute (IFPRI): Washington, DC. Mastrandrea, and L. L. White, IPCC AR5 WGII, Cambridge University ILC, CIRAD, CDE, GIGA, and GIZ (2014). “The Land Matrix Global Obser- Press: Cambridge, UK and New York, NY. vatory.” Available at http://landmatrix.org/en/. IPCC (2014d). Climate Change 2014: Mitigation of Climate Change. ILO (2012). Working Towards Sustainable Development. Opportunities for Contribution of Working Group III to the Fifth Assessment Report Social Inclusion in a Green Economy. International Labor Organiza- of the Intergovernmental Panel on Climate Change. Edited by O. tion: Geneva. Edenhofer, R. Pichs-Madruga, Y. Sokona, E. Farahani, S. Kadner, K. Imbach, P., Molina, L., Locatelli, B., Roupsard, O., Mahé, G., Neilson, R., Seyboth, A. Adler, I. Baum, S. Brunner, P. Eickemeier, B. Kriemann, Corrales, L., Scholze, M., and Ciais, P. (2012). “Modeling Potential J. Savolainen, S. Schlömer, C. von Stechow, T. Zwickel, and J. C. Equilibrium States of Vegetation and Terrestrial Water Cycle of Meso- Minx, IPCC AR5 WGIII, Cambridge University Press: Cambridge america under Climate Change Scenarios.” Journal of Hydrometeorol- and New York, NY. ogy, 13(2), 665–80. Isaac, M., and van Vuuren, D.P. (2009). “Modeling Global Residential Sec- Immerzeel, W., Droogers, P., Terink, W., Hoogeveen, J., Hellegers, P., tor Energy Demand for Heating and Air Conditioning in the Context Bierkens, M., and van Beek, R. (2011). “Middle East and North Africa of Climate Change.” Energy Policy, 37(2), 507–21. Water Outlook.” Wageningen, Netherlands. Isaev, A.S., and Korovin, G.N. (2014). “Forests as a National Treasure of Immerzeel, W.W., and Bierkens, M.F.P. (2012). “Asia’s Water Balance.” Russia.” Contemporary Problems of Ecology, 6(7), 677–82. Nature Geoscience, 5(12), 841–42. Islami, B., Kamberi, M., Bruci, E.D., and Fida, E. (2009). “Albania’s Sec- International Energy Agency (2012). “World Energy Outlook 2012: Execu- ond National Communication to the Conference of Parties under the tive Summary.” IEA: Paris. United Nations Framework Convention on Climate Change.” Ministry International Energy Agency (2013). “World Energy Outlook Special of Environment, Forestry, and Water Administration: Tirana, Albania. Report—Redrawing the Energy-Climate Map.” IEA: Paris. Ivanic, M., and Martin, W. (2008). “Implications of Higher Global Food IOM (2011). “IOM Central Asia Operational Strategy 2011-2015.”Interna- Prices for Poverty in Low-Income Countries.” Policy Research Working tional Organization for Migration Paper 4594, World Bank: Washington DC. IPCC (2007). Climate Change 2007: Mitigation. Contribution of Working Ivanova G.A., Ivanov V.A., Kukavskaya E.A., and Soja, A.J. (2010). “The Group III to the Fourth Assessment Report of the Intergovernmen- Frequency of Forest Fires in Scots Pine Stands of Tuva, Russia.” Envi- tal Panel on Climate Change. Edited by B. Metz, O. R. Davidson, ronmental Research Letters, 5(1), 015002. P. R. Bosch, R. Dave, and L. A. Meyer, IPCC AR4 WGIII, Cambridge Ivins, E.R., Watkins, M.M., Yuan, D.N., Dietrich, R., Casassa, G., and University Press: Cambridge, UK and New York, NY. Rülke, A. (2011). “On-land Ice Loss and Glacial Isostatic Adjustment IPCC (2012). “Special Report on Managing the Risks of Extreme Events and at the Drake Passage: 2003–2009.” Journal of Geophysical Research, Disasters to Advance Climate Change Adaptation (SREX).” A Special 116(B2), B02403. Report of Working Groups I and II of the Intergovernmental Panel on Izaurralde, R.C., Thomson, A.M., Morgan, J.A. , Fay, P.A., Polley, H.W., and Climate Change, edited by C. B. Field, V. Barros, T. F. Stocker, and Hatfield, J.L. (2011). “Climate Impacts on Agriculture: Implications for Q. Dahe, Cambridge University Press: Cambridge, UK. Forage and Rangeland Production.” Agronomy Journal, 103(2), 371. IPCC (2013a). “Summary for Policymakers.” Climate Change 2013: The Jacob, T., Wahr, J., Pfeffer, W.T., and Swenson, S. (2012). “Recent Physical Science Basis. Contribution of Working Group I to the Fifth Contributions of Glaciers and Ice Caps to Sea-level Rise.” Nature, Assessment Report of the Intergovernmental Panel on Climate Change, 482(7386), 514–18. edited by T. F. Stocker, D. Qin, G.-K. Plattner, M. Tignor, S. K. Allen, J. Jacoby, H., Rabassa, M., and Skoufias, E. (2011). “Distributional Implica- Boschung, A. Nauels, Y. Xia, V. Bex, and P. M. Midgley, IPCC AR5 tions of Climate Change in India.” Policy Research Working Paper WGI, Cambridge University Press: Cambridge UK and New York, NY. 5623, World Bank: Washington, DC. IPCC (2013b). Climate Change 2013: The Physical Science Basis. Con- Jaeger, J., Fruehmann, J., and Gemenne, F., (2009). “Environment and tribution of Working Group I to the Fifth Assessment Report of the Migration Scenarios for Ferghana Valley.” EACH-FOR Environmental Intergovernmental Panel on Climate Change. Edited by T. F. Stocker, D. Change and Forced Migration Scenarios No. 2.2.3. Qin, G.-K. Plattner, M. Tignor, S. K. Allen, J. Boschung, A. Nauels, Jaime, P., and Menéndez, A.N. (2002). “Análisis del Régimen Hidrológico Y. Xia, V. Bex, and P. M. Midgley, Cambridge University Press: Cam- de los Rìos Paraná y Uruguay.” Informe. bridge, UK and New York, NY. Jeong, S.-J., Ho, C.H., Gim, H.J., and Brown, M.E. (2011). “Phenology IPCC (2014a). “Summary for Policymakers.” Climate Change 2014: Miti- Shifts at Start vs. End of Growing Season in Temperate Vegetation over gation of Climate Change. Contribution of Working Group III to the the Northern Hemisphere for the period 1982–2008.” Global Change Fifth Assessment Report of the Intergovernmental Panel on Climate Biology, 17(7), 2385–99. Change, edited by O. Edenhofer, R. Pichs-Madruga, Y. Sokona, E. Jeschke, M., Popp, A., and Lotze-Campen, H. (2012). “Adaptation Options to Farahani, S. Kadner, K. Seyboth, A. Adler, I. Baum, S. Brunner, P. Climate-Induced Glacier Retreat in Bolivia.” Climate Change, Justice and Eickemeier, B. Kriemann, J. Savolainen, S. Schlömer, C. von Stechow, Sustainability, edited by O. Edenhofer, J. Wallacher, H. Lotze-Campen, T. Zwickel, and J. C. Minx, IPCC AR5 WGIII, Cambridge University M. Reder, B. Knopf, and J. Müller, Springer Netherlands: Dordrecht. Press: Cambridge, UK and New York, NY. Jetz, W., Wilcove, D.S., and Dobson, A.P. (2007). “Projected Impacts of IPCC (2014b). Climate Change 2014: Impacts, Adaptation, and Vulner- Climate and Land-use Change on the Global Diversity of Birds.” PLoS ability. Part A: Global and Sectoral Aspects. Contribution of Working Biology, 5(6), e157. Group II to the Fifth Assessment Report of the Intergovernmental Johansson, M.A., Dominici, F., and Glass, G.E. (2009). “Local and Global Panel on Climate Change. Edited by C. B. Field, V. R. Barros, D. J. Effects of Climate on Dengue Transmission in Puerto Rico.” PLoS Negl Dokken, K. J. Mach, M. D. Mastrandrea, T. E. Bilir, M. Chatterjee, Trop Dis, 3(2), e382. 254 B I BLIOGRA P HY Johnstone S., and Mazo, J. (2011). “Global Warming and the Arab Spring.” Kashef, A.A. (1983). “Salt Water Intrusion in the Nile Delta.” Ground Survival, 53(2), 11–17. Water, 21(2), 160–67. Johnstone, S., and Mazo, J. (2013). “Global Warming and the Arab Spring.” Kelly, P.M., and Adger, W.N. (2000). “Theory and Practice in Assessing The Arab Spring and Climate Change. A Climate and Security Correla- Vulnerability to Climate Change and Facilitating Adaptation.” Climatic tion Series, edited by C. E. Werrell and F. Femia, Center for American Change, (47), 325–52. Progress, Stimson Center: Washington, DC. Kerrou, J., Renard, P., and Tarhouni, J. (2010). “Status of the Korba Ground- Jones, A., Breuning-Madsen, H., Brossard, M., Dampha, A. J. D., Dewitte, water Resources (Tunisia): Observations and Three-dimensional Mod- O., Gallali, T., Hallett, S., Jones, R., Kilasara, M.,P. L.R., Micheli, E., elling of Seawater Intrusion.” Hydrogeology Journal, 18(5), 1173–90. Montanarella, L., Spaargaren, O., Thiombiano, L. Van, Ranst, E., Kgope, B., Bond, W.J., and Midgley, G.F. (2009). “Growth Responses of Yemefack, M., and Zougmoré, R. (2013). Soil Atlas of Africa. European African Savanna Trees Implicate Atmospheric [CO2] as a Driver of Commission, Publications Office of the European Union: Luxembourg. Past and Current Changes in Savanna Tree Cover.” Austral Ecology, Jones, C., and Carvalho, L.M. V. (2013). “Climate Change in the South 35(4), 451–63. American Monsoon System: Present Climate and CMIP5 Projections.” Khakimov, P., and Mahmadbekov, M. (2009). “Republic of Tajikistan. Case Journal of Climate, 26(17), 6660–78. Study Report.” United Nations University Institute for Environment Jones, P.D., Lister, D.H., Osborn, T.J., Harpham, C., Salmon, M., and and Human Security: Bonn. Morice, C.P. (2012). “Hemispheric and Large-scale Land-surface Air Khan, A.E., Xun, W.W., Ahsan, H., and Vineis, P. (2011). “Climate Change, Temperature Variations: An Extensive Revision and an Update to Sea-Level Rise, and Health Impacts in Bangladesh.” Environment: 2010.” J. Geophys. Res., 117(D5), D05127. Science and Policy for Sustainable Development, 53(5), 18–33. De Jong, R., de Bruin, S., de Wit, A., Schaepman, M.E., and Dent, D.L. Kharin, V. V., Zwiers, F.W., Zhang, X., and Wehner, M. (2013). “Changes (2011). “Analysis of Monotonic Greening and Browning Trends from in Temperature and Precipitation Extremes in the CMIP5 Ensemble.” Global NDVI Time-series.” Remote Sensing of Environment, 115(2), Climatic Change, 119(2), 345–57. 692–702. Kharuk, V.I., Im, S.T., Oskorbin, P.A., Petrov, I.A., and Ranson, K.J. (2013). Jónsson, G. (2010). “The Environmental Factor in Migration Dynamics—A “Siberian Pine Decline and Mortality in Southern Siberian Mountains.” Review of African Case Studies.” Forest Ecology and Management, 310, 312–20. Joseph, G., and Wodon, Q. (2013). “Is Internal Migration in Yemen Driven Kharuk, V.I., Ranson, K.J., Im, S.T., and Naurzbaev, M.M. (2006). “Forest- by Climate or Socioeconomic Factors?” Review of International Eco- tundra Larch Forests and Climatic Trends.” Russian Journal of Ecology, nomics, 21(2), 295–310. 37(5), 291–98. Joughin, I., Smith, B.E., and Medley, B. (2014). “Marine Ice Sheet Collapse Khattabi, A. (2009). “Impact of Climate Change: Vulnerability and Adapta- Potentially Underway for the Thwaites Glacier Basin, West Antarctica.” tion: Tourism.” Arab Environment: Climate Change, edited by Tolba and Science (New York, N.Y.), 344(May), 735–38. Saab, Arab Foundation for Environment and Development: 121-128. Joyner, T.A., Lukhnova, L., Pazilov, Y., Temiralyeva, G., Hugh-Jones, M.E., Kicklighter, D.W., Cai, Y., Zhuang, Q., Parfenova, E.I., Paltsev, S., Sokolov, Aikimbayev, A., and Blackburn, J.K. (2010). “Modeling the Potential A.P., Melillo, J.M., Reilly, J.M., Tchebakova, N.M., and Lu, X. (2014). Distribution of Bacillus anthracis under Multiple Climate Change “Potential Influence of Climate-induced Vegetation Shifts on Future Scenarios for Kazakhstan.” PloS one, 5(3), e9596. Land Use and associated Land Carbon Fluxes in Northern Eurasia.” JRC (2009). “Salinisation and Sodification. Sustainable Agriculture and Environmental Research Letters, 9, 14. Soil Conservation. Soil Degradation Processes. Fact sheet No. 4.” JRC Kim, H.M., Webster, P.J., and Curry, J.A. (2011). “Modulation of North European Commission: Brussels. Pacific Tropical Cyclone Activity by Three Phases of ENSO.” Journal Juen, I., Kaser, G., and Georges, C. (2007). “Modelling Observed and Future of Climate, 24(6), 1839–49. Runoff from a Glacierized Tropical Catchment (Cordillera Blanca, Kingdom of Morocco (2010). “Seconde Communication Nationale a la Perú).” Global and Planetary Change, 59(1-4), 37–48. Convention Cadre des Nations Unies sur les Changements Clima- Jung, E., Albrecht, B., Prospero, J.M., Jonsson, H.H., and Kreidenweis, tiques.” Rabat. S.M. (2013). “Vertical Structure of Aerosols, Temperature, and Moisture Kingdom of Saudi Arabia (2011). “Second National Communication.” Associated with an Intense African Dust Event Observed over the Kinouchi, T., Liu, T., Mendoza, J., and Asaoka, Y. (2013). “Modeling Eastern Caribbean.” Journal of Geophysical Research: Atmospheres, Glacier Melt and Runoff in a High-altitude Headwater Catchment in 118(10), 4623–43. the Cordillera Real, Andes.” Hydrology and Earth System Sciences Kahn, M. (2005). “The Death Toll from Natural Disasters: The Role of Discussions, 10(11), 13093–144. Income, Geography, and Institutions.” Review of Economics and Kjellstrom, T., Butler, A.J., Lucas, R.M., and Bonita, R. (2010). “Public Health Statistics, 87(2), 271–84. Impact of Global Heating due to Climate Change: Potential Effects on Kallis, G., and Zografos, C. (2012). “Towards an Improved Understanding Chronic Non-communicable Diseases.” Int J Public Health, 55, 97–103. of the Hydroclimatic and Security Nexus.” CLICO Working Paper. Kjellstrom, T., and Crowe, J. (2011). “Climate Change, Workplace Heat Kanatani, K., and Ito, I. (2010). “Desert Dust Exposure is Associated with Exposure, and Occupational Health and Productivity in Central Increased Risk of Asthma Hospitalization in Children.” Am J Respir America.” Int J Occup Environ Health, 17, 270–81. Crit Care Med, 182, 1475–81. Kjellstrom, T., Holmer, I., and Lemke, B. (2009). “Workplace Heat Stress, Kaniewski, D., Van Campo, E., and Weiss, H. (2012). “Drought is a Recurring Health, and Productivity—An Increasing Challenge for Low and Middle- Challenge in the Middle East.” Proceedings of the National Academy income Countries During Climate Change.” Global Health Action. of Sciences of the United States of America, 109(10), 3862–67. Kjellstrom, T., Kovats, S., Lloyd, S.J., Holt, T., and Tol, R.S.J. (2009). “The Karmalkar, A. V, Taylor, M.A., Campbell, J., Stephenson, T., New, M., Direct Impact of Climate Change on Regional Labor Productivity.” Centella, A., Benzanilla, A., and Charlery, J. (2013). “A Review of Archives of Environment and Occupational Health, 64(4), 217–27. Observed and Projected Changes in Climate for the Islands in the Kjellstrom, T., and McMichael, A. (2013). “Climate Change Threats to Caribbean.” Atmósfera, 26(2), 283–309. Population Health and Well-being: The Imperative of Protective Solu- Kartiki K. (2011). “Climate Change and Migration: A Case Study from Rural tions that Will Last.” Global Health Action., 6(20816). Bangladesh.” Gender & Development, 19(1), 23–38. Klinenberg, E. (2002). Heat Wave: A Social Autopsy of Disaster in Chicago. Kaser, G., Juen, I., Georges, C., Gómez, J., and Tamayo, W. (2003). “The University of Chicago Press: Chicago. Impact of Glaciers on the Runoff and the Reconstruction of Mass Bal- Klok, E.J., and Tank, A.M.G.K. (2009). “Short Communication Updated ance History from Hydrological Data in the Tropical Cordillera Blanca, and Extended European Dataset of Daily Climate Observations.” Perú.” Journal of Hydrology, 282(1-4), 130–44. International Journal of Climatology, 29, 1182–91. 255 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Kloos, J., Gebert, N., Rosenfeld, T., and Renaud, F. (2013). “Climate Chapter 23—Europe.” Climate Change 2014: Impacts, Adaptation, and Change, Water Conflicts, and Human Security: Regional Assessment Vulnerability. Contribution of Working Group II to the Fifth Assess- and Policy Guidelines for the Mediterranean , Middle East, and Sahel ment Report of the Intergovernmental Panel on Climate Change. Final Regions.” Climate Change, Hydro Conflicts, and Human Security Draft, IPCC AR5 WGII, Cambridge University Press: Cambridge, UK (CLICO) No. 10, UNU-EHS. and New York, NY. Knabb, R.D., Rhome, J.R., and Brown, D.P. (2005). “Tropical Cyclone Koven, C.D., Ringeval, B., Friedlingstein, P., Ciais, P., Cadule, P., Khvo- Report: Hurricane Katrina.” NOAA—National Hurricane Center. rostyanov, D., Krinner, G., and Tarnocai, C. (2011). “Permafrost Knapp, K.R., Kruk, M.C., Levinson, D.H., Diamond, H.J., and Neumann, Carbon-Climate Feedbacks Accelerate Global Warming.” Proceedings C.J. (2010). “The International Best Track Archive for Climate Stew- of the National Academy of Sciences of the United States of America, ardship (IBTrACS),” Bulletin of the American Meteorological Society, 108(36), 14769–74. 91(3), 363–76. Kriegel, D., Mayer, C., Hagg, W., Vorogushyn, S., Duethmann, D., Gafurov, Kniveton, D., Schmidt-Verkerk, K., Smith, C., and Black, R. (2008). “Cli- A. and Farinotti, D. (2013). “Changes in Glacierisation, Climate and mate Change and Migration: Improving Methodologies to Estimate Runoff in the Second Half of the 20th Century in the Naryn Basin, Flows.” IOM Migration Research Series, International Organization Central Asia.” Global and Planetary Change, 110, 51–61. for Migration. Kriegler, E., Riahi, K., Bauer, N., Schwanitz, J., Petermann, N., Bosetti, Knutson, T.R., McBride, J.L., Chan, J., Emanuel, K., Holland, G., Landsea, V., Marcucci, A., Otto, S., Paroussos, L., Rao, S., Arroyo-Curras, T., C., Held, I., Kossin, J.P., Srivastava, A.K., and Sugi, M. (2010). “Tropi- Ashina, S., Bollen, J., Eom, J., Hamdi-Cherif, M., Kitous, A., Mejean, A., cal Cyclones and Climate Change.” Nature Geoscience, 3(3), 157–63. Schaeffer, M., Wada, K., Capros, P., Vuuren, D. van, and Edenhofer, O. Knutson, T.R., Sirutis, J.J., Vecchi, G.A., Garner, S., Zhao, M., Kim, (2013). “The Difficult Road to Global Cooperation on Climate Change: H.S., Bender, M., Tuleya, R.E., Held, I.M., and Villarini, G. (2013). The AMPERE Study on Staged Accession Scenarios for Climate Policy.” “Dynamical Downscaling Projections of Twenty-First-Century Atlantic Technological Forecasting & Social Change, In Review. Hurricane Activity: CMIP3 and CMIP5 Model-Based Scenarios.” Journal Kriegler, E., Tavoni, M., Aboumahboub, T., Luderer, G., Calvin, K., DeMaere, of Climate, 26(17), 6591–6617. G., Krey, V., Riahi, K., Rosler, H., Schaeffer, M. and van Vuuren, D. Knutti, R., and Sedlá ek, J. (2012). “Robustness and Uncertainties in the (2014). “What does the 2C Target Imply for a Global Climate Agree- New CMIP5 Climate Model Projections.” Nature Climate Change, ment in 2020? The LIMITS Study on Durban Platform Scenarios.” (October), 1–5. Climate Change Economics, in press. Kocha, C., Lafore, J.P., Tulet, P., and Seity, Y. (2012). “High-resolution Kriegler, E., Weyant, J., Blanford, G., Krey, V., Clarke, L., Edmonds, J., Simulation of a Major West African Dust Storm: Comparison with Fawcett, A., Luderer, G., Riahi, K., Richels, R., Rose, S., Tavoni, M., and Observations and Investigation of Dust Impact.” Quarterly Journal van Vuuren, D. (2014). “The Role of Technology for Achieving Climate of the Royal Meteorological Society, 138(663), 455–70. Policy Objectives: Overview of the EMF 27 Study on Global Technology Koelle K. (2009). “The Impact of Climate on the Disease Dynamics of and Climate Policy Strategies.” Climatic Change, 123(3–4), 353–67. Cholera.” Clin Microbial Infect, 15(Supp. 1), 29–31. Kroeker, K.J., Kordas, R.L., Crim, R., Hendriks, I.E., Ramajo, L., Singh, Kolahi, A.A., Rastagarpour, A., Abadi, A., and Gachkar, L. (2010). “An G.S., Duarte, C.M., and Gattuso, J.P., (2013). “Impacts of Ocean Unexpectedly High iIncidence of Acute Childhood Diarrhea in Koot- Acidification on Marine Organisms: Quantifying Sensitivities and Abdollah, Ahwaz, Iran.” International Journal of Infectious Diseases, Interaction with Warming.” Global Change Biology. 14(7), e618–e621. Krol, M., Jaeger, A., Bronstert, A., and Güntner, A. (2006). “Integrated Kolmannskog, V.O. (2009). “Future Floods of Refugees.” Norwegian Modelling of Climate, Water, Soil, Agricultural and Socioeconomic Refugee Council: Oslo. Processes: A General Introduction of the Methodology and some Kolstad, E.W., and Johansson, K.A. (2011). “Uncertainties Associated With Exemplary Results from the Semi-arid Northeast of Brazil.” Journal Quantifying Climate Change Impacts on Human Health: A Case Study of Hydrology, 328(3-4), 417–31. for Diarrhea.” Environmental Health Perspectives, 119(3), 299–305. Krol, M.S., and Bronstert, A. (2007). “Regional Integrated Modelling of Köppen, W. (1936). “Das Geographische System der Klimate.” Handbuch Climate Change Impacts on Natural Resources and Resource Usage der Klimatologie, edited by W. Köppen and R. Geiger, Gebrüder in Semi-arid Northeast Brazil.” Environmental Modelling & Software, Borntraeger: Berlin. 22(2), 259–68. Kos, J., Mastilovi, J., Jani, E. and Sari, B. (2013). “Natural Occurrence Kronik, J., and Verner, D. (2010),.Indigenous Peoples and Climate Change of Aflatoxins in Maize Harvested in Serbia during 2009-2012.” Food in Latin America and the Caribbean. The World Bank: Washington, DC. Control, 34, 31–34. Krysanova, V., Dickens, C., Timmerman, J., Varela-Ortega, C., Schlüter, M., Kossin, J.P., Emanuel, K.A., and Vecchi, G.A. (2014), “The Poleward Roest, K., Huntjens, P., Jaspers, F., Buiteveld, H., Moreno, E., Pedraza Migration of the Location of Tropical Cyclone Maximum Intensity.” Carrera, J., Slámová, R., Martínková, M., Blanco, I., Esteve, P., Pringle, Nature, 509(7500), 349–52. K., Pahl-Wostl, C., and Kabat, P. (2010), “Cross-Comparison of Climate Kossin, J.P., Olander, T.L., and Knapp, K.R. (2013). “Trend Analysis with Change Adaptation Strategies Across Large River Basins in Europe, a New Global Record of Tropical Cyclone Intensity.” 26(24), 9960–76. Africa, and Asia.” Water Resources Management, 24(14), 4121–60. Koubi, V., Bernauer, T., Kalbhenn, A., and Spilker, G. (2012). “Climate Ksiksi, T.S., Youssef, T., and Abdelmawla, E. (2012). “Ecosystem and Ecog- Variability, Economic Growth, and Civil Conflict.” Journal of Peace raphy Sea-level Rise and Abu Dhabi Coastlines: An Initial Assessment Research, 49(1), 113–27. of the Impact on Land and Mangrove Areas,” Journal of Ecosystem & Kouzana, L., Ben, A., and Sfar, M. (2009). “Seawater Intrusion and Asso- Ecography, 2(4). ciated Processes : Case of the Korba Aquifer (Cap-Bon, Tunisia).” Kuhlbrodt, T., and Gregory, J. (2012). “Ocean heat Uptake and its Con- Comptes Rendus Geoscience, 341(1), 21–35. sequences for the Magnitude of Sea-level Rise and Climate Change.” Kovats, R.S., Bouma, M.J., Hajat, S., Worrall, E., and Haines, A. (2003). Geophysical Research Letters, 39(18). “El Niño and Health.” Lancet, 362(9394), 1481–89. Kumar, A. (2013). “Natural Hazards of the Arabian Peninsula: Their Causes Kovats, R.S., Edwards, S.J., Hajat, S., Armstrong, B.G., Ebi, K.L., and and Possible Remediation.” inEarth System Processes and Disaster Menne, B. (2004). “The Effect of Temperature on Food Poisoning: A Management, edited by R. Sinha and R. Ravindra, Springer-Verlag: Time-series Analysis of Salmonellosis in Ten European Countries.” Berlin, Heidelberg. Epidemiol. Infect., 132, 443–53. Kumar, K.K., Rajagopalan, B., Hoerling, M., Bates, G., and Cane, M. (2006). Kovats, S., Valentini, R., Bouwer, L.M., Georgopoulou, E., Jacob, D., Mar- “Unraveling the Mystery of Indian Monsoon Failure during El Niño.” tin, E., Rounsevell, M., and Soussana, J.F. (2014). “Regional Aspects. Science (New York, N.Y.), 314(5796), 115–19. 256 B I BLIOGRA P HY Kumar, N., and Quisumbing, A.R. (2011). “Gendered Impacts of the 2007–08 Leakey, A.D.B. (2009). “Rising Atmospheric Carbon Dioxide Concentration Food Price Crisis Evidence Using Panel Data from Rural Ethiopia.” and the Future of C4 Crops for Food and Fuel.” Proceedings of the Washington, DC. Royal Society B, 276(February), 2333–43. Kumetat, D. (2012). “Climate Change on the Arabian Peninsula—Regional Leichenko, R., and Silva, J.A. (2014). “Climate Change and Poverty: Vulner- Security, Sustainability Strategies, and Research Needs.” Climate ability, Impacts, and Alleviation Strategies.” Wiley Interdisciplinary Change, Human Security, and Violent Conflict. Challenges for Social Reviews: Climate Change, 5(4), 539–56. Stability, edited by J. Scheffran, M. Brozka, H. G. Brauch, M. Link, Lelieveld, J., Hadjinicolaou, P., Kostopoulou, E., Chenoweth, J., El Maayer, and J. Schilling, Springer: Berlin, Heidelberg. M., Giannakopoulos, C., Hannides, C., Lange, M.A., Tanarhte, M., Kuriakose, A.T., Heltberg, R., Wiseman, W., Costella, C., Cipryk, R., and Tyrlis, E., and Xoplaki, E. (2012). “Climate Change and Impacts in Cornelius, S. (2012). “Climate-Responsive Social Protection.” Back- the Eastern Mediterranean and the Middle East.” Climatic Change, ground Paper No. 1210 for the World Bank 2012–2022 Social Protection 114, 667–87. and Labor Strategy. Lelieveld, J., Hadjinicolaou, P., Kostopoulou, E., Giannakopoulos, C., Kurz, W.A., Dymond, C.C., Stinson, G., Rampley, G.J., Neilson, E.T., Pozzer, A., Tanarhte, M., and Tyrlis, E. (2013). “Model Projected Heat Carroll, A.L., Ebata, T., and Safranyik, L. (2008). “Mountain Pine Extremes and Air Pollution in the Eastern Mediterranean and Middle Beetle and Forest Carbon Feedback to Climate Change.” Nature, East in the Twenty-first Century.” Regional Environmental Change, 452(7190), 987–90. Springer–Verlag. Laczko, F., and Piguet, E. (2014). “Regional Perspectives on Migration, Lemes, P., Melo, A.S., and Loyola, R.D. (2014). “Climate Change Threatens the Environment, and Climate Change.” People on the Move in a Protected Areas of the Atlantic Forest.” Biodiversity and Conserva- Changing Climate. The Regional Impact of Environmental Change tion, 23(2), 357–68. on Migration. Global Migration Issues 2, edited by F. Laczko and E. Lenton, T.M., Held, H., Kriegler, E., Hall, J.W., Lucht, W., Rahmstorf, S., Piguet, Springer Netherlands: Dordrecht. and Schellnhuber, H.J. (2008). “Tipping Elements in the Earth’s Laderach, P., Lundy, M., Jarvis, A., Ramirez, J., Portilla, E.P., Schepp, K., Climate System.” Proceedings of the National Academy of Sciences, and Eitzinger A. (2011). “Predicted Impact of Climate Change on Cof- 105(6), 1786–93. fee Supply Chains.” in The Economic, Social and Political Elements of Leone, M., D’Ippoliti, D., De Sario, M., Analitis, A., Menne, B., Katsuoy- Climate Change. Chapter 42. Edited by W. Leal Filho. Springer Berlin anni, K., De’Donato, F.K., Basagana, X., Ben Salah, A., Casimiro, E., Heidelberg: Berlin, Heidelberg. pp 703-723 Dortbudak, Z., Iniguez, C., Peretz, C., Wolf, T., and Michelozzi, P. Lagi, M., Bertrand, K.Z., and Bar-yam, Y. (2011). “The Food Crises and (2013). “A Time Series Study on the Effects of Heat on Mortality and Political Instability in North Africa and the Middle East.” New England Evaluation of Heterogeneity into European and Eastern-Southern Complex Systems Institute, Cambridge, MA. Mediterranean Cities: Results of EU CIRCE Project.” Environmental Lambeck, K., and Purcell, A. (2005). “Sea-level Change in the Mediter- Health, 12, 55. ranean Sea since the LGM: Model Predictions for Tectonically Stable Levermann, A. (2014). “Make Supply Chains Climate-smart.” Nature, Areas.” Quaternary Science Reviews, 24(18-19), 1969–88. 506, 27–29. Lambeck, K., Rouby, H., Purcell, A., Sun, Y., and Sambridge, M. (2014). Levermann, A., Clark, P.U., Marzeion, B., Milne, G.A., Pollard, D., “Sea level and global ice volumes from the Last Glacial Maximum to Radic, V., and Robinson, A. (2013). “The Multimillennial Sea-level the Holocene.” Proceedings of the National Academy of Sciences of Commitment of Global Warming.” Proceedings of the National the United States of America. doi: 10.1073/pnas.1411762111.Landerer, Academy of Sciences of the United States of America, 110(34), F.W., and Volkov D.L. (2013). “The Anatomy of Recent Large Sea 13745–50. Level Fluctuations in the Mediterranean Sea.” Geophysical Research Levermann, A., Winkelmann, R., Nowicki, S., Fastook, J.L., Frieler, K., Letters, 40(3), 553–57. Greve, R., Hellmer, H.H., Martin, M.A., Mengel, M., Payne, A.J., Pol- Langerwisch, F., Rost, S., Gerten, D., Poulter, B., Rammig, A., and Cramer, lard, D., Sato, T., Timmermann, R., Wang, W.L., and Bindschadler, W. (2013). “Potential Effects of Climate Change on Inundation Pat- R.A. (2014). “Projecting Antarctic Ice Discharge Using Response terns in the Amazon Basin.” Hydrology and Earth System Sciences, Functions from SeaRISE Ice-sheet Models.” Earth System Dynamics, 17(6), 2247–62. 5 (2014), 271–93. Lapenis, A., Shvidenko, A., Shepaschenko, D., Nilsson, S., and Aiyyer, A. Lewis, S.L., Brando, P.M., Phillips, O.L., van der Heijden, G.M.F., and (2005). “Acclimation of Russian Forests to Recent Changes in Climate.” Nepstad, D. (2011). “The 2010 Amazon Drought.” Science (New York, Global Change Biology, 11(12), 2090–2102. N.Y.), 331(6017), 554. Lapola, D.M., Schaldach, R., Alcamo, J., Bondeau, A., Msangi, S., Priess, Lhomme, J.P., Mougou, R. and Mansour, M. (2009). “Potential Impact J.A., Silvestrini, R., and Soares-Filho, B.S. (2011). “Impacts of Climate of Climate Change on Durum Wheat Cropping in Tunisia.” Climatic Change and the End of Deforestation on Land Use in the Brazilian Change, 96(4), 549–64. Legal Amazon.” Earth Interactions, 15(16), 1–29. Li, J., Lin, X., Chen, A., Peterson, T., Ma, K., Bertzky, M., Ciais, P., Kapos, Last, P.R., White, W.T., Gledhill, D.C., Hobday, A.J., Brown, R., V., Peng, C. and Poulter, B. (2013). “Global Priority Conservation Areas Edgar, G.J., and Pecl, G. (2011). “Long-term Shifts in Abundance in the Face of 21st Century Climate Change.” PloS one, 8(1), e54839. and Distribution of a Temperate Fish Fauna: A Response to Climate Li, J., Xie, S.P., Cook, E.R., Morales, M.S., Christie, D.A., Johnson, N.C., Change and Fishing Practices.” Global Ecology and Biogeography, Chen, F., D’Arrigo, R., Fowler, A.M., Gou, X., and Fang, K. (2013). 20(1), 58–72. “El Niño Modulations Over the Past Seven Centuries.” Nature Climate Latif, M., and Keenlyside, N.S. (2009). “El Nino/Southern Oscillation Change, 3(9), 822–26. Response to Global Warming.” Proceedings of the National Academy Li, W., Fu, R., Juárez, R.I.N., and Fernandes, K. (2008). “Observed of Sciences of the United States of America, 106(49), 20578–83. Change of the Standardized Precipitation Index, its Potential Cause, Lawler, J.J., Shafer, S.L., White, D., Kareiva, P., Maurer, E.P., Blaustein, and Implications to Future Climate Change in the Amazon Region.” A.R., and Bartlein, P.J. (2009). “Projected Climate-Induced Faunal Philosophical Transactions of the Royal Society of London. Series B, Change in the Western Hemisphere.” Ecology, 90(3), 588–97. Biological Sciences, 363(1498), 1767–72. Lawrence, G.B., Lapenis, A.G., Berggren, D., Aparin, B.F., Smith, K.T., van Lieshout, M., Kovats, R.S., Livermore, M.T.J., and Martens, P. (2004). Shortle, W.C., Bailey, S.W., Varlyguin, D.L., and Babikov, B. (2005), “Climate Change and Malaria: Analysis of the SRES Climate and Socio- “Climate Dependency of Tree Growth Suppressed by Acid Deposi- economic Scenarios.” Global Environmental Change, 14(1), 87–99. tion Effects on Soils in Northwest Russia.” Environmental Science & Lindquist, L., and Vapalahti, O. (2008). “Tick-borne Encephalitis.” Lan- Technology, 39(7), 2004–10. cent, 371, 1861–71. 257 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Lioubimtseva, E., and Henebry, G.M. (2009). “Climate and Environmental De Lucena, A.F.P., Szklo, A.S., Schaeffer, R., de Souza, R.R., Borba, Change in Arid Central Asia: Impacts, Vulnerability, and Adaptations.” B.S.M.C., da Costa, I.V.L., Júnior, A.O.P., and da Cunha, S.H.F. (2009). Journal of Arid Environments, 73, 963–77. “The Vulnerability of Renewable Energy to Climate Change in Brazil.” Lioubimtseva, E., and Henebry, G.M. (2012). “Grain Production Trends in Energy Policy, 37(3), 879–89. Russia, Ukraine, and Kazakhstan: New Opportunities in an Increasingly Lucht, W., Schaphoff, S., Erbrecht, T., Heyder, U., and Cramer, W. (2006). Unstable World?” Frontiers of Earth Science, 6(2), 157–66. “Terrestrial Vegetation Redistribution and Carbon Balance Under Little, C.M., Oppenheimer, M., and Urban, N.M. (2013). “Upper Bounds on Climate Change.” Carbon Balance and Management, 1, 6. Twenty-first-century Antarctic Ice Loss Assessed Using a Probabilistic Luck, J., Spackman, M., Freeman, A., Trebicki, P., Griffiths, W., Finlay, Framework.” Nature Climate Change, 3(3), 1–6. K., and Chakraborty, S. (2011). “Climate Change and Diseases of Food Liu H.H. (2011). “Impact of Climate Change on Groundwater Recharge Crops.” Plant Pathology, 60(1), 113–21. in Dry Areas: An Ecohydrology Approach.” Journal of Hydrology, Luderer, G., Bertram, C., Calvin, K., De Cian, E., and Kriegler, E. (2013). 407(1-4), 175–83. “Implications of Weak Near-term Climate Policies on Long-term Liu, T., Kinouchi, T., and Ledezma, F. (2013). “Characterization of Recent Mitigation Pathways.” Climatic Change. Glacier Decline in the Cordillera Real by LANDSAT, ALOS, and ASTER Ludescher, J., Gozolchiani, A., Bogachev, M.I., Bunde, A., Havlin, S., and Data.” Remote Sensing of Environment, 137, 158–72. Schellnhuber, H.J. (2013). “Improved El Nino Forecasting by Coopera- Lloyd, A.H., and Bunn, A.G. (2007). “Responses of the Circumpolar Boreal tivity Detection.” Proceedings of the National Academy of Sciences Forest to 20th Century Climate Variability.” Environmental Research of the United States of America, 110(47), 19172–73. Letters, 2(4), 045013. Luecke, M. (2011). “Climate Migration: A Strategy to Adapt to Climate Lloyd, A.H., Bunn, A.G., and Berner, L. (2011). “A Latitudinal Gradient Change or a Failure to Adapt?” Climate Change: Current Issues, edited in Tree Growth Response to Climate Warming in the Siberian Taiga.” by N. Trofimenko, Kiel Institute for the World Economy (IfW): Kiel. Global Change Biology, 17(5), 1935–45. Lumpkin, R., and Garzoli, S. (2011). “Interannual to Decadal Changes in the Lloyd, S.J., Kovats, R.S., and Chalabi, Z. (2011). “Climate Change, Crop Western South Atlantic’s Surface Circulation.” Journal of Geophysical Yields, and Undernutrition: Development of a Model to Quantify the Research, 116(C1), C01014. Impact of Climate Scenarios on Child Undernutrition.” Environmental Luo, Q. (2011). “Temperature Thresholds and Crop Production: A Review.” Health Perspectives, 119(12), 1817–23. Climatic Change, (December 2010), 583–98. Lobell, D.B., Burke, M.B., Tebaldi, C., Mastrandrea, M.D., Falcon, W.P., and Lutz, A.F., Immerzeel, W.W., Gobiet, A., Pellicciotti, F., and Bierkens, M.F.P. Naylor, R.L. (2008). “Prioritizing Climate Change Adaptation Needs for (2013a). “Comparison of Climate Change Signals in CMIP3 and CMIP5 Food Security in 2030.” Science (New York, N.Y.), 319(5863), 607–10. Multi-model Ensembles and Implications for Central Asian Glaciers.” Lobell, D.B., and Gourdji, S.M. (2012). “The Influence of Climate Change Hydrology and Earth System Sciences, 17(9), 3661–77. on Global Crop Productivity.” Plant Physiology, 160(4), 1686–97. Lutz, D.A., Shugart, H.H., and White, M.A. (2013b). “Sensitivity of Russian Lobell, D.B., Schlenker, W., and Costa-Roberts, J. (2011). “Climate Trends Forest Timber Harvest and Carbon Storage to Temperature Increase.” and Global Crop Production Since 1980.” Science (New York, N.Y.), Forestry, 86(2), 283–93. 333(6042), 616–20. Lutz, W. (2010). “Emerging Population Issues in Eastern Europe and Central Lombard, A., Garric, G., and Penduff, T. (2009). “Regional Patterns of Asia. Research Gaps on Demographic Trends, Human Capital, and Observed Sea-level Change: Insights from a 1/4° Global Ocean/sea- Climate Change.” Eastern Europe and Central Asia Regional Office ice Hindcast.” Ocean Dynamics, 59(3), 433–49. of the UNFPA: New York. Long, S.P., Ainsworth, E.A., Leakey, A.D.B., Nösberger, J., and Ort, D.R. Maas, A., Briggs, C., Cheterian, V., Fritzsche, K., Lee, B., Paskal, C., Taenzler, (2006). “Food for Thought: Lower-than-expected Crop Yield Stimu- D., and Carius, A. (2010). “Shifting Bases, Shifting Perils: A Scoping lation with Rising CO2 Concentrations.” Science (New York, N.Y.), Study on Security Implications of Climate Change in the OSCE Region 312(5782), 1918–21. and Beyond.” Adelphi Research on behalf of OSCE: Berlin. De Longueville, F., Hountondji, Y.C., Ozer, P., Marticorena, B., Chatenet, B., Maas, A., and Fritzsche, K. (2012). “Altering Security Dynamics? Climate and Henry, S. (2013). “Saharan Dust Impacts on Air Quality: What Change Impacts on Iraq.” Climate Change, Human Security, and Are the Potential Health Risks in West Africa?” Human and Ecological Violent Conflict. Challenges for Social Stability, edited by J. Schef- Risk Assessment: An International Journal, 19(6), 1595–1617. fran, M. Brzoska, H. G. Brauch, P. M. Link, and J. Schilling, Springer Lopez, P., Chevallier, P., Favier, V., Pouyaud, B., Ordenes, F., and Oer- Berlin Heidelberg, 409-422. lemans, J., (2010). “A Regional View of Fluctuations in Glacier Length Mabrouk, M.B., Jonoski, A., Solomatine, D., and Uhlenbrook, S. (2013). in Southern South America.” Global and Planetary Change, 71(1–2), “A Review of Seawater Intrusion in the Nile Delta Groundwater Sys- 85–108. tem—The Basis for Assessing Impacts due to Climate Changes and Lorenz, R., Jaeger, E.B., and Seneviratne, S.I. (2010). “Persistence of Heat Water Resources Development.” Hydrology and Earth System Sciences Waves and its Link to Soil Moisture Memory,” Geophysical Research Discussions, 10(8), 10873–911. Letters, 37(9). Macchi, M., Oviedo, G., Gotheil, S., Cross, K., Boedhihartono, A., Wolfan- Lotfy, W.M. (2013). “Climate Change and Epidemiology of Human Para- gel, C., and Howell, M. (2008). “Indigenous and Traditional Peoples sitosis in Egypt: Review.” Journal of Advanced Research. and Climate Change. Issues Paper.” IUCN. Lowry, S.J., Blecker, H., Camp, J., De Castro, B., Hecker, S., Arbabi, S., MacDonald, G.M., Kremenetski, K.V., and Beilman, D.W. (2008). “Climate and Traven, N. (2010). “Possibilities and Challenges in Occupational Change and the Northern Russian Treeline Zone.” Philosophical Trans- Injury Surveillance of Day Laborers.” American Journal of Industrial actions of the Royal Society of London. Series B, Biological Sciences, Medicine, 53(2), 126–34. 363(1501), 2285–99. Loyola, R.D., Lemes, P., Faleiro, F. V, Trindade-Filho, J., and Machado, Maclean, I.M.D., and Wilson, R.J. (2011). “Recent Ecological Responses R.B. (2012). “Severe Loss of Suitable Climatic Conditions for Marsupial to Climate Change Support Predictions of High Extinction Risk.” Species in Brazil: Challenges and Opportunities for Conservation.” Proceedings of the National Academy of Sciences of the United States PloS one, 7(9), e46257. of America, 108(30), 12337–42. Loyola, R.D., Lemes, P., Nabout, J.C., Trindade-Filho, J., Sagnori, M.D., Magnani, F., Mencuccini, M., Borghetti, M., Berbigier, P., Berninger, F., Dobrovolski, R., and Diniz-Filho, J.A.F. (2013). “A Straightforward Delzon, S., Grelle, A., Hari, P., Jarvis, P.G., Kolari, P., Kowalski, Conceptual Approach for Evaluating Spatial Conservation Priorities A.S., Lankreijer, H., Law, B.E., Lindroth, A., Loustau, D., Manca, G., Under Climate Change.” Biodiversity and Conservation, 22(2), 483–95. Moncrieff, J.B., Rayment, M., Tedeschi, V., Valentini, R., and Grace, 258 B I BLIOGRA P HY J. (2007). “The Human Footprint in the Carbon Cycle of Temperate Marcus, R., Pereznieto, P., Cullen, E., and Jones, N. (2011). “Children and and Boreal Forests.” Nature, 447(7146), 848–50. Social Protection in the Middle East and North Africa. A Mapping Magrin, G., Gay García, C., Cruz Choque, D., Giménez, J.C., Moreno, A.R., Exercise.” London. Nagy, G.J., Nobre, C., and Villamizar, A. (2007). “Chapter 10—Latin Marengo, J., Chou, S., Kay, G., Alves, L., Pesquero, J., Soares, W., Santos, America.” Climate Change 2007: Impacts, Adaptation, and Vulnerability. D., Lyra, A., Sueiro, G., Betts, R., Chagas, D., Gomes, J., Bustamante, Contribution of Working Group II to the Fourth Assessment Report of J., and Tavares, P. (2012). “Development of Regional Future Climate the Intergovermental Panel on Climate Change, edited by M. L. Parry, Change Scenarios in South America using the Eta CPTEC/HadCM3 O. F. Canziano, J. P. Palutikof, van der L. P.J., and C. E. Hanson, IPCC Climate Change Projections: Climatology and Regional Analyses for AR4 WGII, Cambridge University Press: Cambridge, 581–685. the Amazon, São Francisco, and the Paraná River Basins.” Climate Magrin, G., Marengo, J., Boulanger, J.P., Buckeridge, M.S., Castellanos, Dynamics, 38(9-10), 1829–48. E., Poveda, G., Scarano, F., and Vicuña, S. (2014). “Central and South Marengo, J.A., Alves, L.M., Soares, W.R., Rodriguez, D.A., Camargo, H., America.” Climate Change 2014: Impacts, Adaptation, and Vulner- Riveros, M.P., and Pabló, A.D. (2013). “Two Contrasting Severe Seasonal ability. Contribution of Working Group II to the Fifth Assessment Extremes in Tropical South America in 2012: Flood in Amazonia and Report of the Intergovernmental Panel on Climate Change. Regional Drought in Northeast Brazil.” Journal of Climate, 26(22), 9137–54. Aspects. Chapter 27. Final Draft, edited by L. O. Girardin and J. P. Marengo, J.A., Jones, R., Alves, L.M., and Valverde, M.C. (2009). “Future Ometto, IPCC AR5 WGII, Cambridge University Press: Cambridge UK. Change of Temperature and Precipitation Extremes in South America Mahmood, R., Pielke, R.A., Hubbard, K.G., Niyogi, D., Dirmeyer, P.A., as derived from the PRECIS Regional Climate Modeling System.” McAlpine, C., Carleton, A.M., Hale, R., Gameda, S., Beltrán-Przekurat, International Journal of Climatology. A., Baker, B., McNider, R., Legates, D.R., Shepherd, M., Du, J., Blanken, Marengo, J.A., Tomasella, J., Alves, L.M., Soares, W.R., and Rodriguez, P.D., Frauenfeld, O.W., Nair, U.S., and Fall, S. (2013). “Land Cover D.A. (2011). “The Drought of 2010 in the Context of Historical Droughts Changes and their Biogeophysical Effects on Climate.” International in the Amazon Region.” Geophysical Research Letters, 38(12) . Journal of Climatology, 34(4), 929–53. Marin, F.R., Jones, J.W., Singels, A., Royce, F., Assad, E.D., Pellegrino, Main Administation of Hydrometeorology (2009). “Second National Com- G.Q., and Justino, F. (2012). “Climate Change Impacts on Sugarcane munication of the Republic of Uzbekistan under the United Nations Attainable Yield in Southern Brazil.” Climatic Change, 117(1-2), 227–39. Framework Convention on Climate Change.” Centre of Hydrometeo- Marini, M.A., Barbet-Massin, M., Lopes, L.E., and Jiguet, F. (2009). “Pre- rological Service under the Cabinet of Ministers of the Republic of dicted Climate-driven Bird Distribution Changes and Forecasted Con- Uzbekistan: Tashkent. servation Conflicts in a Neotropical Savanna.” Conservation Biology: Malevsky-Malevich, S.P., Molkentin, E.K., Nadyozhina, E.D., and Shk- The Journal of the Society for Conservation Biology, 23(6), 1558–67. lyarevich, O.B. (2008). “An Assessment of Potential Change in Wildfire Mark, B.G., McKenzie, J.M., and Gómez, J. (2005). “Hydrochemical Activity in the Russian Boreal Forest Zone Induced By Climate Warming Evaluation of Changing Glacier Meltwater Contribution to Stream During the Twenty-first Century.” Climatic Change, 86(3-4), 463–74. Discharge: Callejon de Huaylas, Peru/Evaluation Hydrochimique Malhi, Y., Aragão, L.E.O.C., Galbraith, D., Huntingford, C., Fisher, R., de la Contribution Evolutive de la Fonte Glaciaire à L’écoulement Zelazowski, P., Sitch, S., McSweeney, C., and Meir, P. (2009). “Exploring Fluvial: Callejon de Huaylas, Pérou.” Hydrological Sciences Journal, the Likelihood and Mechanism of a Climate-change-induced Dieback 50(6), 37–41. of the Amazon Rainforest.” Proceedings of the National Academy of Markandya, A., and Chiabai, A. (2009). “Valuing Climate Change Impacts Sciences of the United States of America, 106(49), 20610–15. on Human Health: Empirical Evidence from the Literature.” Int. J. Malhi, Y., Wood, D., Baker, T.R., Wright, J., Phillips, O.L., Cochrane, T., Environ. Res. Public Health, 6, 759–86. Meir, P., Chave, J., Almeida, S., Arroyo, L., Higuchi, N., Killeen, T.J., Marshall, N.A., Marshall, P.A., Abdulla, A., Rouphael, T., and Ali, A. (2011). Laurance, S.G., Laurance, W.F., Lewis, S.L., Monteagudo, A., Neill, D. “Preparing for Climate Change: Recognising its Early Impacts through A., Vargas, P.N., Pitman, N.C.A., Quesada, C.A., Salomao, R., Silva, the Perceptions of Dive Tourists and Dive Operators in the Egyptian J.N.M., Lezama, A.T., Terborgh, J., Martinez, R.V., and Vinceti, B. Red Sea.” Current Issues in Tourism, 14(6), 507–18. (2006). “The Regional Variation of Aboveground Live Biomass in Old- Martí, R.Z. (2006). “Los Efectos de los Desastres en 2004 y 2005: La growth Amazonian Forests.” Global Change Biology, 12(7), 1107–38. Necesidad de Adaptación de Largo Plazo.” SERIE Estudios y Per- Mallard, M.S., Lackmann, G.M., Aiyyer, A., and Hill, K. (2013). “Atlantic spectivas—SEDE Subregional de la CEPAL en Mexico no. 54, CEPAL: Hurricanes and Climate Change. Part I: Experimental Design and Isola- Mexico City. tion of Thermodynamic Effects.” Journal of Climate, 26(13), 4876–93. Martinez-Sanchez, J., and Cavazos, T. (2014). “Eastern Tropical Pacific Mangal, T.D., Paterson, S., and Fenton, A. (2008). “Predicting the Impact Hurricane Variability and Landfalls on Mexican Coasts.” Climate of Long-Term Temperature Changes on the Epidemiology and Control Research, 58(3), 221–34. of Schistosomiasis: A Mechanistic Model.” PLoS ONE, 3(1), e1438. Martinez-Urtaza, J., Huapaya, B., Gavilan, R.G., Blanco-Abad, V., Ansede- Mani, M., and Pillai, R. (2010). “Impact of Dust on Solar Photovoltaic (PV) Bermejo, J., Cadarso-Suarez, C., Figueiras, A., and Trinanes, J. (2008). Performance: Research Status, Challenges, and Recommendations.” “Emergence of Asiatic Vibrio Diseases in South America in Phase With Renewable and Sustainable Energy Reviews, 14(9), 3124–31. El Nino.” Epidemiology, 19(6). Mannig, B., Müller, M., Starke, E., Merkenschlager, C., Mao, W., Zhi, X., Marzeion, B., Jarosch, A.H., and Hofer, M. (2012). “Past and Future Podzun, R., Jacob, D., and Paeth, H. (2013). “Dynamical Downscaling Sea-level Change from the Surface Mass Balance of Glaciers.” The of Climate Change in Central Asia.” Global and Planetary Change, Cryosphere, 6, 1295–1322. 110, 26–39. Mascarenhas, A., and Wisner, B. (2012). “Politics, Power, and Disasters.” Mantilla, G., Oliveros, H., and Barnston, A.G. (2009). “The Role of ENSO The Routledge Handbook of Hazards and Disaster Risk Reduction, in Understanding Changes in Colombia’s Annual Malaria Burden by edited by B. Wisner, J. Gaillard, and I. Kelman, Routledge: New York. Region, 1960-2006.” Malaria Journal, 8(6). Mas-Coma, S. (2005). “Epidemiology of fascioliasis in Human Endemic Marcos, M., and Tsimplis, M.N. (2008). “Comparison of Results of AOGCMs Areas.” Journal of Helminthology, 79, 207–16. in the Mediterranean Sea during the 21st Century.” Journal of Geo- Mas-Coma, S., Agramunt, V.H., and Valero, M.A. (2014). “Neurological and physical Research, 113(C12), C12028. Ocular Fascioliasis in Humans.” Advances in Parasitology, 84, 27–149. Marcos, M., Tsimplis, M.N., and Shaw, A.G.P. (2009). “Sea-level Mas-Coma, S., Valero, M.A., and Bargues, M.D. (2009). “Climate Change Extremes in Southern Europe.” Journal of Geophysical Research, Effects on Trematodiases, with emphasis on Zoonotic fascioliasis and 114(C1), C01007. schistosomiasis.” Veterinary Parasitology, 163, 264–80. 259 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Masiokas, M.H., Villalba, R., Christie, D.A., Betman, E., Luckman, B.H., McLeman, R. (2011). “Climate Change, Migration, and Critical International Le Quesne, C., Prieto, M.R., and Mauget, S. (2012), “Snowpack Varia- Security Considerations.” IOM Migration Research Series, International tions since AD 1150 in the Andes of Chile and Argentina (30°–37°S) Organization for Migration, Geneva. Inferred from Rainfall, Tree-ring, and Documentary Records.” Journal McLeman, R., and Hunter, L. (2011). “Migration in the Context of Vulner- of Geophysical Research, 117(D5), D05112. ability and Adaptation to Climate Change: Insights from Analogues.” Masiokas, M.H., Villalba, R., Luckman, B.H., and Mauget, S. (2010). “Intra- Wiley Interdisciplinary Reviews: Climate Change, 1(3), 450–61. to Multidecadal Variations of Snowpack and Streamflow Records in McMichael, A.J., and Lindgren, E. (2011). “Climate Change: Present and the Andes of Chile and Argentina between 30° and 37°S.” Journal of Future Risks to Health, and Necessary Responses.” Journal of Internal Hydrometeorology, 11(3), 822–31. Medecine, (270), 401–13. Masiokas, M.H., Villalba, R., Luckman, B.H., Montaña, E., Betman, E., McMichael, C., Barnett, J., and McMichael, A.J. (2012). “An Ill Wind? Christie, C., Le Quesne, C., and Mauget, S. (2013). “Recent and His- Climate Change, Migration, and Health.” Environmental Health toric Andean Snowpack and Streamflow Variations and Vulnerability Perspectives, 120(5). to Water Shortages in Central-Western Argentina.” Climate Vulner- McPhaden, M.J., Zebiak, S.E., and Glantz, M.H. (2006). “ENSO as an ability: Understanding and Addressing Threats to Essential Resources, Integrating Concept in Earth Science.” Science (New York, NY), Elsevier Inc., Academic Press. 314(5806), 1740–45. Masiokas, M.H., Villalba, R., Luckman, B.H., Le Quesne, C., and Aravena, MEA (2005). Millenium Ecosystem Assessment. Ecosystems and Human J.C. (2006). “Snowpack Variations in the Central Andes of Argentina and Wellbeing: Synthesis. Edited by J. Sarukhan and A. Whyte, Island Chile, 1951-2005: Large-scale Atmospheric Influences and Implications Press: Washington, DC. for Water Resources in the Region.” Journal of Climate, 19, 6334–52. Medvigy, D., Walko, R.L., and Avissar, R. (2011). “Effects of Deforestation Maslac, T. (2012). “Drought Driven Declines in Serbian Crops Increased on Spatiotemporal Distributions of Precipitation in South America.” Food Prices. GAIN Report.” Journal of Climate, 24(8), 2147–63. Masson-Delmotte, V., Schulz, M., Abe-Ouchi, A., Beer, J., Ganopolski, A., Meehl, G.A., Stocker, T.F., Collins, W.D., Friedlingstein, P., Gaye, A.T., Rouco, J.F.G., Jansen, E., Lambeck, K., Luterbacher, J., Naish, T., Gregory, J.M., Kitoh, A., Knutti, R., Murphy, J.M., Noda, A., Raper, Osborn, T., Otto-Bliesner, B., Quinn, T., Ramesh, R., Rojas, M., S.C.B., Watterson, I.G., Weaver, A.J., and Zhao, Z.C. (2007). “Global Shao, X., and Timmermann, A. (2013). “Information from Paleocli- Climate Projections.” Climate Change 2007: The Physical Science mate Archives.” Climate Change 2013: The Physical Science Basis. Basis. Contribution of Working Group I to the Fourth Assessment Contribution of Working Group I to the Fifth Assessment Report of the Report of the Intergovernmental Panel on Climate Change, edited by Intergovernmental Panel on Climate Change, edited by T. F. Stocker, S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Averyt, D. Qin, G.-K. Plattner, M. Tignor, S. K. Allen, J. Boschung, A. Nauels, M. Tignor, and H. L. Miller, IPCC AR4 WGI, Cambridge University Y. Xia, V. Bex, and P. M. Midgley, IPCC AR5 WGI, Cambridge University Press: Cambridge, UK and New York, NY. Press: Cambridge, UK and New York, NY. Meissner, K.J., Lippmann, T., and Sen Gupta, A. (2012). “Large-scale Stress Mata, L.J., Campos, M., Basso, E., Compagnucci, R., Fearnside, P., Factors Affecting Coral Reefs: Open Ocean Sea Surface Temperature Magrin, G., Marengo, J., Moreno, A.R., Suárez, A., Solman, S., Vil- and Surface Seawater Aragonite Saturation over the Next 400 Years.” lamizar, A., and Villers, L. (2001). “Latin America.” Working Group II: Coral Reefs, 31(2), 309–19. Impacts, Adaptation, and Vulnerability, Intergovernmental Panel of Mekonnen, M.M., and Hoekstra, A.Y. (2011). “National Water Footprint Climate Changeanel of Climate Change, edited by J. J. McCarthy, O. F. Accounts: The Green, Blue and Grey Water Footprint of Production Canziani, N. A. Leary, D. J. Dokken, and K. S. White, Cambridge and Consumption. Volume 1: Main Report.” Value of Water Research University Press. Report Series No. 50. UNESCO-IHE Institute for Water Education: Mata, L.J., and Nobre, C. (2006). “Impacts, Vulnerability, and Adaptation to Delft, Netherlands Climate Change in Latin America.” Background Paper, UNFCCC: Lima. Mendelsohn, R. (2007). “Past Climate Change Impacts on Agriculture. Maurer, E.P., Adam, J.C., and Wood, A.W. (2009). “Climate Model-based In Agricultural Development: Farmers, Farm Production and Farm Consensus on the Hydrologic Impacts of Climate Change to the Rio Markets.” Handbook of Agricultural Economics, 3, 3009–31. Lempa basin of Central America.” Hydrology and Earth System Sci- Mendelsohn, R., Emanuel, K., Chonabayashi, S., and Bakkensem L. (2011). ences, (13), 183–94. “The Impact of Climate Change on Global Tropical Storm Damages.” Maynard-Ford, B.M.C., Phillips, E.C., Chirico, P.G., and Survey, U.S.G. Nature Climate Change, Policy Research Working Paper, 2, 205–9. (2008). “Mapping Vulnerability to Disasters in Latin America and the Menéndez, Á., and Berbery, E.H. (2005). “Hydrological Trends.” Climate Caribbean, 1900–2007.” Reston, VA. Change in the La Plata Basin: Raising Awareness and New Tools for McDermott, G.R., and Nilsen, Ø.A. (2014). “Electricity Prices, River Tem- Water Management, edited by V. Barros, R. Clarke, and P. Silva Días, peratures, and Cooling Water Scarcity.” Land Economics, 90, 131–48. Project SGP II 057, Inter American Institute for Global Change Research McDowell, J.Z., and Hess, J.J. (2012). “Accessing Adaptation: Multiple (IAI): Montevideo, Uruguay. Stressors on Livelihoods in the Bolivian Highlands under a Changing Mengel, M., and Levermann, A. (2014). “Ice Plug Prevents Irreversible Climate.” Global Environmental Change, 22(2), 342–52. Discharge from East Antarctica.” May, 1–5. McDowell, M.A., Rafati, A., Ramalho-Ortigao, M., and Ben Salah, A. Mercer, J., Domineyhowes, D., Kelman, I., and Lloyd, K. (2007). “The (2011). “Leishmaniasis: Middle East and North Africa Research and Potential for Combining Indigenous and Western Knowledge in Reduc- Development Priorities.” PLoS Negl Trop Dis, 5(7), e1219. ing Vulnerability to Environmental Hazards in Small Island Developing McDowell, N.G. (2011). “Mechanisms Linking Drought, Hydraulics, States.” Environmental Hazards, 7(4), 245–56. Carbon Metabolism, and Vegetation Mortality.” Plant Physiology, Mercer, J.H. (1978). “West Antarctic Ice Sheet and CO2 Greenhouse Effect: 155(3), 1051–59. A Threat of Disaster.” Nature, 271(5643), 321–25. McEvoy, D., Matczak, P., Banaszak, I., and Chorynski, A. (2010). “Fram- Mergili, M., and Schneider, J.F. (2011). “Regional-scale Analysis of Lake Outburst ing Adaptation to Climate-Related Extreme Events.” Mitigation and Hazards in the Southwestern Pamir, Tajikistan, based on Remote Sensing Adaptation Strategies for Global Change, 15(7), 779-795. and GIS.” Natural Hazards and Earth System Science, 11(5), 1447–62. McGranahan, G., Balk, D., and Anderson, B. (2007). “The Rising Tide: Merrifield, M.A., and Maltrud, M.E. (2011). “Regional Sea-level Trends Assessing the Risks of Climate Change and Human Settlements in Low due to a Pacific Trade Wind Intensification.” Geophysical Research Elevation Coastal Zones.” Environment and Urbanization, 19(1), 17–37. Letters, 38(21), L21605. 260 B I BLIOGRA P HY Mesquita, P.C.M.D., Pinheiro-Mesquita, S.F., and Pietczak, C. (2013). “Are Montenegro, A., and Ragab, R. (2010). “Hydrological Response of a Brazil- Common Species Endangered by Climate Change? Habitat Suitability ian Semi-arid Catchment to Different Land Use and Climate Change Projections for the Royal Ground Snake, Liophis reginae (Serpentes, Scenarios: A Modelling Study.” Hydrological Processes, 24(19), 2705–23. Dipsadidae).” North-Western Journal of Zoology, 9(1), 51–56. Mora, C., Frazier, A.G., Longman, R.J., Dacks, R.S., Walton, M.M., Meyers, W.H., Ziolkowska, J.R., Tothova, M., and Goychuk, K. (2012). Tong, E.J., Sanchez, J.J., Kaiser, L.R., Stender, Y.O., Anderson, J.M., “Issues Affecting the Future of Agriculture and Food Security for Europe Ambrosino, C.M., Fernandez-Silva, I., Giuseffi, L.M., and Giambel- and Central Asia.” Policy Studies on Rural Transition. luca, T.W. (2013). “The Projected Timing of Climate Departure from Meyssignac, B., and Cazenave, A., (2012). “Sea Level: A Review of Recent Variability.” Nature, 502(7470), 183–87. Present-day and Recent-past Changes and Variability.” Journal of Mora, D.E., Campozano, L., Cisneros, F., Wyseure, G., and Willems, P. Geodynamics, 58, 96–109. (2013). “Climate Changes of Hydrometeorological and Hydrological Meza, F.J., and Silva, D. (2009). “Dynamic Adaptation of Maize and Wheat Extremes in the Paute Basin, Ecuadorean Andes.” Hydrology and Production to Climate Change.” Climatic Change, 94(1-2), 143–56. Earth System Sciences Discussions, 10(5), 6445–71. Mika, J. (2013). “Changes in Weather and Climate Extremes: Phenom- Moreno, A., and Amelung, B. (2009). “Climate Change and Coastal and enology and Empirical Approaches.” Climatic Change, 121(1), 15–26. Marine Tourism: Review and Analysis.” Journal of Coastal Research, Miller, S.D., Kuciauskas, A.P., Liu, M., Ji, Q., Reid, J.S., Breed, D.W., Walker, A.L., 2009(56), 1140–44. and Mandoos, A. Al (2008). “Haboob Dust Storms of the Southern Arabian Moreno, A.R. (2006). “Climate Change and Human Health in Latin Peninsula.” Journal of Geophysical Research, 113(D1), D01202. America: Drivers, Effects, and Policies.” Regional Environmental Milly, P.C.D., Dunne, K.A., and Vecchia, A.V. (2005). “Global Pattern of Change, 6, 157–64. Trends in Streamflow and Water Availability in a Changing Climate.” Moreno Sanchez, A.R., and Urbina Soria, J. (2008). “Impactos Sociales Nature, 438(7066), 347–50. del Cambio Climatico en Mexico.” INE-UNDP: Mexico, D.F. Mimura, N., Nurs, L., McLean, R., Agard, J., Briguglio, L., Lefale, P., Morice, C.P., Kennedy, J.J., Rayner, N.A., and Jones, P.D. (2012). “Quanti- Payet, R., and Sem, G. (2007). “Small Islands.” Climate Change 2007: fying Uncertainties in Global and Regional Temperature Change using Impacts, Adaptation, and Vulnerability, Contribution of Working Group an Ensemble of Observational Estimates: The HadCRUT4 Data Set.” II to the Fourth Assessment Report of the Intergovernmental Panel on Journal of Geophysical Research: Atmospheres, 117(D8), D08101. Climate Change, edited by M. L. Parry, O. F. Canziani, J. P. Palutikof, Moritz, C., and Agudo, R. (2013). “The Future of Species under Climate P. J. van der Linden, and C. E. Hanson, IPCC AR4 WGII, Cambridge Change: Resilience or Decline?” Science (New York, N.Y.), 341(6145), University Press: Cambridge. 504–8. Min, S.K., Zhang, X., Zwiers, F.W., and Hegerl, G.C. (2011). “Human Morlighem, M., Rignot, E., Mouginot, J., Seroussi, H., and Larour, E. (2014). Contribution to More-intense Precipitation Extremes.” Nature, “Deeply Incised Submarine Glacial Valleys beneath the Greenland Ice 470(7334), 378–81. Sheet.” 7(May), 18–22. Ministry for Spatial Planning Construction and Ecology (2013). “Second Morrison, A., Ellsberg, M., and Bott, S. (2004). Addressing Gender-based National Communication of Bosnia and Herzegovina under the United Violence in the Latin American and Caribbean Region: A Critical Nations Framework Convention on Climate Change (UNFCCC).” Review of Interventions. World Bank: Washington, DC. Ministry for Spatial Planning, Construction, and Ecology of Bosnia Morton, D.C., Nagol, J., Carabajal, C.C., Rosette, J., Palace, M., Cook, and Herzegovina: Banja Luka. B.D., Vermote, E.F., Harding, D.J., and North, P.R.J. (2014). “Amazon Ministry of Environment and Spatial Planning (2010). “Initial National Forests Maintain Consistent Canopy Structure and Greenness during Communication of the Republic of Serbia under the United Nations the Dry Season.” Nature, 506(7487), 221–24. Framework Convention on Climate Change.” The Ministry of Envi- Morton, D.C., Le Page, Y., Defries, R., Collatz, G.J., and Hurtt, G.C. (2013). ronment and Spatial Planning of the Republic of Serbia: Belgrade. “Understorey Fire Frequency and the Fate of Burned Forests in Southern Ministry of Nature Protection (2003). “Tajikistan National Action Plan for Amazonia.” Philos Trans R Soc Lond B Biol Sci, 368(1619), 20120163. Climate Change Mitigation.” Main Administration on Hydrometeorol- Moser, C., Norton, A., Stein, A., and Georgieva, S. (2010). “Pro-Poor ogy and Environmental Pollution Monitoring, Ministry for Nature Adaptation to Climate Change in Urban Centers: Case Studies of Protection of the Republic Tajikistan: Dushanbe. Vulnerability and Resilience in Kenya and Nicaragua.” World Bank: Miraglia, M., Marvin, H.J.P., Kleter, G.A., Battilani, P., Brera, C., Coni, E., Washington, DC. Cubadda, F., Croci, L., Santis, B. De, Dekkers, S., Filippi, L., Hutjes, Moser, C., and Rogers, D. (2005). “Change, Violence and Insecurity in R.W.A., Noordam, M.Y., Pisante, M., Piva, G., Prandini, A., Toti, L., Non-Conflict Situations.” ODI Working Papers, London. Born, G.J., Van, Den, and Vespermann, A. (2009). “Climate Change Mougou, R., Mansour, M., Iglesias, A., Chebbi, R.Z., and Battaglini, A. and Food Safety: An Emerging Issue with Special Focus on Europe.” (2010). “Climate Change and Agricultural Vulnerability: A Case Study Food and Chemical Toxicology, 47(5), 1009–21. of Rain-fed Wheat in Kairouan, Central Tunisia.” Regional Environ- Mitrovica, J.X., Tamisiea, M.E., Davis, J.L., and Milne, G.A. (2001). “Recent mental Change, 11(S1), 137–42. Mass Balance of Polar Ice Sheets inferred from Patterns of Global Moustadraf, J., Razack, M., and Sinan, M. (2008). “Evaluation of the Sea-level Change.” Nature, 409(6823), 1026–29. Impacts of Climate Changes on the Coastal Chaouia Qquifer, Morocco, Mizik, T. (2010). “The Diversity of Agriculture in Former Soviet and Western using Numerical Modeling.” Hydrogeology Journal, 16(7), 1411–26. Balkan Countries.” FAO Regional Office for Europe and Central Asia, Muggeo, V.M., and Hajat, S. (2009). “Modelling the Non-linear Multiple-lag Policy Studies on Rural Transition No. 2010-2. Effects of Ambient Temperature on Mortality in Santiago and Palermo: Mizyed, N. (2008). “Impacts of Climate Change on Water Resources A Constrained Segmented Distributed Lag Approach.” Occup Environ Availability and Agricultural Water Demand in the West Bank.” Water Med, 66, 584–91. Resources Management, 23(10), 2015–29. Mukheibir, P. (2013). “Potential Consequences of Projected Climate Mokhov, I.I., Chernokulsky, A. V., and Shkolnik, I.M. (2006). “Regional Change Impacts on Hydroelectricity Generation.” Climatic Change, Model Assessments of Fire Risks under Global Climate Changes.” 121(1), 67–78. Doklady Earth Sciences, 411(2), 1485–88. Mulitza, S., Heslop, D., Pittauerova, D., Fischer, H.W., Meyer, I., Stuut, Molina Barrios, R., Figueroa, M., and Quisbert, I. (2005). “Los Pueblos J.B., Zabel, M., Mollenhauer, G., Collins, J.A., Kuhnert, H., and Schulz, Indígenas de Bolivia: Diagnóstico Sociodemográfico a Partir del Censo M. (2010). “Increase in African Dust Flux at the Onset of Commercial del 2001.” Santiago de Chile. Agriculture in the Sahel Region.” Nature, 466(7303), 226–28. 261 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Müller, C., Elliott, J., and Levermann, A. (2014). “Food Security: Fertilizing Nepstad, D., Carvalho, G., Barros, A.C., Alencar, A., Capobianco, J.P., Hidden Hunger.” Nature Climate Change, 4(7), 540–41. Bishop, J., Moutinho, P., Lefebvre, P., Silva, U.L., and Prins, E. (2001). Murakami, H., Mizuta, R., and Shindo, E. (2011). “Future Changes in “Road paving, Fire Regime Feedbacks, and the Future of Amazon Tropical Cyclone Activity Projected by Multi-physics and Multi-SST Forests.” Forest Ecology and Management, 154(3), 395–407. Ensemble Experiments using the 60-km-mesh MRI-AGCM.” Climate Nepstad, D., Stickler, C.M., Filho, B.S., and Merry, F. (2008). “Interactions Dynamics, 39(9-10), 2569–84. among Amazon Land Use, Forests and Climate: Prospects for a Near- Murakami, H., and Wang, B. (2010). “Future Change of North Atlantic term Forest Tipping Point.” Philosophical Transactions of the Royal Tropical Cyclone Tracks: Projection by a 20-km-Mesh Global Atmo- Society of London. Series B, Biological Sciences, 363(1498), 1737–46. spheric Model.” Journal of Climate, 23(10), 2699–2721. Nepstad, D.C., de Carvalho, C.R., Davidson, E.A., Jipp, P.H., Lefebvre, P.A., Murakami, H., Wang, Y., Yoshimura, H., Mizuta, R., Sugi, M., Shindo, Negreiros, G.H., da Silva, E.D., Stone, T.A., Trumbore, S.E., and Vieira, E., Adachi, Y., Yukimoto, S., Hosaka, M., Kusunoki, S., Ose, T., and S. (1994). “The Role of Deep Roots in the Hydrological and Carbon Kitoh, A. (2012). “Future Changes in Tropical Cyclone Activity Pro- Cycles of Amazonian Forests and Pastures. Nature, 372(6507), 666–69. jected by the New High-Resolution MRI-AGCM.” Journal of Climate, Nepstad, D.C., Tohver, I.M., Ray, D., Moutinho, P., and Cardinot, G. (2007). 25(9), 3237–60. “Mortality of Large Trees and Lianas Following Experimental Drought Myers, N., Mittermeier, R.A., Mittermeier, C.G., da Fonseca, G.A., and in an Amazon Forest.” Ecology, 88(9), 2259–69. Kent, J. (2000). “Biodiversity Hotspots for Conservation Priorities.” Neumayer, E., and Plumper, T. (2007). “The Gendered Nature of Natural Nature, 403(6772), 853–58. Disasters: The Impact of Catastrophic Events on the Gender Gap in Myers, S.S., Zanobetti, A., Kloog, I., Huybers, P., Leakey, A.D.B., Bloom, Life Expectancy, 1981-2002.” Annals of the Association of American A.J., Carlisle, E., Dietterich, L.H., Fitzgerald, G., Hasegawa, T., Holbrook, Geographers, 97(3), 551–66. N.M., Nelson, R.L., Ottman, M.J., Raboy, V., Sakai, H., Sartor, K.A., NYT (2014). “A Coffee Crop Withers. Fungus Cripples Coffee Production Schwartz, J., Seneweera, S., Tausz, M., and Usui, Y. (2014). “Increas- Across Central America” New York Times (accessed September 10, ing CO2 Threatens Human Nutrition.” Nature, 510(7503), 139–42. 2014), [available at http://www.nytimes.com/2014/05/06/business/ Myneni, R.B., Dong, J., Tucker, C.J., Kaufmann, R.K., Kauppi, P.E., Liski, international/fungus-cripples-coffee-production-across-central-america. J., Zhou, L., Alexeyev, V., and Hughes, M.K. (2001). “A Large Carbon html?_r=2] Sink in the Woody Biomass of Northern Forests.” Proceedings of Niang, I., Dansokho, M., Faye, S., Gueye, K., and Ndiaye, P. (2010). “Impacts the National Academy of Sciences of the United States of America, of Climate Change on the Senegalese Coastal Zones: Examples of 98(26), 14784–89. the Cap Vert Peninsula and Saloum Estuary.” Global and Planetary Myneni, R.B., Keeling, C.D., Tucker, C.J., Asrar, G., and Nemani, R.R. Change, 72(4), 294–301. (1997). “Increased Plant Growth in the Northern High Latitudes from Niang, I., Ruppel, O.C., Abdrabo, M., Essel, A., Lennard, C., Padgham, 1981 to 1991.” Nature, 386(6626), 698–702. J., and Urquhart, P. (2014). “Africa.” Climate Change 2014: Impacts, Naderi Beni, A., Lahijani, H., Mousavi Harami, R., Arpe, K., Leroy, S.A.G., Adaptation, and Vulnerability. IPCC Working Group II Contribution Marriner, N., Berberian, M., Andrieu-Ponel, V., Djamali, M., Mahboubi, to AR5. Volume I: Global and Sectoral Aspects. Final Draft. A., and Reimer, P.J. (2013). “Caspian Sea-level Changes during the Nicholls, R.J., Marinova, N., Lowe, J.A., Brown, S., Vellinga, P., de Gusmão, Last Millennium: Historical and Geological Evidence from the South D., Hinkel, J., and Tol, R.S.J. (2011). “Sea-level Rise and its Possible Caspian Sea.” Climate of the Past, 9(4), 1645–65. Impacts Given a ‘Beyond 4°C World’ in the Twenty-first Century.” Nakaegawa, T., Kitoh, A., and Hosaka, M. (2013). “Discharge of Major Philosophical Transactions. Series A, Mathematical, Physical, and Global Rivers in the Late 21st Century Climate Projected with the Engineering Sciences, 369(1934), 161–81. High Horizontal Resolution MRI-AGCMs.” Hydrological Processes, Nicoldi, J.L., and Mueller Petermann, R. (2010). “Potential Vulnerability of 27, 3301–18. the Brazilian Coastal Zone in its Environmental, Social, and Technologi- Nakaegawa, T., and Vergara, W. (2010). “First Projection of Climatological cal Aspects,” Pan-Americal Journal of Aquatic Sciences, 5(2), 12–32. Mean River Discharges in the Magdalena River Basin, Colombia, in Nijeri, Njiru B. (2012). “Climate Change, Resource Competition, and Con- a Changing Climate during the 21st Century.” Hydrological Research flict amongst Pastoral Commumities in Kenya.” in Climate Change, Letters, 4, 50–54. Human Security and Violent Conflict. Challenges for Social Stability., Nateghi, R., Guikema, S., and Quiring, S.M. (2013). “Power Outage edited by J. Scheffran, M. Brzoska, H. G. Brauch, P. M. Link, and J. Estimation for Tropical Cyclones: Improved Accuracy with Simpler Schilling, Springer: Berlin, Heidelberg. Models.” Risk Analysis: An Official Publication of the Society for Nóbrega, M.T., Collischonn, W., Tucci, C.E.M., and Paz, A.R. (2011). Risk Analysis, 34(6). “Uncertainty in Climate Change Impacts on Water Resources in the Nayar, A. (2009). “When the Ice Melts.” Nature, 461(October), 1042–46. Rio Grande Basin, Brazil.” Hydrology and Earth System Sciences, Necco Carlomagno, G. V (2012). “Impactos Potenciales del Cambio Climático 15(2), 585–95. en la Seguridad Regional en América Latina.” Friedrich Ebert Stiftung. Nomura Global Economics and Strategy (2010). “The Coming Surge in Nel, P., and Righarts, M. (2008). “Natural Disasters and the Risk of Violent Food Prices.” Nomura: Tokyo. Civil Conflict.” International Studies Quarterly, 52(1), 159–85. North A. (2010). “Drought, Drop Out and Early Marriage: Feeling the Effects Nelson, G.C., Rosegrant, M.W., Koo, J., Robertson, R., Sulser, T., Zhu, of Climate Change in East Africa.” Equals, (24), 6. T., Ringler, C., Msangi, S., Palazzo, A., Batka, M., Magalhaes, M., Notaro, M., Alkolibi, F., Fadda, E., and Bakhrjy, F. (2013). “Trajectory Valmonte-Santos, R., Ewing, M., and Lee, D. (2009). “Climate Change. Analysis of Saudi Arabian Dust Storms.” Journal of Geophysical Impact on Agriculture and Costs of Adaptation.” International Food Research: Atmospheres, 118(12), 6028–43. Policy Research Institute: Washington DC. Novelo, J.A.M., and Tapia, L.R. (2011). “The Growth of Water Demand Nelson, G.C., Rosegrant, M.W., Koo, J., Robertson, R., Sulser, T., Zhu, in Mexico City and the Over-exploitation of its Aquifers.” Water T., Ringler, C., Msangi, S., Palazzo, A., Batka, M., Magalhaes, M., Resources in Mexico, Hexagon Series on Human and Environmental Valmonte-Santos, R., Ewing M., and Lee, D. (2010). “The Costs of Agri- Security and Peace, edited by Ú. Oswald Spring, Springer Berlin cultural Adaptation to Climate Change.” World Bank: Washington, DC. Heidelberg. Nelson, G.C., Rosegrant, M.W., Palazzo, A., Gray, I., Ingersoll, C., Robert- Novikov, I., Kalter-Leibovici, O., Chetrit, A., Stav, N., and Epstein, Y. (2012). son, R., Tokgoz, S., and Zhu, T. (2010). “Food Security and Climate “Weather Conditions and Visits to the Medical Wing of Emergency Change.” IFPRI Issue Brief 66, International Food Policy Research Rooms in a Metropolitan Area during the Warm Season in Israel: A Institute: Washington, DC. Predictive Model.” Int J Biometeorol, 56, 121–27. 262 B I BLIOGRA P HY Novikov, V., Simonett, O., Beilstein, M., Bournay, E., Berthiaume, C., Osman-Elasha, B. (2010). “Mapping of Climate Change Threats and Kirby, A., Daniel, C., Martino, L. De, Denisov, N., Rajabov, I., For- Human Development Impacts in the Arab Region. Arab Human ster, H., Baliev, B., Kayumov, A., Mustaeva, N., Abaikhanova, Z., Development Report.” UNDP Regional Bureau for Arab States: Kuzmichenok, V., Myagkov S., Erokhin, S., Isabaev, K., Reznikova, L., New York, NY. Safarov, M., Safarov, N., and Makhmadaliev, B. (2009). “Climate Ososkova, T., Gorelkin, N., and Chub, V. (2000), “Water Resources of Central Change in Central Asia—A Visual Synthesis.” Swiss Federal Office Asia and Adaptation Measures for Climate Change.” Environmental for the Environment (FOEN), Zoï Environment Network. Monitoring and Assessment, 61(1), 161–66. O’Brien, K., Leichenko, R., and Vogel, C. (2008). “Disaster Risk Reduction, Ostanciaux, É., Husson, L., Choblet, G., Robin, C., and Pedoja, K. (2012). Climate Change Adaptation, and Human Security.” A commissioned “Present-day Trends of Vertical Ground Motion along the Coast Lines.” report for the Norwegian Ministry of Foreign Affairs. Earth-Science Reviews, 110(1–4), 74–92. O’Halloran ,T.L., Law, B.E., Goulden, M.L., Wang, Z., Barr, J.G., Schaaf, C., Ostberg, S., Lucht, W., Schaphoff, S., and Gerten, D. (2013). “Critical Brown, M., Fuentes, J.D., Göckede, M., Black, A., and Engel, V. (2012), Impacts of Global Warming on Land Ecosystems.” Earth System “Radiative Forcing of Natural Forest Disturbances.” Global Change Dynamics, 4(2), 347–57. Biology, 18(2), 555–65. Østby, G. (2007). “Horizontal Inequalities, Political Environment, and Civil Odemis, B., Sangun, M.K., and Evrendilek, F. (2010). “Quantifying Long-term Conflict—Evidence from 55 Developing Countries, 1986–2003.” World Changes in Water Quality and Quantity of Euphrates and Tigris Rivers, Bank Policy Research Working Paper, Washington, DC. Turkey.” Environmental Monitoring and Assessment, 170(1-4), 475–90. Otto, I.M., Shkaruba, A., and Kireyeu, V. (2011). “The Rise of Multilevel OECD (2013). “Renewable Energies in the Middle East and North Africa.” Governance for Biodiversity Conservation in Belarus.” Environment Organization for Economic Cooperation and Development: Paris. and Planning C: Government and Policy, 29, 113–132. OECD/FAO (2013). “Agricultural Outlook 2013.” OECD/FAO: Paris/Rome. Painter, J. (2007). “Deglaciation in the Andean Region.” Human Develop- Okada, M., Iizumi, T., and Hayashi, Y. (2011). “Modeling the Multiple ment Report 2007/2008. Effects of Temperature and Radiation on Rice Quality.” Environmental Pan, Y., Birdsey, R. A, Fang, J., Houghton, R., Kauppi, P.E., Kurz, W.A., Research Letters, 6(034031). Phillips, O.L., Shvidenko, A., Lewis, S.L., Canadell, J.G., Ciais, P., Okin, G., Parsons, A., and Wainwright, J. (2009). “Do Changes in Con- Jackson, R.B., Pacala, S.W., McGuire, A.D., Piao, S., Rautiainen, A., nectivity Explain Desertification?” BioScience, 59(3), 237–44. Sitch, S., and Hayes, D. (2011), “A Large and Persistent Carbon Sink Oliveira, A., Trezza, R., Holzapfel, E., Lorite, I., and Paz, V. (2009). in the World’s Forests.” Science (New York, NY), 333(6045), 988–93. “Irrigation Water Management in Latin America.” Chilean Journal of Pardaens, A.K., Gregory, J.M., and Lowe, J.A. (2011). “A Model Study of Agricultural Research, 69, 7–16. Factors Influencing Projected Changes in Regional Sea Level over the Olson, D.M., and Dinerstein, E. (2002). “The Global 200: Priority Ecore- Twenty-first Century.” Climate Dynamics, 36(9-10), 2015–33. gions for Global Conservation.” Annals of the Missouri Botanical Parizek, B.R., Christianson, K., Anandakrishnan, S., Alley, R.B., Walker, Garden, 89(2), 199–224. R.T., Edwards, R. A., Wolfe, D.S., Bertini, G.T., Rinehart, S.K., Bind- Olsson, L., Opondo, M., Tschakert, P., Agrawal, A., Eriksen, S., Ma, S., schadler, R.A., and Nowicki, S.M.J. (2013). “Dynamic (In)stability of Perch, L., and Zakieldeen, S. (2014). “Livelihoods and Poverty.” Climate Thwaites Glacier, West Antarctica.” Journal of Geophysical Research: Change 2014: Impacts, Adaptation, and Vulnerability. Contribution of Earth Surface, 118(2), 638–55. Working Group II to the Fifth Assessment Report of the Intergovern- Parry, M. (2007). “The Implications of Climate Change for Crop Yields, mental Panel on Climate Change. Chapter 13. Final Draft, edited by Global Food Supply, and Risk of Hunger.” ICRISAT eJournal, 4(1), 1–44. C. Field, V. Barros, K. Mach, and M. Mastrandrea, IPCC AR5 WGII, Parry, M. (2010). “Copenhagen Number Crunch.” Nature Reports Climate Cambridge University Press: Cambridge. Change, 4(1002), 18–19. Olsson, O., Gassmann, M., Wegerich, K., and Bauer, M. (2010). “Identification Parry, M., Evans, A., Rosegrant, M., and Wheeler, T. (2009), “Climate of the Effective Water Availability from Streamflows in the Zerafshan Change and Hunger Responding to the Challenge.” World Food river Basin, Central Asia.” Journal of Hydrology, 390(3-4), 190–97. Programme: Rome. ONU-Habitat (2012). Estado de las Ciudades de América Latina y El Pasicko, R., Brankovic, C., and Simic, Z. (2012). “Assessment of Climate Caribe 2012: Rumbo a una Nueva Transición Urbana. Programa Change Impacts on Energy Generation from Renewable Sources in de las Naciones Unidas para los Asentamientos Humanos, ONU- Croatia.” Renewable Energy, 46(2012), 224–31. Habitat: Nairobi. Patz, J.A., Campbell-Lendrum, D., Holloway, T., and Foley, J.A. (2005). Oppenheimer, M. (1998). “Global Warming and the Stability of the West “Impact of Regional Climate Change on Human Health.” Nature, Antarctic Ice Sheet.” Nature, 393, 325–32. 438(7066), 310–17. Orlowsky, B., and Seneviratne, S.I. (2011). “Global Changes in Extreme Paz, C. (2012), “Atlas Sociodemográfico de los Pueblos Indígenas y Events: Regional and Seasonal Dimension.” Climatic Change, 110(3–4), Afrodescendientes en Colombia.” CELADE—División de Población 669–96. de la CEPAL: Santiago de Chile. Orlowsky, B., and Seneviratne, S.I., (2013). “Elusive Drought: Uncertainty Peltier, W.R., and Andrews, J.T. (1976). “Glacial-Isostatic Adjustment—I. in Observed Trends and Short- and Long-term CMIP5 Projections.” The Forward Problem.” Geophysical Journal of the Royal Astronomical Hydrology and Earth System Sciences, 17(5), 1765–81. Society, 46(3), 605–46. Ortiz, I., Burke, S., Berrada, M., and Cortes, H. (2013). “World Protests Peña, E., Zúñiga, O., and Peña, J. (2011). “Accounting the Carbon Storage 2006–2013.” Institute for Policy Dialogue and Friedrich-Ebert-Stiftung: in Disturbed and Non-Disturbed Tropical Andean Ecosystems.” in New York. Planet Earth 2011—Global Warming Challenges and Opportunities for Ortiz, R., Sayre, K.D., Govaerts, B., Gupta, R., Subbarao, G.V., Ban, T., Policy and Practice. Edited by E. G. Carayannis. InTech Hodson, D., Dixon, J.M., Iván Ortiz-Monasterio, J., and Reynolds, Perch-Nielsen, S., Bättig, M., and Imboden, D. (2008). “Exploring the M. (2008), “Climate Change: Can Wheat Beat the Heat?” Agriculture, Link between Climate Change and Migration.” Climatic Change, Ecosystems & Environment, 126(1–2), 46–58. 91(3-4), 375–93. Oshaghi, M.A., Ravasan, N.M., Javadian, E., Rassi, Y., Sadraei, J., Enayati, Perch-Nielsen, S.L. (2010). “The Vulnerability of Beach Tourism to Climate A.A., Vatandoost, H., Zare, Z., and Emami, S.N. (2009). “Application Change—An Index Approach.” Climatic Change, 100(3-4), 579–606. of Predictive Degree Day Model for Field Development of Sandfly Peretz, C., Biggeri, A., Alpert, P., and Baccini, M. (2012). “The Effect of Vectors of Visceral Leishmaniasis in Northwest of Iran.” Journal of Heat Stress on Daily Mortality in Tel Aviv, Israel.” National Security Vector Borne Disease, 46, 247–54. and Human Health Implications of Climate Change, edited by H. 263 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Fernando, NATO Science for Peace and Security Series C: Environ- los Objetivos de Desarrollo del Milenio.” Notas de poblacion No. 86, mental Security: Springer Science and Business Media. CEPAL: Santiago de Chile. Perrette, M., Landerer, F., Riva, R., Frieler, K., and Meinshausen, M. Porter, J.R., Liyong, X., Challinor, A., Cochrane, K., Howden, M., Iqbal, (2013). “A Scaling Approach to Project Regional Sea-level Rise and M.M., Lobell, D., and Travasso, M.I. (2014). “Food Security and Food its Uncertainties.” Earth System Dynamics, 4(1), 11–29. Production Systems.” IPCC 2014: Climate Change 2014: Impacts, Perry, A.L., Low, P.J., Ellis, J.R., and Reynolds, J.D. (2005). “Climate Adaptation, and Vulnerability. Contribution of Working Group II to the Change and Distribution Shifts in Marine Fishes,” Science (New York, Fifth Assessment Report of the Intergovernmental Panel on Climate N.Y.), 308(5730), 1912–15. Change. Chapter 7. Final Draft, IPCC AR5 WGII, Cambridge University Peters, G.P., Andrew, R.M., Boden, T., Canadell, J.G., Ciais, P., Quéré, C. Le, Press: Cambridge and New York. Marland, G., Raupach, M.R., and Wilson, C. (2013). “The Challenge to Portmann, F.T., Döll, P., Eisner, S., and Flörke, M. (2013). “Impact of Cli- keep Global Warming Below 2°C.” Nature Climate Change 3(1) 2–4. mate Change on Renewable Groundwater Resources: Assessing the Pey, J., Querol, X., Alastuey, A., Forastiere, F., and Stafoggia, M. (2013). Benefits of Avoided Greenhouse Gas Emissions using Selected CMIP5 “African Dust Outbreaks over the Mediterranean Basin during 2001–2011: Climate Projections.” Environmental Research Letters, 8(2), 1–14. PM10 Concentrations, Phenomenology, and Trends, and its Relation Pörtner, H.O., and Farrell, A.P. (2008). “Physiology and Climate Change.” with Synoptic and Mesoscale Meteorology.” Atmospheric Chemistry Science, 322(October), 690–92. and Physics, 13(3), 1395–1410. Pörtner, H.O., Karl, D., Boyd, P.W., Cheung, W., Lluch-Cota, S.E., Nojiri, Peyrouse, S. (2013). “Food Security in Central Asia. A Public Policy Y., Schmidt, D., and Zavialov, P. (2014). “Ocean Systems.” Climate Challenge.” PONARS Eurasia Policy Memo No. 300, Elliott School of Change 2014: Impacts, Adaptation, and Vulnerability. Contribution of International Affairs, George Washington University: Washington, DC. Working Group II to the Fifth Assessment Report of the Intergovern- Phillips, O.L., Aragão, L.E.O.C., Lewis, S.L., Fisher, J.B., Lloyd, J., López- mental Panel on Climate Change. Chapter 6. Final Draft, edited by González, G., Malhi, Y., Monteagudo, A., Peacock, J., Quesada, C.A., K. Drinkwater and A. Polonsky, IPCC AR5 WGII, Cambridge University van der Heijden, G., Almeida, S., Amaral, I., Arroyo, L., Aymard, G., Press: Cambridge, UK. Baker, T.R., Bánki, O., Blanc, L., Bonal, D., Brando, P., Chave, J., de Portnov, B.A., Paz, S., and Linn, S. (2011). “What Does the Inflow of Patients Oliveira, A.C.A., Cardozo, N.D., Czimczik, C.I., Feldpausch, T.R., into the Rambam Medical Center in Haifa Tells Us about Outdoor Tem- Freitas, M.A., Gloor, E., Higuchi, N., Jiménez, E., Lloyd, G., Meir, P., peratures and Air Pollution?” Geography Research Forum, 31, 39–52. Mendoza, C., Morel, A., Neill, D.A., Nepstad, D., Patiño, S., Peñuela, Postigo, J.A.R. (2010). “Leishmaniasis in the World Health Organization M.C., Prieto, A., Ramírez, F., Schwarz, M., Silva, J., Silveira, M., Eastern Mediterranean Region.” International Journal of Antimicrobial Thomas, A.S., Steege, H. Ter, Stropp, J., Vásquez, R., Zelazowski, P., Agents, 365, 562–65. Alvarez Dávila, E., Andelman, S., Andrade, A., Chao, K.J., Erwin, T., Poulter, B., Aragão, L., Heyder, U., Gumpenberger, M., Heinke, J., Langer- Di Fiore, A., Honorio, C E., Keeling, H., Killeen, T.J., Laurance, W.F., wisch, F., Rammig, A., Thonicke, K., and Cramer, W. (2010). “Net Peña Cruz, A., Pitman, N.C.A., Núñez Vargas, P., Ramírez-Angulo, H., Biome Production of the Amazon Basin in the 21st Century.” Global Rudas, A., Salamão, R., Silva, N., Terborgh, J., and Torres-Lezama, Change Biology, 16(7), 2062–75. A. (2009). “Drought Sensitivity of the Amazon Rainforest.” Science Pousa, J.L., D’Onofrio, E.E., Fiore, M.M.E., and Kruse, E.E. (2013). “Envi- (New York, N.Y.), 323(5919), 1344–47. ronmental Impacts and Simultaneity of Positive and Negative Storm Pierre, C., Bergametti, G., Marticorena, B., Mougin, E., Bouet, C., and Sch- Surges on the Coast of the Province of Buenos Aires, Argentina.” mechtig, C. (2012). “Impact of Vegetation and Soil Moisture Seasonal Environmental Earth Sciences, 68(8), 2325–35. Dynamics on Dust Emissions over the Sahel.” Journal of Geophysical Poveda, G. (2004). “La Hidroclimatología de Colombia: Una Síntesis desde Research, 117(D6), D06114. la Escala Inter-decadal hasta la Escala Diurna.” Rev. Acad.Colomb. Piguet, É., Pécoud, A., and de Guchteneire, P. (2011). Migration and Climate Cienc., 28(107), 201–22. Change, Cambridge University Press: Cambridge. Poveda, G., Estrada-Restrepo, O.A., Morales, J.E., Hernandez, O.O., Galeano, Pincha, P.C. (2008). “Gender Sensitive Disaster Management. A Toolkit A., and Osorio, S. (2011). “Integrating Knowledge and Management for Practitioners.” Earthworm Books for Oxfam America and NAN- regarding the Climate–Malaria Linkages in Colombia.” Current Opinion BAN Trust. in Environmental Sustainability, 3, 448–60. Pinti, V., Marcolli, C., Zobrist, B., Hoyle, C.R., and Peter, T. (2012). “Ice Poveda, G., Jaramillo, A., Gil, M.M., Quiceno, N., and Mantilla, R.I. Nucleation Efficiency of Clay Minerals in the Immersion Mode.” (2001). “Seasonality in ENSO-related Precipitation, River Discharges, Atmospheric Chemistry and Physics, 12(13), 5859–78. Soil Moisture, and Vegetation Index in Colombia.” Water Resources Pinto da Cunha, J. (2011). “Movilidad Espacial, Vulnerabilidad y Seg- Research, 37(8), 2169–78. regación Socioespacial: Reflexiones a Partir del Estudio de la Región Poveda, G., and Pineda, K. (2010). “Advances in Geosciences Reassess- Metropolitana de Campinas, 2007.” ECLAC, Notas de Poblacion 93. ment of Colombia’s Tropical Glaciers Retreat Rates: Are They Bound Pleijel, H., and Uddling, J. (2012). “Yield vs. Quality Trade-offs for Wheat to Disappear during the 2010–2020 Decade?” Advances in Geosciences, in Response to Carbon Dioxide and Ozone.” Global Change Biology, 22(2009), 107–16. 18(2), 596–605. Power, S., Delage, F., Chung, C., Kociuba, G., and Keay, K. (2013). “Robust Poland, J.F., and Davis, G.H. (1969). “Land Subsidence due to Withdrawal Twenty-first-century Projections of El Niño and Related Precipitation of Fluids.” Rev Eng Geol, 2, 187–269. Variability.” Nature, 502(7472), 541–45. Ponce-Reyes, R., Reynoso-Rosales, V.-H., Watson, J.E.M., VanDerWal, J., Pradhan, E.K., West, K., Katz, J., LeClerq, S.C., Khatry, S.K., and Shres- Fuller, R.A., Pressey, R.L., and Possingham, H.P. (2012). “Vulnerabil- tha, S.R. (2007). “Risk of Food-related Mortality in Nepal.” Disasters, ity of Cloud Forest Reserves in Mexico to Climate Change.” Nature 31(1), 57–70. Climate Change, 2(6), 448–52. Prato, B., and Longo, R. (2012). “Empowerment of Poor Rural People through Popescu, I., Brandimarte, L., and Peviani, M. (2014). “Effects of Climate Initiatives in Agriculture and Natural Resource Management.” Paris. Change over Energy Production in La Plata Basin.” International Priestley, M., and Hemingway, L. (2006). “Disability and Disaster Recov- Journal of River Basin Management, (May 2014), 1–9. ery: A Tale of Two Cities?” Journal of Social Work in Disability and del Popolo, F., and Oyarce, A.M. (2005). “Población Indígena de América Rehabilitation, 5(3-4), 23–42. Latina: Perfil Sociodemográfico en el Marco de la CIPD y de las Metas del Prudhomme, C., Giuntoli, I., Robinson, E.L., Clark, D.B., Arnell, N.W., Milenio.” CELADE—División de Población, CEPAL: Santiago de Chile. Dankers, R., Fekete, B.M., Franssen, W., Gerten, D., Gosling, S.N., del Popolo, F., Oyarce, M.A., and Ribotta, B. (2009). “Indígenas Urbanos Hagemann, S., Hannah, D.M., Kim, H., Masaki, Y., Satoh, Y., Stacke, T., en América Latina: Algunos Resultados Censales y su Relación con Wada, Y., and Wisser, D. (2013). “Hydrological Droughts in the 21st 264 B I BLIOGRA P HY Century, Hotspots, and Uncertainties from a Global Multimodel Ensem- Raven, J. (2005). “Ocean Acidification due to Increasing Atmospheric ble Experiment.” Proceedings of the National Academy of Sciences. Carbon Dioxide.” The Royal Society: Cardiff. Pryor, S.C., and Barthelmie, R.J. (2013). “Assessing the Vulnerability Ray, D., Nepstad, D., and Moutinho, P. (2005). “Micrometeorological and of Wind Energy to Climate Change and Extreme Events.” Climatic Canopy Controls of Fire Susceptibility in a Forested Amazon Land- Change, 121(1), 79–91. scape.” Ecological Applications, 15(5), 1664–78. Quiring, S.M., Schumacher, A.B., and Guikema, S.D. (2013). “Incorporating Renssen, H., Lougheed, B.C., Aerts, J.C.J.H., de Moel, H., Ward, P.J., Hurricane Forecast Uncertainty into a Decision Support Application and Kwadijk, J.C.J. (2007). “Simulating Long-term Caspian Sea-level for Power Outage Modeling.” Bulletin of the American Meteorologi- Changes: The Impact of Holocene and Future Climate Conditions.” cal Society. Earth and Planetary Science Letters, 261(3–4), 685–93. Rabatel, A., Francou, B., Soruco, A., Gomez, J., Cáceres, B., Ceballos, Republic of Lebanon (2011). “Lebanon’s Second National Communication J.L., Basantes, R., Vuille, M., Sicart, J.E., Huggel, C., Scheel, M., to the United Nations Framework Convention on Climate Change.” Lejeune, Y., Arnaud, Y., Collet, M., Condom, T., Consoli, G., Favier, Beirut. V., Jomelli, V., Galarraga, R., Ginot, P., Maisincho, L., Mendoza, J., Republique Algerienne Democratique et Populaire (2010). “Seconde Com- Ménégoz, M., Ramirez, E., Ribstein, P., Suarez, W., Villacis, M., and munication Nationale de l`Algerie sur les Changements Climatiques Wagnon, P. (2013), “Current State of Glaciers in the Tropical Andes: A a la CCNUCC.” Ministere de l´Amenagement du Territoire et de Multi-century Perspective on Glacier Evolution and Climate Change.” l´Environment: Alger. The Cryosphere, 7(1), 81–102. Restrepo, J.C., Ortíz, J.C., Pierini, J., Schrottke, K., Maza, M., Otero, L., and Radic´, V., Bliss, A., Beedlow, A.C., Hock, R., Miles, E., and Cogley, J.G. Aguirre, J. (2014). “Freshwater Discharge into the Caribbean Sea from (2013). “Regional and Global Projections of Twenty-first Century the Rivers of Northwestern South America (Colombia): Magnitude, Glacier Mass Changes in Response to Climate Scenarios from Global Variability, and Recent Changes (Accepted Manuscript).” Journal of Climate Models.” Climate Dynamics, 142(1–2), 37–58. Hydrology, 509, 266–81. El Raey, M. (2011). “Mapping Areas Affected by Sea-level Rise due to Reuveny, R. (2007). “Climate-change-induced Migration and Violent Climate Change in the Nile Delta until 2100.” Coping with Global Conflict.” Political Geography, 26(6), 656–73. Environmental Change, Disasters, and Security, edited by H. G. Reyer, C.P.O., Leuzinger, S., Rammig, A., Wolf, A., Bartholomeus, R.P., Brauch, Ú. Oswald Spring, C. Mesjasz, J. Grin, P. Kameri-Mbote, Bonfante, A., de Lorenzi, F., Dury, M., Gloning, P., Abou Jaoudé, B. Chourou, P. Dunay, and J. Birkmann, Springer-Verlag: Berlin, R., Klein, T., Kuster, T.M., Martins, M., Niedrist, G., Riccardi, M., Heidelberg, New York. Wohlfahrt, G., de Angelis, P., de Dato, G., François, L., Menzel, A., El Raey, M., Dewidar, K., and El Hattab, M. (1999). “Adaptation to the and Pereira, M. (2013). “A Plant’s Perspective of Extremes: Terrestrial Impacts of Sea-level Rise in Egypt.” Climate Research, 12, 117–28. Plant Responses to Changing Climatic Variability.” Global change Rahmstorf, S. (2007). “A Semi-empirical Approach to Projecting Future biology, 19(1), 75–89. Sea-level Rise.” Science, 315(5810), 368–70. Reynolds, J.F., Smith, D.M.S., Lambin, E.F., Turner, B.L., Mortimore, M., Rakhmatullaev, S., Huneau, F., Kazbekov, J., Celle-Jeanton, H., Motelica- Batterbury, S.P.J., Downing, T.E., Dowlatabadi, H., Fernández, R.J., Heino, M., Coustumer, P. Le, and Jumanov, J. (2012). “Groundwater Herrick, J.E., Huber-Sannwald, E., Jiang, H., Leemans, R., Lynam, T., Resources of Uzbekistan: An Environmental and Operational Overview.” Maestre, F.T., Ayarza, M., and Walker, B. (2007). “Global Desertifica- Central European Journal of Geosciences, 4(1), 67–80. tion: Building a Science for Dryland Development.” Science (New Raleigh, C., Jordan, L., and Salehyan, I. (2008), “Assessing the Impact York, NY), 316(5826), 847–51. of Climate Change on Migration and Conflict.” World Bank: Wash- Reynolds, J.F., and Stafford Smith, D.M. (Eds.) (2002). Global Desertification: ington, DC. Do Humans Cause Deserts? Dahlem University Press: Berlin. Raleigh, C., and Urdal, H. (2007). “Climate Change, Environmental Deg- Riahi, K., Kriegler, E., Johnson, N., Bertram, C., den Elzen, M., Eom, J., radation, and Armed Conflict.” Political Geography, 26(6), 674–94. Schaeffer, M., Edmonds, J., Isaac, M., Krey, V., Longdon, T., Luderer, Ramirez-Villegas, J., Salazar, M., Jarvis, A., and Navarro-Racines, C.E. G., Mejean, A., McCollum, D.L., Mima, S., Turton, H., van Vuuren, (2012). “A Way Forward on Adaptation to Climate Change in Colom- D.P., Wada, K., Bosetti, V., Capros, P., Criqui, P., and Kainuma, M. bian Agriculture: Perspectives Towards 2050.” Climatic Change, (2013). “Locked into Copenhagen Pledges—Implications of Short-term 115(3–4), 611–28. Emission Targets for the Cost and Feasibility of Long-term Climate Rammig, A., Jupp, T., Thonicke, K., Tietjen, B., Heinke, J., Ostberg, S., Goals.” Technological Forecasting & Social Change, In Review. Lucht, W., Cramer, W., and Cox, P. (2010). “Estimating the Risk of Ribot, J. (2010). “Vulnerability Does Not Just Fall From the Sky: Amazonian Forest Dieback.” The New Phytologist, 187(3), 694–706. Toward Multi-scale Pro-poor Climate Policy.” Social Dimensions Randerson, J.T., Liu, H., Flanner, M.G., Chambers, S.D., Jin, Y., Hess, of Climate Change: Equity and Vulnerability in a Warming World. P.G., Pfister, G., Mack, M.C., Treseder, K.K., Welp, L.R., Chapin, F.S., Chapter 7, edited by R. Mearns and A. Norton, World Bank: Harden, J.W., Goulden, M.L., Lyons, E., Neff, J.C., Schuur, E.A.G., Washington, DC. and Zender C.S. (2006). “The Impact of Boreal Forest Fire on Climate Ribotta, B. (2011). “Atlas Sociodemográfico de los Pueblos Indígenas del Warming.” Science (New York, NY), 314(5802), 1130–32. Peru.” CELADE—División de Población de la CEPAL: Santiago de Chile. Randolph, S.E., and Rogers, D.J. (2000). “Fragile Transmission Cycles of Ribotta, B. (2012a). “Atlas Sociodemográfico de los Pueblos Indígenas de Tick-Borne Encephalitis Virus may be Disrupted by Predicted Climate Chile.” CELADE—División de Población de la Comisión Económica Change.” Proceedings: Biological Sciences, 267(1454), 1741–44. para América Latina y el Caribe (CEPAL): Santiago de Chile. Rangel, M. (2006). “Pueblos Indígenas y Afrodescendientes de América Ribotta, B. (2012b). “Atlas Diagnóstico Sociodemográfico de los Pueblos Latina y el Caribe: Información Sociodemográfica para Políticas y Indigenas Argentina.” CELADE—División de Población de la CEPAL: Programas.” Pueblos indígenas y afrodescendientes de América Latina Santiago de Chile. y el Caribe, Comisión Económica para América Latina y el Caribe Rignot, E., Jacobs, S., Mouginot, J., and Scheuchl, B. (2013). “Ice-shelf (CEPAL): Santiago de Chile. Melting around Antarctica.” Science, 341(6143), 266–70. Rauscher, S.A., Giorgi, F., Diffenbaugh, N.S., and Seth, A. (2008). “Exten- Rignot, E., Mouginot, J., Morlighem, M., Seroussi, H., and Scheuchl, B. sion and Intensification of the Meso-American Mid-summer Drought (2014). “Widespread, Rapid Grounding Line Retreat of Pine Island, in the Twenty-first Century.” Climate Dynamics, 31(5), 551–71. Thwaites, Smith, and Kohler Glaciers, West Antarctica, from 1992 to Ravallion, M., Chen, S., and Sangraula, P. (2008). “New Evidence on the 2011.” Geophysical Research Letters. Urbanization of Global Poverty.” Background Paper to The World RIMISP (2011). “Poverty and Inequality Latin American Report.” Latin Development Report 2008, World Bank: Washington, DC. American Center for Rural Development (RIMISP). 265 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Rinnan, R., Michelsen, A., Bååth, E., and Jonasson, S. (2007). “Fifteen Years Rowell, D.P. (2011). “Sources of Uncertainty in Future Changes in Local of Climate Change Manipulations Alter Soil Microbial Communities in Precipitation.” Climate Dynamics, 39(7–8), 1929–50. a Subarctic Heath Ecosystem.” Global Change Biology, 13(1), 28–39. Ruane, A.C., Cecil, L.D., Horton, R.M., Gordón, R., McCollum, R., Brown, Rio, M.H., Pascual, A., Poulain, P.M., Menna, M., Barceló, B., and Tintoré, D., Killough, B., Goldberg, R., Greeley, A.P., and Rosenzweig, C. (2013). J. (2014). “Computation of a New Mean Dynamic Topography for the “Climate Change Impact Uncertainties for Maize in Panama: Farm Mediterranean Sea from Model Outputs, Altimeter Measurements and Information, Climate Projections, and Yield Sensitivities.” Agricultural Oceanographic In-situ Data.” Ocean Science Discussions, 11(1), 655–92. and Forest Meteorology, 170, 132–45. Riyad, M., Chiheb, S., and Soussi-Abdallaoui, M. (2013). “Cutaneous Rubin, O. (2011). “Conflict and Climate Change.” Reducing Poverty, Pro- Leishmaniasis caused by Leishmania Major in Morocco: Still a Topical tectingLivelihoods, and Building Assets in a Changing Climate Social qQuestion.” Eastern Mediterranean Health Journal, 19(5). Implications of Climate Change for Latin America and the Caribbean, Roberts, M., and Schlenker, W. (2010). “Identifying Supply and Demand edited by D. Verner, The World Bank: Washington, DC. Elasticities of Agricultural Commodities: Implications for the US Ethanol Ruml, M., Vukovi c c, M., Djurdjevi´ ´, A., Vujadinovi´ c, V., Rankovi´ c, Z., c-Vasi´ Mandate.” American Economic Review, 103(6), 2265–95. Atanackovi´ cev, B., Markovi´ c, Z., Sivˇ c, N., Matijaševi´ c, S., and Petrovi´c, Rodrigues, L.C., van den Bergh, J.C.J.M., and Ghermandi, A. (2013). N. (2012). “On the Use of Regional Climate Models: Implications of “Socioeconomic Impacts of Ocean Acidification in the Mediterranean Climate Change for Viticulture in Serbia.” Agricultural and Forest Sea.” Marine Policy, 38, 447–56. Meteorology, 158–159, 53–62. Roessig, J.M., Woodley, C.M., Cech, J.J., and Hansen, L.J. (2005). “Effects Runyan, C. (2012). “Physical and Biological Feedbacks of Deforestation.” of Global Climate Change on Marine and Estuarine Fishes and Fisher- Reviews of Geophysics, 50(RG4006), 1–32. ies.” Fish Biology and Fisheries, (14), 251–75. Ruth, M., and Ibarran, M. (Eds.) (2009). Distributional Impacts of Climate Rogelj, J., Meinshausen, M., and Knutti, R. (2012). “Global Warming Change and Disasters. Edward Elgar: Cheltenham and Northampton, MA. under Old and New Scenarios using IPCC Climate Sensitivity Range Saab, N. (2012). “Survival Options: Ecological Footprint of Arab Countries, Estimates.” Nature Climate Change, 2(4), 248–53. Arab Forum for Environment and Development.” Beirut. Rogelj, J., McCollum, D. L., Reisinger, A., Meinshausen, M., and Riahi, K. Saatchi, S.S., Harris, N.L., Brown, S., Lefsky, M., Mitchard, E.T.A., Salas, (2013a). “Probabilistic cost estimates for climate change mitigation”. W., Zutta, B.R., Buermann, W., Lewis, S.L., Hagen, S., Petrova, S., Nature, 493, 79–83. White, L., Silman, M., and Morel, A. (2011). “Benchmark Map of Rogelj, J., McCollum, D. L., O‘Neill, B. C., and Riahi, K. (2013b). “2020 Forest Carbon Stocks in Tropical Regions across Three Continents.” emissions levels required to limit warming to below 2 °C”. Nature Proceedings of the National Academy of Sciences of the United States Climate Change, 3, 405–412. of America, 108(24), 9899–9904. Rogelj, J., Meinshausen, M., Sedlᡠcek, J., and Knutti, R. (2014). “Implica- Safa, A., and Hilmi, N. (2012). “The Impact of Climate Change on Coastal tions of Potentially Lower Climate Sensitivity on Climate Projections Tourism in MENA Countries.” Topics in Middle Eastern and African and Policy.” Environmental Research Letters, 9(3), 031003. Economies, 14, 245–64. Rojas-Soto, O.R., Sosa, V., and Ornelas, J.F. (2012). “Forecasting Cloud Safriel, U. (2009). “Status of Desertification in the Mediterranean Region.” Forest in Eastern and Southern Mexico: Conservation Insights under Water Scarcity, Land Degradation, and Desertification in the Medi- Future Climate Change Scenarios.” Biodiversity and Conservation, terranean Region. NATO Science for Peace and Security Series C: 21(10), 2671–90. Environmental Security, edited by J. L. Rubio, U. Safriel, R. Daussa, Romanovsky, V.E., Drozdov, D.S., Oberman, N.G., Malkova, G.V., Kholodov, W. Blum, and F. Pedrazzini, Springer Netherlands: Dordrecht. A.L., Marchenko, S.S., Moskalenko, N.G., Sergeev, D.O., Ukraintseva, Safriel, U., and Adeel, Z. (2005). “Dryland Systems.” Millenium Assess- N.G., Abramov, A.A., Gilichinsky, D.A., and Vasiliev, A.A. (2010). ment Ecosystem. Ecosystems and Human Well-Being: Current State “Thermal State of Permafrost in Russia.” Permafrost and Periglacial and Trends, edited by R. Hassan, R. Scholes, and N. Ash, Island Press: Processes, 21(2), 136–55. Washington, DC. Romero Lankao, P. (2010). “Water in Mexico City: What Will Climate Change Safriel, U., and Adeel, Z. (2008). “Development Paths of Drylands: Bring to its History of Water-related Hazards and Vulnerabilities?” Thresholds and Sustainability.” Sustainability Science, 3(1), 117–23. Environment and Urbanization, 22(1), 157–78. Salazar-Lindo, E. (2008). “ENSO and Cholera in South America: What Can Ronchail J., Bourrel L., Cochonneau G., Vauchel P., Phillips L., Castro A., We Learn about it from the 1991 Cholera Outbreak?” International Guyot J.-L. and de Oliveira E. (2005), “Inundations in the Mamoré basin Journal of Environment and Health, 2(1). (south-western Amazon—Bolivia) and sea-surface temperature in the Sale, P.F., Butler IV, M.J., Hooten, A.J., Kritzer, J.P., Lindeman, K.C., Pacific and Atlantic Oceans,” Journal of Hydrology, 302(1–4), 223–38. Sadovy de Mitcheson, Y.J., Steneck, R.S., and van Lavieren, H. (2008). Rosenzweig, C., Elliott, J., Deryng, D., Ruane, A.C., Müller, C., Arneth, “Stemming Decline of the Coastal Ocean: Rethinking Environmental A., Boote, K.J., Folberth, C., Glotter, M., Khabarov, N., Neumann, Management.” UNU-UNWEH: Hamilton, Canada. K., Piontek, F., Pugh, T.A.M., Schmid, E., Stehfest, E., Yang, H., and Salehi, M., Mohammad, K., Farahani, M.M., Zeraati, H., Nourijelyani, K., Jones, J.W. (2013a). “Assessing Agricultural Risks of Climate Change and Zayeri, F. (2008). “Spatial Modeling of Malaria Incidence Rates in the 21st Century in a Global Gridded Crop Model Intercomparison.” in Sistan and Baluchistan Province, Islamic Republic of Iran.” Saudi Proceedings of the National Academy of Sciences of the United States Med J., 29(2), 1791–96. of America, (14), 1–6. Salick, J., and Ross, N. (2009). “Traditional Peoples and Climate Change.” Rosenzweig, C., Jones, J.W., Hatfield, J.L., Ruane, A.C., Boote, K.J., Thor- Global Environmental Change, 19(2), 137–39. burn, P., Antle, J.M., Nelson, G.C., Porter, C., Janssen, S., Asseng, Sallenger, A.H., Doran, K.S., and Howd, P.A. (2012). “Hotspot of Acceler- S., Basso, B., Ewert, F., Wallach, D., Baigorria, G., and Winter, J.M. ated Sea-level Rise on the Atlantic Coast of North America.” Nature (2013b). “The Agricultural Model Intercomparison and Improvement Climate Change, 2(8), 1–5. Project (AgMIP): Protocols and Pilot Studies.” Agricultural and Forest Samaniego, J. (2009). “ECLAC Climate Change and Development in Latin Meteorology, 170, 166–82. America and the Caribbean. Overview 2009.” ECLAC. Rossing, T., and Rubin, O. (2011). “Climate Change, Disaster Hot Spots, Samanta, A., Ganguly, S., and Myneni, R.B. (2011). “MODIS Enhanced and Asset Erosion.” Reducing Poverty, Protecting Livelihoods, and Vegetation Index Data Do Not Show Greening of Amazon Forests Dur- Building Assets in a Changing Climate Social Implications of Climate ing the 2005 Drought.” The New Phytologist, 189(1), 11–15. Change for Latin America and the Caribbean, edited by D. Verner, Sampaio, G., Nobre, C., Costa, M.H., Satyamurty, P., Soares-Filho, B.S., and The World Bank: Washington, DC. Cardoso, M. (2007). “Regional Climate Change over Eastern Amazonia 266 B I BLIOGRA P HY Caused by Pasture and Soybean Cropland Expansion.” Geophysical Transition and Global Change.” Environmental Science & Policy, Research Letters, 34(17), L17709. 13(7), 620–36. Samuels, R., Rimmer, A., Hartmann, A., Krichak, S., and Alpert, P. (2010). Schmidhuber, J., and Tubiello, F.N., (2007). “Global Food Security under “Climate Change Impacts on Jordan River Flow: Downscaling Applica- Climate Change.” Proceedings of the National Academy of Sciences tion from a Regional Climate Model.” Journal of Hydrometeorology, of the United States of America, 104(50), 19703–8. 11(4), 860–79. Schneider, C., Laize, C.L.R., Acreman, M., and Florke, M. (2013). “How Satterthwaite, D., Huq, S., Pelling, M., Reid, H., and Lankao, P.R. (2007). Will Climate Change Modify River Flow Regimes in Europe?” Hydrol- “Adapting to Climate Change in Urban Areas: The Possibilities and ogy and Earth System Sciences, 17, 325–39. Constraints in Low- and Middle-Income Nations.” International Institute Scholes, R., Settele, J., Betts, R., Bunn, S., Leadley, P., Nepstad, D., Over- for Environment and Development: London. peck, J., and Taboada, M.A. (2014). “Terrestrial and Inland Water El Sayed Frihy, O., Deabes, E.A., Shereet, S.M., and Abdalla, F.A. (2010). Systems.” Climate Change 2014: Impacts, Adaptation, and Vulner- “Alexandria-Nile Delta Coast, Egypt: Update and Future Projection of ability. Contribution of Working Group II to the Fifth Assessment Relative Sea-level Rise.” Environmental Earth Sciences, 61(2), 253–73. Report of the Intergovernmental Panel on Climate Change. Chapter Schaefer, K., Zhang, T., Bruhwiler, L., and Barrett, A.P. (2011). “Amount 4. Final Draft, edited by A. Fischlin, J. M. Moreno, and T. Root, IPCC and Timing of Permafrost Carbon Release in Response to Climate AR5 WGII, Cambridge University Press: Cambridge. Warming.” Tellus B, 63(2), 165–80. Schroth, G., Laderach, P., Dempewolf, J., Philpott, S., Haggar, J., Eakin, Schaefer, M., Machguth, H., Falvey, M., and Casassa, G. (2013). “Modeling H., Castillejos, T., Garcia Moreno, J., Soto Pinto, L., Hernandez, R., Past and Future Surface Mass Balance of the Northern Patagonia Ice- Eitzinger, A., and Ramirez-Villegas, J. (2009). “Towards a Climate field.” Journal of Geophysical Research: Earth Surface, 118(2), 571–88. Change Adaptation Strategy for Coffee Communities and Ecosystems Schaeffer, M., Hare, W., Rahmstorf, S., and Vermeer, M. (2012). “Long-term in the Sierra Madre de Chiapas, Mexico.” Mitigation and Adaptation Sea-level Rise Implied by 1.5°C and 2°C Warming Levels.” Nature Strategies for Global Change, 14(7), 605–25. Climate Change, advance on. Schubert, R., Schellnuhber, H.J., Buchmann, N., Epiney, A., Griesshammer, Schaphoff, S., Heyder, U., Ostberg, S., Gerten, D., Heinke, J., and Lucht, R., Kulessa, M., Messner, D., Rahmstorf, S., and Schmid, J. (2007). W. (2013). “Contribution of Permafrost Soils to the Global Carbon “Climate Change as a Security Risk.” German Advisory Council on Budget.” Environmental Research Letters, 8(1), 014026. Global Change (WGBU): Berlin. Schaphoff, S., Lucht, W., Gerten, D., Sitch, S., Cramer, W., and Prentice, Scott, D., Lemieux, C., and Malone, L. (2011). “Climate Services to Sup- I.C. (2006). “Terrestrial BiosphereCarbon Storage under Alternative port Sustainable Tourism and Adaptation to Climate Change.” Climate Climate Projections.” Climatic Change, 74(1–3), 97–122. Research, 47(1), 111–22. Schär, C., Vidale, P.L., Lüthi, D., Frei, C., Häberli, C., Liniger, M.A., and Seely, M., Al-Amoud, A.I., Dawn, C., Henschel, J., Kinahan, J.J., Klin- Appenzeller, C. (2004). “The Role of Increasing Temperature Variability tenberg, P., León, A., Morrison, S., Roedern, C., Abraham, E., Felger, in European Summer Heatwaves.” Nature, 427(January), 3926–28. R.S., Laureano, P., Mouat, D., Portnov, B.A., Safriel, U., Schmidt, S., Scheffer, M., Hirota, M., Holmgren, M., Van Nes, E.H., and Chapin, F.S. Sciortino, M., Warren, A., and Zohary, D. (2006). “People and Desert.” (2012). “Thresholds for Boreal Biome Transitions.” Proceedings of Global Deserts Outlook, edited by E. Ezcurra, UNEP: Nairobi. the National Academy of Sciences of the United States of America, Selby, J., and Hoffmann, C. (2014). “Beyond ‘Scarcity,’ ‘State Failure,’ 109(52), 21384–89. and ‘Under-development’: Rethinking Water, Climate Change and Scheffran, J., Marmer, E., and Sow, P. (2011). “Migration as a Resource Conflict in the New Sudans.” Global Environmental Change, Article for Resilience and Innovation in Climate Adaptation: Social Networks in Press: Feb 2014. and Co-development in Northwest Africa.” Research Group Climate Selvaraju, R. (2013). “Implications of Climate Change for Agriculture Change and Security, University of Hamburg. and Food Security in the Western Asia and Northern Africa Region.” Scheiter, S., and Higgins, S. (2009). “Impacts of Climate Change on the Climate Change and Food Security in West Asia and North Africa, Vegetation of Africa: An Adaptive Dynamic Vegetation Modelling edited by M. V. K. S. et Al., Springer Science and Business Media: Approach.” Global Change Biology, 15(9), 2224–46. Dordrecht, pp 27-51. Schepaschenko, D.G., Mukhortova, L. V., Shvidenko, A.Z., and Vedrova, Semenov, M.A., Mitchell, R.A.C., Whitmore, A.P., Hawkesford, M.J., E.F. (2013). “The Pool of Organic Carbon in the Soils of Russia.” Parry, M.A.J., and Shewry, P.R. (2012). “Shortcomings in Wheat Yield Eurasian Soil Science, 46(2), 107–16. Predictions.” Nature Climate Change, 2(6), 380–82. Schewe, J., Heinke, J., Gerten, D., Haddeland, I., Arnell, N.W., Clark, Semenza, J., Suk, J.E., Estevez, V., Ebi, K.L., and Lindgren, E. (2012). D.B., Dankers, R., Eisner, S., Fekete, B., Colón-González, F.J., Gosling, “Mapping Climate Change Vulnerabilities to Infectious Diseases in S.N., Kim, H., Liu, X., Masaki, Y., Portmann, F.T., Satoh, Y., Stacke, Europe.” Environmental Health Perspectives, 120(3), 385–92. T., Tang, Q., Wada, Y., Wisser, D., Albrecht, T., Frieler, K., Piontek, Seneviratne, S.I., Donat, M.G., Mueller, B,. and Alexander, L. V. (2014). F., Warszawski, L., and Kabat, P. (2013). “Multi-model Assessment of “No Pause in the Increase of Hot Temperature Extremes.” Nature Water Scarcity under Climate Change.” Proceedings of the National Climate Change, 4(3), 161–63. Academy of Sciences of the United States of America, 104(51), 1–6. Seneviratne, S.I., Nicholls, I., Easterling, D., Goodess, C.M., Kanae, S., Schilling, J., Freier, K.P., Hertig, E., and Scheffran, J. (2012). “Climate Kossin, J., Luo, Y., Marengo, J., McInnes, K., Rahimi, M., Reichstein, Change, Vulnerability, and Adaptation in North Africa with Focus M., Sorteberg, A., Vera, C., and Zhang, X. (2012). “Changes in Climate on Morocco.” Agriculture, Ecosystems & Environment, 156, 12–26. Extremes and their Impacts on the Natural Physical Environment.” Schleussner, C.F., Frieler, K., Meinshausen, M., Yin, J., and Levermann, Managing the Risks of Extreme Events and Disasters to Advance A. (2011). “Emulating Atlantic Overturning Strength for Low Emission Climate Change Adaptation. A Special Report of Working Groups I Scenarios: Consequences for Sea-level Rise along the North American and II of the Intergovernmental Panel on Climate Change (IPCC)., East Coast.” Earth System Dynamics, 2(2), 191–200. edited by C. . Field, V. Barros, T. F. Stocker, D. Qin, D. J. Dokken, K. Schloss, C.A., Nunez, T.A., and Lawler, J.J. (2012). “Dispersal Will Limit Abil- L. Ebi, M. D. Mastrandrea, K. J. Mach, K.-G. Plattner, S. K. Allen, M. ity of Mammals to Track Climate Change in the Western Hemisphere.” Tignor, and P. M. Midgley, Cambridge University Press: Cambridge, Proceedings of the National Academy of Sciences, 109(22), 8606–11. UK, and New York. Schlüter, M., Hirsch, D., and Pahl-Wostl, C. (2010). “Coping with Change: Seo, S.N., McCarl, B.A., and Mendelsohn, R. (2010). “From Beef Cattle Responses of the Uzbek Water Management Regime to Socioeconomic to Sheep Under Global Warming? An Analysis of Adaptation by 267 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Livestock Species Choice in South America.” Ecological Economics, Part 2. Future Climate Projections.” Journal of Geophysical Research: 69(12), 2486–94. Atmospheres, 118(6), 2473–93. Seoane, R., and López, P. (2007). “Assessing the Effects of Climate Change Silvestrini, R.A., Soares-Filho, B.S., Nepstad, D., Coe, M., Rodrigues, H., on the Hydrological Regime of the Limay River Basin.” GeoJournal, and Assunção, R. (2011). “Simulating Fire Regimes in the Amazon in 70(4), 251–56. Response to Climate Change and Deforestation.” Ecological Applica- Shaban, A. (2008), “Indicators and Aspects of Hydrological Drought in tions: A publication of the Ecological Society of America, 21(5), 1573–90. Lebanon.” Water Resources Management, 23(10), 1875–91. Simon, L.M., Oliveira, G. de, Barreto, B. de S., Nabout, J.C., Rangel, Shakesby, R.A. (2011). “Post-wildfire Soil Erosion in the Mediterranean: T.F.L.V.B., and Diniz-Filho, J.A.F. (2013). “Effects of Global Climate Review and Future Research Directions.” Earth-Science Reviews, Changes On Geographical Distribution Patterns of Economically 105(3), 71–100. Important Plant Species in Cerrado.” Viçosa-MG, 37(2), 267–74. Shanin, V.N., Komarov, A.S., Mikhailov, A. V., and Bykhovets, S.S. (2011). Simpson, M., Scott, D., and Trotz, U. (2011). “Climate Change’s Impact “Modelling Carbon and Nitrogen Dynamics in Forest Ecosystems of on the Caribbean’s Ability to Sustain Tourism, Natural Assets, and Central Russia under Different Climate Change Scenarios and Forest Livelihoods.” Technical Note IDB-TB-238, Inter-American Develop- Management Regimes.” Ecological Modelling, 222(14), 2262–75. ment Bank: Washington, DC. Sharmina, M., Anderson, K., and Bows-Larkin, A. (2013). “Climate Change Simpson, M.C., Scott, D., Harrison, M., Silver, N., O’Keeffe, E., Sim, R., Har- Regional Review: Russia.” Wiley Interdisciplinary Reviews: Climate rison, S., Taylor, M., Lizcano, G., Rutty, M., Stager, H., Oldham, J., Wil- Change, 4(5), 373–96. son, M., New, M., Clarke, J., Day, O.J., Fields, N., Georges, J., Waithe, R., Shea, K.M., Truckner, R.T., Weber, R.W., and Peden, D.B. (2008). “Climate and McSharry, P. (2010). “Quantification and Magnitude of Losses and Change and Allergic Disease.” The Journal of Allergy and Clinical Damages Resulting from the Impacts of Climate Change: Modeling the Immunology, 122(3), 443–53; quiz 454–5. Transformational Impacts and Costs of Sea-level Rise in the Caribbean.” Sheffield, J., Wood, E.F., and Roderick, M.L. (2012). “Little Change in United Nations Development Programm: Barbados, West Indies. Global Drought over the Past 60 Years.” Nature, 491(7424), 435–38. Sinervo, B., Méndez-de-la-Cruz, F., Miles, D.B., Heulin, B., Bastiaans, E., Shepherd, A., Mitchell, T., Lewis, K., Lenhardt, A., Jones, L., Scott, L., Villagrán-Santa Cruz, M., Lara-Resendiz, R., Martínez-Méndez, N., and Muir-Wood, R. (2013). “The Geography of Poverty, Disasters, and Calderón-Espinosa, M.L., Meza-Lázaro, R.N., Gadsden, H., Avila, L.J., Climate Extremes in 2030.” Overseas Development Institute: London. Morando, M., De la Riva, I.J., Victoriano Sepulveda, P., Rocha, C.F.D., Shi, L., Escobar, M., Joyce, B., and Kostaras, J. (2013). “Strategic Land Use Ibargüengoytía, N., Aguilar Puntriano, C., Massot, M., Lepetz, V., Planning for Climate Change-Driven Water Shortages in El Alto, Bolivia.” Oksanen, T.A., Chapple, D.G., Bauer, A.M., Branch, W.R., Clobert, J., Shkol’nik, I.M., Meleshko, V.P., Efimov, S.V., and Stafeeva, E.N. (2012). and Sites, J.W. (2010). “Erosion of Lizard Diversity by Climate Change “Changes in Climate Extremes on the Territory of Siberia by the Middle and Altered Thermal Niches.” Science (New York, NY), 328(5980), of the 21st Century: An Ensemble Forecast based on the MGO Regional 894–99. Climate Model.” Russian Meteorology and Hydrology, 37(2), 71–84. Singh, B.K., Bardgett, R.D., Smith, P., and Reay, D.S. (2010). “Microor- Shuman, J.K., Shugart, H.H., and O’Halloran, T.L. (2011). “Sensitivity of ganisms and Climate Change: Terrestrial Feedbacks and Mitigation Siberian Larch Forests to Climate Change.” Global Change Biology, Options.” Nature Reviews. Microbiology, 8(11), 779–90. 17(7), 2370–84. Siraj, A. S., Santos-Vega, M., Bouma, M. J., Yadeta, D., Carrascal, D. R., Shvidenko, A., and Nilsson, S. (2002). “Dynamics of Russian Forests and and Pascual, M. (2014). “Altitudinal Changes in Malaria Incidence in the Carbon Budget in 1961–1998: An Assessment based on Long-term Highlands of Ethiopia and Colombia.” Science, 343, 1154-1158 Forest Inventory Data.” Climatic Change, (2), 5–37. Sito-Sucic, D. (2014). “INTERVIEW-World Bank: Floods Could Disrupt Shvidenko, A., and Nilsson, S. (2003). “A Synthesis of the Impact of Balkans’ Power Supplies for at least Six Months.” Reuters UK, (accessed Russian Forests on the Global Carbon Budget for 1961–1998.” Tellus, June 28, 2014). Available at http://uk.reuters.com/article/2014/05/26/ 55B, 391–415. balkans-flood-worldbank-idUKL6N0OC2MP20140526. Shvidenko, A.Z., and Schepaschenko, D.G. (2013). “Climate Change and Skansi, M.D.L.M., Brunet, M., Sigró, J., Aguilar, E., Arevalo Groening, J.A., Wildfires in Russia.” Contemporary Problems of Ecology, 6(7), 683–92. Bentancur, O.J., Castellón Geier, Y.R., Correa Amaya, R.L., Jácome, H., Shvidenko, A.Z., Shchepashchenko, D.G., Vaganov, E.A., Sukhinin, A.I., Malheiros Ramos, A., Oria Rojas, C., Pasten, A.M., Sallons Mitro, S., Maksyutov, S.S., McCallum, I., and Lakyda, I.P. (2012). “Impact of Villaroel Jiménez, C., Martínez, R., Alexander, L.V., and Jones P.D. Wildfire in Russia between 1998–2010 on Ecosystems and the Global (2013). “Warming and Wetting Signals Emerging from Analysis of Carbon Budget,” Doklady Earth Sciences, 441(2), 1678–82. Changes in Climate Extreme Indices over South America.” Global Sieber, J. (2013). “Impacts of, and Adaptation Options to, Extreme Weather and Planetary Change, 100, 295–307. Events and Climate Change concerning Thermal Power Plants.” Cli- Skinner, E. (2011). Gender and Climate Change Overview Report. Institute matic Change, 121(1), 55–66. of Development Studies. Siebert, S., Burke, J., Faures, J.M., Frenken, K., Hoogeveen, J., Döll, P., Skoufias, E. (ed) (2012). The Poverty and Welfare Impacts of Climate and Portmann, F.T. (2010). “Groundwater Use for Irrigation—A Global Change. Quantifying the Effects, Identifying the Adaptation Strategies. Inventory.” Hydrology and Earth System Sciences, 14(10), 1863–80. World Bank: Washington, DC. Siegfried, T., Bernauer, T., Guiennet, R., Sellars, S., Robertson, A.W., Skoufias, E., Rabassa, M., and Olivieri, S. (2011). “The Poverty Impacts Mankin, J., Bauer-Gottwein, P., and Yakovlev, A. (2012). “Will Climate of Climate Change: A Review of the Evidence.” World Bank Policy Change Exacerbate Water Stress in Central Asia?” Climatic Change, Research Working Paper 5622: Washington, DC. 112(3-4), 881–99. Slangen, A.B.A., Katsman, C.A., Wal, R.S.W., Vermeersen, L.L.A., Riva, Sillmann, J., Donat, M.G., Fyfe, J.C., and Zwiers, F.W. (2014). “Observed R.E.M., and van de Wal, R.S.W. (2011). “Towards Regional Projec- and Simulated Temperature Extremes during the Recent Warming tions of Twenty-first Century Sea-level Change based on IPCC SRES Hiatus.” Environmental Research Letters, 9(6), 064023. Scenarios.” Climate Dynamics, 38(5-6), 1191–1209. Sillmann, J., Kharin V.V., Zhang, X., Zwiers, F.W., and Bronaugh, D. (2013a). Slater, H., and Michael, E. (2012). “Predicting the Current and Future “Climate Extreme Indices in the CMIP5 Multi-model Ensemble. Part 1 : Potential Distributions of Lymphatic Filariasis in Africa Using Maxi- Model Evaluation in the Present.” Journal of Geophysical Research: mum Entropy Ecological Niche Modelling.” PLoS ONE, 7(2), e32202. Atmospheres, 118(4), 1716–33. Slay, B., and Hughes, J. (2011). “The Glaciers of Central Asia: A Disappear- Sillmann, J., Kharin, V.V., Zhang, X., Zwiers, F.W., and Bronaugh, D. ing Resource. Chapter 3—Understanding the Repercussions.” United (2013b). “Climate Extremes Indices in the CMIP5 Multi-model Ensemble: Nations Development Programme. 268 B I BLIOGRA P HY Slettebak, R.T. (2012). “Don’t Blame the Weather! Climate-related Natural Sowers, J., Vengosh, A., and Weinthal, E. (2011). “Climate Change, Water Disasters and Civil Conflict.” Journal of Peace Research, 49(1), 163–76. Resources, and the Politics of Adaptation in the Middle East and North Smith, D. and Vivekananda, J. (2012). “Climate Change, Conflict, and Africa.” Climatic Change, 104(3-4), 599–627. Fragility: Getting the Institutions Right.” Climate Change, Human Spiess, A. (2012). “Environmental Degradation, Climate Uncertainties Security and Violent Conflict. Challenges for Social Stability, Hexagon and Human Vulnerabilities.” Climate Change, Human Security and Series on Human and Environmental Security and Peace, edited by Violent Conflict. Challenges for Social Stability, edited by J. Schef- J. Scheffran, M. Brzoska, H. G. Brauch, P. M. Link, and J. Schilling, fran, M. Brzoska, H. G. Brauch, P. M. Link, and J. Schilling, Springer Springer Berlin Heidelberg. Berlin Heidelberg. Smith, J., Mann, R., Haddad, D., Polack, S., Kurylo, E., and Brooker, Squires, V. (2002). “Physics, Mechanics and Processes of Dust and Sand- S. (2013). “Global Atlas of Trachoma.” International Trachoma stroms. Dust and Sandstorms: An Early Warning of Impending Disas- Initiative, (accessed December 13, 2013), Available at http://www ter.” Global Alarm: Dust and Sandstorms from the World´s Drylands, .trachomaatlas.org. edited by Y. Youlin, V. Squires, and L. Qi, The United Nations, Asia Smith, K.R., Woodward, A., Campbell-Lendrum, D., Chadee, D., Honda, Y., RCU of the UNCCD: Bangkok. Liu, Q., Olwoch, J., Revich, B., and Sauerborn, R. (2014). “Human Stafford Smith, M. (2008). “The ‘Desert Syndrome’—Causally-linked Health: Impacts, Adaptation, and Co-Benefits.” Climate Change 2014: Factors that Characterise Outback Australia.” The Rangeland Jour- Impacts, Adaptation, and Vulnerability. Contribution of Working nal, 30(1), 3. Group II to the Fifth Assessment Report of the Intergovernmental Stern, N. (2013). “The Structure of Economic Modeling of the Potential Panel on Climate Change. Chapter 11. Final Draft, edited by U. Impacts of Climate Change: Grafting Gross Underestimation of Risk Confalonieri and A. Haines, IPCC AR5 WGII, Cambridge University onto Already Narrow Science Models.” Journal of Economic Litera- Press: Cambridge. ture, 51(3), 838–59. Snoussi, M., Ouchani, T., Khouakhi, A., and Niang-Diop, I. (2009). Sternberg, T. (2011). “Regional Drought Has a Global Impact.” Nature, “Impacts of Sea-level rise on the Moroccan Coastal Zone: Quantifying 472(7342), 169. Coastal Erosion and Flooding in the Tangier Bay.” Geomorphology, Sternberg, T. (2012). “Chinese Drought, Bread and the Arab Spring.” 107(1–2), 32–40. Applied Geography, 34, 519–24. Snoussi, M., Ouchani, T., and Niazi, S. (2008). “Vulnerability Assessment Sternberg, T. (2013). “Chinese Drought, Wheat, and the Egyptian Uprising: of the Impact of Sea-level Rise and Flooding on the Moroccan Coast: How a Localized Hazard Became Globalized.” The Arab Spring and The Case of the Mediterranean Eastern Zone.” Estuarine, Coastal and Climate Change. A Climate and Security Correlation Series, edited by Shelf Science, 77(2), 206–13. C. E. Werrell and F. Femia, Center for American Progress, Stimson Soares-Filho, B., Silvestrini, R., Nepstad, D., Brando, P., Rodrigues, H., Center: Washington D.C. Alencar, A., Coe, M., Locks, C., Lima, L., Hissa, L., and Stickler, C. Sterzel, T., Lüdeke, M.K.B., Kok, M.J.T., Soysa, I. de, Walther, C., Sietz, (2012). “Forest Fragmentation, Climate Change and Understory Fire D., Lucas, P., and Janssen, P. (2013). “Armed Conflict Distribution in Regimes on the Amazonian Landscapes of the Xingu Headwaters,” Global Drylands Through the Lens of a Typology of Socio-ecological Landscape Ecology, 27(4), 585–98. Vulnerability.” Regional Environmental Change (accepted). Soares-Filho, B.S., Nepstad, D.C., Curran, L.M., Cerqueira, G.C., Gar- Stevenson, S., Fox-Kemper, B., Jochum, M., Neale, R., Deser, C., and cia, R.A., Ramos, C.A., Voll, E., McDonald, A., Lefebvre, P., and Meehl, G. (2012). “Will There Be a Significant Change to El Niño in Schlesinger, P. (2006). “Modelling Conservation in the Amazon Basin.” the Twenty-First Century?” Journal of Climate, 25(6), 2129–45. Nature, 440(7083), 520–23. Stocker, T.F., Qin, D., Plattner, G.K., Alexander, L.V., Allen, S.K., Bindoff, Soja, A.J., Tchebakova, N.M., French, N.H.F., Flannigan, M.D., Shugart, H.H., N.L., Bréon, F.M., Church, J.A., Cubasch, U., Emori, S., Forster, P., Stocks, B.J., Sukhinin, A.I., Parfenova, E.I., Chapin, F.S., and Stack- Friedlingstein, P., Gillett, N., Gregory, J.M., Hartmann, D.L., Jansen, E., house, P.W. (2007). “Climate-induced Boreal Forest Change: Predic- Kirtman, B., Knutti, R., Kumar, K.K., Lemke, P., Marotzke, J., Masson- tions versus Current Observations.” Global and Planetary Change, Delmotte, V., Meehl, G.A., Mokhov, I.I., Piao, S., Ramaswamy, V., 56(3–4), 274–96. Randall, D., Rhein, M., Rojas, M., Sabine, C., Shindell, D., Talley, L.D., Sommer, R., Glazirina, M., Yuldashev, T., Otarov, A., Ibraeva, M., Mar- Vaughan, D.G., and Xie, S.-P. (2013). “Technical Summary.” Climate tynova, L., Bekenov, M., Kholov, B., Ibragimov, N., Kobilov, R., Change 2013: The Physical Science Basis. Contribution of Working Karaev, S., Sultonov, M., Khasanova, F., Esanbekov, M., Mavlyanov, D., Group I to the Fifth Assessment Report of the Intergovernmental Panel Isaev, S., Abdurahimov, S., Ikramov, R., Shezdyukova, L., and de on Climate Change, edited by T. F. Stocker, D. Qin, G.-K. Plattner, M. Pauw, E. (2013). “Impact of Climate Change on Wheat Productivity Tignor, S. K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex, and P. M. in Central Asia,” Agriculture, Ecosystems & Environment, 178, 78–99. Midgley, IPCC AR5 WGI, Cambridge University Press: Cambridge, Somot, S., Sevault, F., Déqué, M., and Crépon, M. (2008). “21st Century UK and New York, NY. Climate Change Scenario for the Mediterranean Using a Coupled Stocks, B.J., Fosberg, M.A., Lynham, T.J., Mearns, L., Wotton, B.M., Yang, Atmosphere–Ocean Regional Climate Model.” Global and Planetary Q., Lawrence, K., Hartley, G.R., Mason, J.A., and Mckenney, D.W. Change, 63(2–3), 112–26. (1998). “Climate Change and Forest Fire Potential in Russian and Soruco, A., Vincent, C., Francou, B., and Gonzalez, J.F. (2009). “Glacier Canadian Boreal Forests.” Climatic Change, 38, 1–13. Decline between 1963 and 2006 in the Cordillera Real, Bolivia.” Geo- Strobl, E. (2012). “The Economic Growth Impact of Natural Disasters in physical Research Letters, 36(3), 1–6. Developing Countries: Evidence from Hurricane Strikes in the Central Sosa-Rodriguez, F.S. (2013). “From Federal to City Mitigation and Adapta- American and Caribbean Regions.” Journal of Development Econom- tion: Climate Change Policy in Mexico City.” Mitigation and Adaptation ics, 97(1), 130–41. Strategies for Global Change, 1–28. Sutton, W., Srivastava, J., and Neumann, J. (2013a). “Looking Beyond Sousa, P.M., Trigo, R.M., Aizpurua, P., Nieto, R., Gimeno, L., and Garcia- the Horizon. How Climate Change Impacts and Adaptation Responses Herrera, R. (2011). “Trends and Extremes of Drought Indices Throughout Will Reshape Agriculture in Eastern Europe and Central Asia.” World the 20th Century in the Mediterranean.” Natural Hazards and Earth Bank: Washington, DC. System Science, 11(1), 33–51. Sutton, W., Srivastava, J., Neumann, J., Droogers, P., and Boehlert, B. Souza, T.V. De, Lorini, M.L., Alves, M.A.S., Cordeiro, P., and Vale, M.M. (2013b). “Reducing the Vulnerability of Uzbekistan’s Agricultural (2011). “Redistribution of Threatened and Endemic Atlantic Forest Systems to Climate Change. Impact Assessment and Adaptation Birds Under Climate Change.” Natureza & Conservação, 9(2), 214–18. Options.” World Bank: Washington DC. 269 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL Sutton, W., Srivastava, J., Neumann, J., Strz, K.M., and Boehlert, B.B. Teixeira, E., Fischer, G., van Velthuizen, H., van Dingenen, R., Dentener, (2013c). “Reducing the Vulnerability of the Former Yugoslav Republic of F., Mills, G., Walter, C., and Ewert, F. (2011). “Limited Potential of Macedonia’s Agricultural Systems to Climate Change,” Washington DC. Crop Management for Mitigating Surface Ozone Impacts on Global Sun, X., Sun, Q., Zhou, X., Li, X., Yang, M., Yu, A. and Geng, F. (2014). Food Supply.” Atmospheric Environment, 45(15), 2569–76. “Heat Wave Impact on Mortality in Pudong New Area, China in 2013.,” Teixeira, E.I., Fischer, G., van Velthuizen, H., Walter, C., and Ewert, F. Science of the Total Environment, 493, 789–94. (2013). “Global Hot-spots of Heat Stress on Agricultural Crops due to Swarup, A., Dankelman, I., and Ahluwalia, K. (2011). “Weathering the Storm: Climate Change.” Agricultural and Forest Meteorology, 170, 206–15. Adolescent Girls and Climate Change.” PLAN International: Surrey. Teixeira, M.G., Costa, M. de C.N., Barreto, F., and Barreto, M.L. (2009). Swedish Government (2007). “Sweden Facing Climate Change. Threats “Dengue: Twenty-five Years Since Remeergence in Brazil.” Cad. Saúde and Opportunities.” Stockholm. Pública, 35(Sup 1), S7–S18. Swinnen, J.F.M., van Herck, K., and Vranken, L. (2010). “Shifting Patterns Tekken, V., Costa, L., and Kropp, J.P. (2009). “Assessing the Regional of Agricultural Production and Productivity in the Former Soviet Union Impacts of Climate Change on Economic Sectors in the Low-lying and Central and Eastern Europe.” The Shifting Patterns of Agricultural Coastal Zone of Mediterranean East Morocco.” Journal of Coastal Production and Productivity Worldwide., The Midwest Agribusiness Research, 2009(56), 272–76. Trade Research and Information Center, Iowa State University: Ames, IA. Tekken, V., Costa, L., and Kropp, J.P. (2013). “Increasing Pressure, Declin- Syvitski, J.P.M., Kettner, A.J., Overeem, I., Hutton, E.W.H., Hannon, M.T., ing Water and Climate Change in Northeastern Morocco.” Journal of Brakenridge, G.R., Day, J., Vörösmarty, C., Saito, Y., Giosan, L., and Coastal Conservation, 17(3), 379–88. Nicholls, R.J. (2009). “Sinking Deltas Due To Human Activities.” Tekken, V., and Kropp, J. (2012). “Climate-Driven or Human-Induced: Indi- Nature Geoscience, 2(10), 681–86. cating Severe Water Scarcity in the Moulouya River Basin (Morocco).” Tacoli, C. (2009). “Crisis or Adaptation? Migration and Climate Change in a Water, 4(4), 959–82. Context of High Mobility.” Environment and Urbanization, 21(2), 513–25. The Economist (2012). “Democracy Index 2012. Democracy at a Standstill.” Tacoli, C. (2011). “Migration and Global Environmental Change. CR2: The The Economist: London, New York, Hong Kong, Geneva. links Between Environmental Change and Migration: A Livelihoods Theisen, O.M. (2012). “Climate Clashes? Weather Variability, Land Pres- Approach.” International Institute for Environment and Develop- sure, and Organized Violence in Kenya, 1989-2004.” Journal of Peace ment: London. Research, 49(1), 81–96. Tacoli, C., Bukhari, B., and Fisher, S. (2013). “Urban Poverty, Food Security Theisen, O.M., Holtermann, H., and Buhaug, H. (2011). “Climate Wars? and Climate Change.” Human Settlements Working Paper No . 37. Assessing the Claim that Drought Breeds Conflict.” International Tamer, A., and Jäger, J. (2010). Environment, Forced Migration and Social Security, 36(3), 79–106. Vulnerability. Springer. Thomas, C.D., Cameron, A., Green, R.E., Bakkenes, M., Beaumont, L.J., Tanner, T., and Mitchell, T. (2009), “Entrenchment or Enhancement: Collingham, Y.C., Erasmus, B.F.N., De Siqueira, M.F., Grainger, A., Could Climate Change Adaptation Help to Reduce Chronic Poverty?” Hannah, L., Hughes, L., Huntley, B., Van Jaarsveld, A.S., Midgley, IDS Bulletin, 39(4), 6–15. G.F., Miles, L., Ortega-Huerta, M.A, Peterson, A.T., Phillips, O.L., Tapia-Conyer, R., Méndez-Galván, J.F., and Gallardo-Rincón, H. (2009). and Williams, S.E. (2004). “Extinction Risk From Climate Change.” “The Growing Burden of Dengue in Latin America.” Journal of Clini- Nature, 427(6970), 145–48. cal Virology, 46(S2), S3–S6. Thornton, P.K., Jones, P.G., Ericksen, P.J., and Challinor, A.J. (2011). Tarnocai, C., Canadell, J.G., Schuur, E.A.G., Kuhry, P., Mazhitova, G., “Agriculture and Food Systems in Sub-Saharan Africa in a 4°C+ and Zimov, S. (2009). “Soil Organic Carbon Pools in the Northern World.” Philosophical Transactions. Series A, Mathematical, Physical, Circumpolar Permafrost Region.” Global Biogeochemical Cycles, 23(2). and Engineering Sciences, 369(1934), 117–36. Tavoni, M., Kriegler, E., Aboumahboub, T., Calvin, K., DeMaere, G., Jewell, Thornton, P.K., van de Steeg, J., Notenbaert, A., and Herrero, M. (2009). J., Kober, T., Lucas, P., Luderer, G., McCollum, D., Marangoni, G., “The Impacts of Climate Change on Livestock and Livestock Systems Riahi, K. and van Vuuren D. (2014). “The Distribution of the Major in Developing Countries: A Review of What We Know and What We Economies’ Effort in the Durban Platform Scenarios.” Climate Change Need To Know.” Agricultural Systems, 101(3), 113–27. Economics, 5(1), 1340009. Thurman, M. (2011). “Natural Disaster Risks in Central Asia: A Synthesis.” Taylor, K.E., Stouffer, R.J., and Meehl, G.A. (2012). “An Overview of CMIP5 United Nations Development Programme and the Experiment Design.” Bulletin of the American Meteorological Thurner, M., Beer, C., Santoro, M., Carvalhais, N., Wutzler, T., Society, 93(4), 485–98. Schepaschenko, D., Shvidenko, A., Kompter, E., Ahrens, B., Levick, Taylor, G.T., Muller-Karger, F.E., Thunell, R.C., Scranton, M.I., Astor, Y., S.R., and Schmullius, C. (2013). “Carbon Stock and Density of Northern Varela, R., Ghinaglia, L.T., Lorenzoni, L., Fanning, K.A., Hameed, Boreal and Temperate Forests.” Global Ecology and Biogeography, S. and Doherty, O. (2012b). “Ecosystem Responses in the Southern 23(3), 297–310. Caribbean Sea to Global Climate Change.” Proceedings of the National Timmermann, A., McGregor, S., and Jin, F.F. (2010). “Wind Effects on Past Academy of Sciences of the United States of America, 109(47), 19315–20. and Future Regional Sea Level Trends in the Southern Indo-Pacific.” Taylor, M. A., Whyte, F.S., Stephenson, T.S., and Campbell, J.D. (2013). Journal of Climate, 23(16), 4429–37. “Why Dry? Investigating the Future Evolution of the Caribbean Low Le Tissier, M. (2013). “Identification and Implementation of Adaptation Level Jet to Explain Projected Caribbean Drying,” International Journal Response Measures in the Drini—Mati River Deltas. Project Synthesis of Climatology, 33(3), 784–92. Report.” United Nations Development Programme: Tirana. Tchebakova, N.M., Parfenova, E., and Soja, A.J. (2009). “The Effects of Törnros, T., and Menzel, L. (2014). “Addressing Drought Conditions Under Climate, Permafrost and Fire on Vegetation Change in Siberia in a Current and Future Climates in the Jordan River Region.” Hydrology Changing Climate.” Environmental Research Letters, 4(4), 045013. and Earth System Sciences, 18(1), 305–18. Tchebakova, N.M., and Parfenova, E.I. (2012). “The 21st Century Climate Torres, R., Pablo Jayat, J., and Pacheco, S. (2012). “Modelling Potential Change Effects on the Forests and Primary Conifers in Central Siberia.” Impacts of Climate Change on the Bioclimatic Envelope and Conser- Bosque, 33(3), 253–59. vation of the Maned Wolf (Chrysocyon brachyurus).” Mammalian Tchebakova, N.M., Parfenova, E.I., and Soja, A.J. (2011). “Climate Change Biology—Zeitschrift für Säugetierkunde, 78(1), 41–49. and Climate-induced Hot Spots in Forest Shifts in Central Siberia from Torres, R.R., Lapola, D.M., Marengo, J.A., and Lombardo, M.A. (2012). Observed Data.” Regional Environmental Change, 11(4), 817–27. “Socio-climatic Hotspots in Brazil.” Climatic Change, 115(3-4), 597–609. 270 B I BLIOGRA P HY Tory, K.J., Chand, S.S., McBride, J.L., Ye, H., and Dare, R.A. (2013). UNDP (2012). “GII: Gender Inequality Index.” United Nation Development “Projected Changes in Late-Twenty-First-Century Tropical Cyclone Programme (accessed March 17, 2014). Available at https://data.undp. Frequency in 13 Coupled Climate Models from Phase 5 of the Coupled org/dataset/GII-Gender-Inequality-Index-value/bh77-rzbn. Model Intercomparison Project.” Journal of Climate, 26(24), 9946–59. UNDP (2014). “Climate Change Adaptation and Low Emission Development Toumi, A., Chlif, S., Bettaieb, J., Ben Alaya, N., Boukthir, A., Ahmadi, Z.E., Strategy for Bosnia and Herzegovina.” United Nations Development and Ben Salah, A. (2012). “Temporal Dynamics and Impact of Climate Programme, Ministry of Foreign Trade and Economic Relations of Factors on the Incidence of Zoonotic Cutaneous Leishmaniasis in Bosnia and Herzegovina. Central Tunisia.” PLoS Negl Trop Dis, 6(5), e1633. UNDP RBAS (2009). “Arab Human Development Report 2009. Challenges Trab Nielsen, S. (2010). “Coastal Livelihoods and Climate Change.” Reduc- to Human Security in the Arab Countries.” United Nations Develop- ing Poverty, ProtectingLivelihoods, and Building Assets in a Changing ment Programme Regional Bureau for Arab States: New York, NY. Climate Social Implications of Climate Change for Latin America and UNEP (1992). World Atlas of Desertification. Edward Arnold: London. the Caribbean, edited by D. Verner, The World Bank: Washington DC. UNEP (2007). “Climate Change Hits Hard on Latin America and the Transparency International (2012). “Corruption Perceptions Index 2012.” Caribbean.” Press Releases April 2007, United Nations Environment (Accessed March 17, 2014.) Available at http://www.transparency Programme: Brussels. .org/cpi2012/results. UNEP (2008). “Climate Change in the Caribbean and the Challenge of Trenberth, K. (2011). “Changes in Precipitation with Climate Change.” Adaptation.” UNEP Regional Office for Latin America and the Carib- Climate Research, 47(1), 123–38. bean: Panama City, Panama. Trenberth, K., Dai, A., Schrier, G. Van Der, Jones, P.D., Barichivich, J., UNEP (2013). “The Emissions Gap Report 2013.” Nairobi. Briffa, K.R., and Sheffield, J. (2014). “Global Warming and Changes UNESCO (2010). “UNESCO Science Report 2010. The Current Status of in Drought.” Nature Climate Change, 4(January), 3–8. Science around the World.” UNESCO: Paris. Trenberth, K.E., and Shea, D.J. (2006). “Atlantic Hurricanes and Natural UNESCWA (2007). “The Millennium Development Goals in the Arab Variability in 2005.” Geophysical Research Letters, 33(12), L12704. Region 2007: A Youth Lens.” United Nations Economic and Social Tsimplis, M.N., and Baker, T.F. (2000). “Sea Level Drop in the Mediterranean Commission for Western Asia: Beirut. Sea: An Indicator of Deep Water Salinity and Temperature Changes?” Unger-Shayesteh, K., Vorogushyn, S., Farinotti, D., Gafurov, A., Dueth- Geophysical Research Letters, 27(12), 1731–34. mann, D., Mandychev, A., and Merz, B. (2013). “What Do We Know Tsimplis, M.N.N., Calafat, F.M., Marcos, M., Jordà, G., Gomis, D., Fenoglio- About Past Changes in the Water Cycle of Central Asian Headwaters? Marc, L., Struglia, M. V., Josey, S.A., and Chambers, D.P. (2013). A Review.” Global and Planetary Change, 110, 4–25. “The Effect of the NAO on Sea Level and on Mass Changes in the UN-HABITAT (2013). “Urban Indicators.” United Nations Human Settle- Mediterranean Sea.” Journal of Geophysical Letters, 118(1), 944–52. ments Programme, UrbanInfo Database v2.0 (accessed March 25, Tubiello, F.N., Amthor, J.S., Boote, K.J., Donatelli, M., Easterling, W., Fischer, 2013). Available at http://www.devinfo.info/urbaninfo/. G., Gifford, R.M., Howden, M., Reilly, J., and Rosenzweig, C. (2007). UNHCR (2010). “Convention and Protocol Relating to the Status of Refugees.” “Crop Response to Elevated CO2 and World Food Supply. A Comment UNHCR Communications and Public Information Service: Geneva. on ‘Food for Thought...’ by Long et al. Science 312:1918–1921, 2006.” UNICEF (2008). “Our Climate, Our Children, Our Responsibility. The European Journal of Agronomy, 26(3), 215–23. Implications of Climate Change for the World’s Children.” Climate Tubiello, F.N., Soussana, J.-F., and Howden, M.S. (2007). “Crop and Pasture Change Report, UNICEF UK. Response to Climate Change.” Proceedings of the National Academy UNICEF (2011). “Latin America and the Caribbean.” Humanitarian Action of Sciences of the United States of America, 104(50), 19686–90. for Children (accessed March 25, 2014). Available at http://www. Turley, C., and Boot, K. (2010). “UNEP Emerging Issues: Environmental unicef.org/arabic/hac2011/files/HAC2011_4pager_TACRO.pdf. Consequences of Ocean Acidification: A Threat to Food Security.” UNICEF (2013). “Childinfo. Monitoring the Situation of Children and United Nations Environment Programme. Women.” (accessed June 2, 2014). Available at http://www.childinfo. Turley, C., and Gattuso, J.P. (2012). “Future Biological and Ecosystem org/malnutrition_status.html. Impacts of Ocean Acidification and Their Socioeconomic-Policy Implica- UNISDR (2009). “Disaster Risk Reduction in Central Asia, Building Partner- tions.” Current Opinion in Environmental Sustainability, 4(3), 278–86. ships to Secure Development Gains.” The United Nations International UBC (2011). “Sea Around Us Project.” University of British Columbia and Strategy for Disaster Reduction: Manilla the Pew Environment Group, (accessed December 13, 2013). Available United Nations Serbia, EU, and World Bank. (2014). “Serbia Floods 2014.” Belgrade. at http://seaaroundus.org/lme/12/1.aspx. Unnikrishnan, A.S., and Shankar, D. (2007). “Are Sea-level-rise Trends UK Government Office for Science (2011). “The Future of Food and Farming: along the Coasts of the North Indian Ocean Consistent with Global Challenges and Choices for Global Sustainability.” London. Estimates?” Global Planetary Change, 57, 301–7. UK Government Office for Science (2012). “Reducing Risks of Future UNPFII (2009). “Human Development Reports and Indigenous Peoples. A Disasters: Priorities for Decision Makers.” Foresight Project, UK Gov- Desk Review.” United Nations Permanent Forum on Indigenous Issues. ernment Office for Science: London. UN-WATER (2006). “Water: A Shared Responsibility. The United Nations UN (2013). “Sand and Dust Storms—Fact Sheet.” Joint Analysis and Policy World Water Development Report 2.” World Water Development Unit (JAPU), United Nations Iraq. Report 2, UNESCO, UN-WATER and the United Nations World Water UNCCD (1994). “Elaboration of an International Convention to Combat Assessment Programme (WWAP). Desertification in Countries Experiencing Serious Drought and/or Valderrama-Ardilla, C., Alexander, N., Ferro, C., Cadena, H., Marin, D., Desertification, Particularly in Africa.,” United Nations Convention Holford, T.R., Munstermann, L.E., and Ocampo, C.B. (2010). “Envi- to Combat Desertification, U.N. Doc. A/AC.241/27, 33 I.L.M. 1328. ronmental Risk Factors for the Incidence of American Cutaneous UNCCD Secretariat (2013). “A Stronger UNCCD for a Land-Degradation Leishmaniasis in a Sub-Andean Zone of Colombia.” American Journal Neutral World.” Issue brief, Bonn, Germany: Bonn, Germany. of Tropical Medicine and Hygiene, 82(2), 243–50. UN-DESA (2014). “World Urbanization Prospects. The 2014 Revision, Valdés, A., and Foster, W. (2010). “Reflections on the Role of Agriculture Highlights (ST/ESA/SER.A/352).” The Population Division of the in Pro-Poor Growth.” World Development, 38(10), 1362–74. United Nations Department of Economic and Social Affairs: New York. Vandergert, P., and Newell, J. (2003). “Illegal Logging in the Russian Far UNDP (2007). “Fighting Climate Change: Human Solidarity in a Divided East and Siberia.” International Forestry Review, 5(3), 303–6. World.” Human Development Report 2007/2008, United Nations Vandentorren, S., Bretin, P., Zeghnoun, A, Mandereau-Bruno, L., Croisier, A., Development Programme, New York. Cochet, C., Ribéron, J., Siberan, I., Declercq, B., and Ledrans, M. 271 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL (2006). “August 2003 Heat Wave in France: Risk Factors for Death of Villanoy, C., David, L., Cabrera, O., Atrigenio, M., Siringan, F., Aliño, P., Elderly People Living at Home.” European Journal of Public Health, and Villaluz, M. (2012). “Coral Reef Ecosystems Protect Shore from 16(6), 583–91. High-energy Waves Under Climate Change Scenarios.” Climatic Vanwey, L. (2009), “Social and Distributional Impacts of Biofuel Produc- Change, 112(2), 493–505. tion.” Biofuels: Environmental Consequences and Interactions with Villarini, G., and Vecchi, G. A. (2013). “Projected Increases in North Changing Land Use. Chapter 12, edited by R. W. Howarth and S. Atlantic Tropical Cyclone Intensity from CMIP5 Models.” Journal of Bringezu, Proceedings of the Scientific Committee on Problems of the Climate, 26(10), 3231–40. Environment (SCOPE) International Biofuels Project Rapid Assess- Vivchar, A.V., Moiseenko, K.B., and Pankratova, N.V. (2010). “Estimates ment: Gummersbach. of Carbon Monoxide Emissions from Wildfires in Northern Eurasia for Vásquez-Léon, M. (2009). “Hispanic Farmers and Farmworkers: Social Air Quality Assessment and Climate Modeling.” Izvestiya Atmospheric Networks, Institutional Exclusion, and Climate Vulnerability in and Oceanic Physics, 46(3), 281–293. Southeastern Arizona.” American Anthropologist, 111(3), 289–301. Viviroli, D., Archer, D.R., Buytaert, W., Fowler, H.J., Greenwood, G.B., Vautard, R., Gobiet, A., Jacob, D., Belda, M., Colette, A., Déqué, M., Hamlet, A. F., Huang, Y., Koboltschnig, G., Litaor, M.I., López-Moreno, Fernández, J., García-Díez, M., Goergen, K., Güttler, I., Halenka, J.I., Lorentz, S., Schädler, B., Schreier, H., Schwaiger, K., Vuille, M., and T., Karacostas, T., Katragkou, E., Keuler, K., Kotlarski, S., Mayer, S., Woods, R. (2011). “Climate Change and Mountain Water Resources: Meijgaard, E., Nikulin, G., Patarci´ ˇ c, M., Scinocca, J., Sobolowski, Overview and Recommendations for Research, Management and S., Suklitsch, M., Teichmann, C., Warrach-Sagi, K., Wulfmeyer, V., Policy.” Hydrology and Earth System Sciences, 15(2), 471–504. and Yiou, P. (2013). “The Simulation of European Heat Waves from Van Vliet, M.T.H., Yearsley, J.R., Ludwig, F., Vögele, S., Lettenmaier, D.P., an Ensemble of Regional Climate Models within the EURO-CORDEX and Kabat, P. (2012). “Vulnerability of U.S. and European Electricity Project.” Climate Dynamics, 41(9–10), 2555–75. Supply to Climate Change.” Nature Climate Change, 2(9), 676–81. Vecchi, G. A., and Wittenberg, A.T. (2010). “El Niño and Our Future Climate: Volk, T. (2010), “Agriculture in the Western Balkan Countries.” IAMO— Where Do We Stand?” Wiley Interdisciplinary Reviews: Climate Change. Leibniz-Institut für Agrarentwicklung in Mittel und Osteuropa: Halle, Vengosh, A. (2014). “Salinization and Saline Environments.” Treatise on Germany. Geochemistry, edited by H. D. Holland and K. K. Turekian, Elsevier Vörösmarty, C.J., Bravo de Guenni, L., Wollheim, W.M., Pellerin, B., Ltd: Amsterdam. Bjerklie, D., Cardoso, M., D’Almeida, C., Green P., and Colon, L. Vergara, W., Deeb, A., Leino, I., Kitoh, A., and Escobar, M. (2011). Assess- (2002). “Extreme Rainfall, Vulnerability and Risk: A Continental-scale ment of the Impacts of Climate Change on Mountain Hydrology: Assessment for South America.” Philosophical Transactions of the Development of a Methodology through a Case Study in the Andes Royal Society A: Mathematical, Physical and Engineering Sciences, 371. of Peru. World Bank: Washington, DC. Vörösmarty, C.J., Guenni, L.B. De, Wollheim, W.M., Bjerklie, D., Car- Vergara, W., and Scholz, S.M. (2010). Assessment of the Risk of Amazon doso, M., Almeida, C.D., Green, P., and Pellerin, B. (2013). “Extreme Dieback. World Bank Publications: Washington, DC. Rainfall, Vulnerability and Risk : A Continental-scale Assessment for Vergara, W., Toba, N., Mira-Salama, D., and Deeb, A. (2009). “Latin America South America.” Philosophical Transactions of the Royal Society A: and Caribbean Region Sustainable Development Working Paper 32 Mathematical, Physical and Engineering Sciences, 371, 1–17. Assessing the Potential Consequences of Climate Destabilization in Vos, R., Velasco, M., and Labastida, R.E. de (1999). Economic and Social Latin America.” World Bank: Washington, DC. Effects of El Niño in Ecuador, 1997–1998. Vergara, W., Deeb, A., Valencia, A., Bradley, R., Francou, B., Zarzar, A., De Vries, A. J., Tyrlis, E., Edry, D., Krichak, S.O., Steil, B., and Lelieveld, Grünwaldt, A., and Haeussling, S.M. (2007). “Economic Impacts of J. (2013). “Extreme Precipitation Events in the Middle East: Dynam- Rapid Glacier Retreat in the Andes”. EOS, Transactions American ics of the Active Red Sea Trough.” Journal of Geophysical Research: Geophysical Union, 88(25), 261–264. Atmospheres, 118(13), 7087–7108. Verner, D. (2010). “Reducing Poverty, Protecting Livelihoods, and Building Vuille, M. (2013). “Climate Change and Water Resources in the Tropical Assets in a Changing Climate.” D. Verner, ed., World Bank: Washing- Andes.” Inter-American Development Bank, Environmental Safeguards ton, DC. Unit,Technical Note IDB-TN-515. Verner, D. (2012). “Adaptation to a Changing Climate in the Arab Coun- Vuille, M., Francou, B., Wagnon, P., Juen, I., Kaser, G., Mark, B.G., and tries: A Case for Adaptation Governance and Leadership in Building Bradley, R.S. (2008). “Climate Change and Tropical Andean Glaciers: Climate Resilience.” The World Bank: Washington, DC. Past, Present, and Future.” Earth-Science Reviews, 89(3–4), 79–96. Verner, D., and Breisinger, C. (2013). Economics of Climate Change in the Vuuren, D.P., Stehfest, E., Elzen, M.G.J., Kram, T., Vliet, J., Deetman, Arab World: Case Studies from the Syrian Arab Republic, Tunisia, and S., Isaac, M., Klein Goldewijk, K., Hof, A., Mendoza Beltran, A., the Republic of Yemen. The World Bank: Washington, DC. Oostenrijk, R., and Ruijven, B. (2011). “RCP2.6: Exploring the Possibil- Verstraete, M.M., Brink, A.B., Scholes, R.J., Stafford Smith, M., and ity to Keep Global Mean Temperature Increase Below 2°C.” Climatic Beniston, M. (2008). “Climate Change and Desertification: Where Change, 109(1–2), 95–116. Do We Stand, Where Should We Go?” Global and Planetary Change, Vuuren, D.P. Van, Sala, O.E., and Pereira, H.M. (2006). “The Future of 64(3–4), 105–10. Vascular Plant Diversity Under Four Global Scenarios.” Ecology And Verstraete, M.M., Scholes, R.J., and Smith, M.S. (2009). “Climate and Society, 11(2), 25. Desertification: Looking at an Old Problem Through New Lenses.” Walker, M.D., Wahren, C.H., Hollister, R.D., Henry, G.H.R., Ahlquist, L.E., Alatalo, Frontiers in Ecology and the Environment, 7(8), 421–28. J.M., Bret-Harte, M.S., Calef, M.P., Callaghan, T.V, Carroll, A.B., Epstein, Vicuña, S., Garreaud, R.D., and McPhee, J. (2010). “Climate Change H.E., Jónsdóttir, I.S., Klein, J.A., Magnússon, B., Molau, U., Oberbauer, Impacts on the Hydrology of a Snowmelt-driven Basin in Semiarid S.F., Rewa, S.P., Robinson, C.H., Shaver, G.R., Suding, K.N., Thompson, Chile.” Climatic Change, 105(3–4), 469–88. C.C., Tolvanen, A., Totland, Ø., Turner, P.L., Tweedie, C.E., Webber, P.J. Vicuña, S., Gironás, J., Meza, F.J., Cruzat, M.L., Jelinek, M., Bustos, E., and Wookey, P.A. (2006). “Plant Community Responses to Experimental Poblete, D., and Bambach, N. (2013). “Exploring Possible Connec- Warming Across the Tundra Biome.” Proceedings of the National Academy tions between Hydrological Extreme Events and Climate Change in of Sciences of the United States of America, 103(5), 1342–46. Central South Chile.” Hydrological Sciences Journal, 58(8), 1598–1619. Walker, R., Moore, N.J., Arima, E., Perz, S., Simmons, C., Caldas, M., Vignoli, J.R. (2012). “How is Internal Migration Reshaping Metropolitan Vergara, D., and Bohrer, C. (2009). “Protecting the Amazon with Populations in Latin America? New Methodologies and New Evidence.” Protected Areas.” Proceedings of the National Academy of Sciences Centro Latinoamericano y Caribeño de Demografia, ECLAC. of the United States of America, 106(26), 10582–86. 272 B I BLIOGRA P HY Wang, M., Overland, J.E., and Bond, N. A. (2010). “Climate Projections WHO (2009). “Protecting Health from Climate Change: Connecting Science, for Selected Large Marine Ecosystems.” Journal of Marine Systems, Policy, and People.” World Health Organisation: Geneva. 79(3–4), 258–66. WHO (2013a). “World Malaria Report 2013.” World Health Organization, Wang, R., and Wu, L. (2013). “Climate Changes of Atlantic Tropical Cyclone Geneva. Formation Derived from Twentieth-Century Reanalysis.” Journal of WHO (2013b). “Global Health Observatory: Cholera.” World Health Orga- Climate, 26(22), 8995–9005. nization (accessed December 13, 2013). Available at http://www.who Wang, X., Otto, I.M., and Lu, Y. (2013). “How Physical and Social Fac- .int/gho/epidemic_diseases/cholera/en/index.html. tors Affect Village-level Irrigation: An Institutional Analysis of Water WHO (2014). “Essential Leishmaniasis Maps: Visceral and Mucocutaneous Governance in Northern China.” Agricultural Water Management, Leishmaniasis.” World Health Organization (accessed March 17, 2014). 19, 110–18. Available at http://www.who.int/leishmaniasis/leishmaniasis_maps/en/. Ward, D.M. (2013). “The Effect of Weather on Grid Systems and the Reli- WHO and EMRO (2009). “The Work of WHO in the Eastern Mediterra- ability of Electricity Supply.” Climatic Change, 121(1), 103–13. nean Region: Annual Report of the Regional Director—1 January to Warner, K. (2010). “Global Environmental Change and Migration: Gov- 31 December 2008.” World Health Organization, Geneva. ernance Challenges.” Global Environmental Change, 20(3), 402–13. WHO/WMO (2012). “Atlas of Health and Climate.” World Health Organiza- Warner, K., Dun, O., and Stal, M. (2008). “Field Observations and Empirical tion and World Meteorological Organization: Geneva. Research.” Forced Migration Review. Climate Change and Displace- Wilhite, D., Wood, D., and Meyer, S. (1987). “Climate-related Impacts in ment, (31), 13–16. the United States during the 1982–83 El Nino.” M. Glantz, R. Katz, Warner, K., Hamza, M.A., Oliver-Smith, A., Renaud, F., and Julca, A. and M. Krenz (eds.), Climate Crisis: The Societal Impacts Associated (2010), “Climate Change, Environmental Degradation and Migration.” with the 1982-83 Worldwide Climate Anomalies, pp. 75–78. UNEP, Natural Hazards, 55, 689–715. Nairobi; and NCAR, Boulder, Colorado, 1987, UNEP: Nairobi. Warren, R., VanDerWal, J., Price, J., Welbergen, J.A., Atkinson, I., Ramirez- Williams, C. (2010). “An Investigation of the Applicability of Hydroinfor- Villegas, J., Osborn, T.J., Jarvis, A., Shoo, L.P., Williams, S.E., and matics in Assessing the Impact of Flooding on Castries, St. Lucia.” Lowe, J. (2013). “Quantifying the Benefit of Early Climate Change University of the West Indies, Barbados. Mitigation in Avoiding Biodiversity Loss.” Nature Climate Change, Wilson, S.K., Graham, N.J., Pratchett, M.S., Jones, G.P., and Polunin, V.C. 3(7), 678–82. (2006), “Multiple Disturbances and the Global Degradation of Coral Warszawaski, L. (2013). “Research Design of the Intersectoral Impact Model Reefs: Are Reef Fishes At Risk Or Resilient?” Global Change Biology, Intercomparison Project (ISI-MIP).” Proc. Natl. Acad. Sci. USA, in prep. 12(11), 2220–34. Warszawaski, L., Friend, A., Ostberg, S., Frieler, K., Lucht, W., Scha- Winchester, P. (2000). “Cyclone Mitigation, Resource Allocation, and Post- phoff, S., Beerling, D., Cadule, P., Ciais, P., Clark, D.B., Kahana, R., disaster Reconstruction in South India: Lessons from Two Decades of Ito, A., Keribin, R., Kleidon, A., Lomas, M., Nishina, K., Pavlick, R., Research.” Disasters, 24(1), 18–37. Rademacher, T.T., Buechner, M., Piontek, F., Schewe, J., Serdeczny, O., Wisner B., Blaikie P., Cannon T. and Davis I. (2004), At Risk: Natural and Schellnhuber, H.J. (2013). “A Multi-model Analysis of Risk of Hazards, People’ s Vulnerability, and Disasters. Second edition. Hyogo Ecosystem Shifts Under Climate Change,” Environmental Research Framework for Action 2005, UNDP. Letters, 8(4), 044018. Wisner, B., Gaillard, J. and Kelman, I. (2012). The Routledge Handbook of Wassmann, R., Jagadish, S.V.K., Heuer, S., Ismail, A., Redona, E., Ser- Hazards and Disaster Risk Reduction. Routledge: London. raj, R., Singh, R.K., Howell, G., Pathak, H., and Sumfleth, K. (2009). Wittenberg, A.T. (2009). “Are Historical Records Sufficient To Constrain “Climate Change Affecting Rice Production: The Physiological and ENSO Simulations?” Geophysical Research Letters, 36(12), L12702. Agronomic Basis for Possible Adaptation Strategies.” Advances in Wittmann, A.C., and Pörtner, H.O. (2013). “Sensitivities of Extant Animal Agronomy, Vol 101, edited by D. L. Sparks, Elsevier Academic Press. Taxa to Ocean Acidification.” Nature Climate Change, 3(11), 995–1001. Webster, P.J., Holland, G.J., Curry, J.A., and Chang, H.R. (2005). “Changes WMO (2007). “Climate Change and Desertification.” World Meteorological in Tropical Cyclone Number, Duration, and Intensity in a Warming Organization, Geneva, Switzerland. Environment.” Science (New York, N.Y.), 309(5742), 1844–46. WMO (2014). “The State of Greenhouse Gases in the Atmosphere Based on Welch, J.R., Vincent, J.R., Auffhammerc, M., Moyae, P.F., Dobermannf, Global Observations through 2013.” Greenhouse Gas Bulletin No. 10, A., and Dawe, D. (2010). “Rice Yields in Tropical/Subtropical Asia World Meteorological Organisation: Geneva Exhibit Large but Opposing Sensitivities to Minimum and Maximum Wodon, Q., Burger, N., Grant, A., Joseph, G., Liverani, A., and Olesya, Temperatures,” Proceedings of the National Academy of Sciences of T. (2014). “Climate Change, Extreme Weather Events, and Migra- the United States of America, 107(33), 14562–67. tion: Review of the Literature for Five Arab Countries.” People on Werner, A.D., and Simmons, C.T. (2009). “Impact of Sea-Level Rise on Sea the Move in a Changing Climate. The Regional Impact of Environ- Water Intrusion in Coastal Aquifers.” Ground Water, 47(2), 197–204. mental Change on Migration. Global Migration Issues 2, edited by Werz, M., and Conley, L. (2012). “Climate Change, Migration, and Conflict E. Piguet and F. Laczko, International Organization of Migration, in Northwest Africa. Rising Dangers and Policy Options Across the Springer: Dordrecht. Arc of Tension.” Center for American Progress, Heinrich Böll Stiftung. Wodon, Q., Liverani, A., Joseph, G., and Bougnoux, N. (2013). “Climate Wessels, K.J., Prince, S.D., Malherbe, J., Small, J., Frost, P.E., and VanZyl, D. Change and Migration. Evidence Fraom the Middle East and North (2007). “Can Human-induced Land Degradation Be Distinguished Africa.” The World Bank: Washington, DC. from the Effects of Rainfall Variability? A Case Study in South Africa.” Woodruff, J.D., Irish, J.L., and Camargo, S.J. (2013). “Coastal Flooding Journal of Arid Environments, 68(2), 271–97. by Tropical Cyclones and Sea-level Rise.” Nature, 504(7478), 44–52. Westra, S., Alexander, L. V., and Zwiers, F.W. (2013). “Global Increasing Wöppelmann, G., Le Cozannet, G., de Michele, M., Raucoules, D., Trends in Annual Maximum Daily Precipitation.” Journal of Climate, Cazenave, A., Garcin, M., Hanson, S., Marcos, M., and Santamaría- 26(11), 3904–18. Gómez, A. (2013). “Is Land Subsidence Increasing the Exposure to White, J.W., Hoogenboom, G., Kimball, B.A., and Wall, G.W. (2011). Sea-level Rise in Alexandria, Egypt?” Geophysical Research Letters, “Methodologies for Simulating Impacts of Climate Change on Crop 40(12), 2953–57. Production.” Field Crops Research, 124(3), 357–68. Wöppelmann, G., and Marcos, M. (2012). “Coastal Sea-level Rise in WHO (2002). “The World Health Report 2002. Reducing Risks, Promoting Southern Europe and the Nonclimate Contribution of Vertical Land Healthy Life.” World Health Organization, Geneva. Motion.” Journal of Geophysical Research, 117(C1), C01007. 273 TUR N DO W N THE HE AT: CONF R ONT I N G T HE NE W CLI MATE NO R MAL World Bank (2008). “Adaptation to the Impact of Rapid Glacier Retreat in World Bank (2013n). “Electricity Production from Oil Sources (% of Total).” the Tropical Andes.” The World Bank, Report No: 39172-LAC. World Development Indicators (accessed March 25, 2014). Available World Bank (2010a). “World Development Report 2010—Development and at http://data.worldbank.org/indicator/EG.ELC.PETR.ZS] Climate Change.” World Bank Publications: Washington DC. World Bank (2013o). “Electricity Production from Nuclear Sources (% of World Bank (2010b). “The Former Yugoslav Republic of Macedonia. Climate Total).” World Development Indicators (accessed March 25, 2014). Change and Agriculture—Country Note.” World Bank: Washington, DC. Available at http://data.worldbank.org/indicator/EG.ELC.NUCL.ZS. World Bank (2011a). “Gender and Climate Change: Three Things You World Bank (2013p). “Electricity production from Hydroelectric Sources (% Should Know.” World Bank: Washington, DC. of total).” World Development Indicators (accessed March 25, 2014). World Bank (2011b). “North African Coastal Cities Address Natural Disasters Available at http://data.worldbank.org/indicator/EG.ELC.HYRO.ZS. and Climate Change. Climate Change Adaptation and Natural Disaster World Bank (2013q). “Electricity Production from Coal Sources (% of Preparedness in the Coastal Cities of North Africa: Summary of the total).” World Development Indicators (accessed March 25, 2014). Regional Study.” Washington, DC. Available at http://data.worldbank.org/indicator/EG.ELC.COAL.ZS. World Bank (2011c). “Albania—Climate Change and Agriculture. Country World Bank (2013r). “Kyrgyz Republic—Overview of Climate Change Note,” World Bank: Washington, DC. Activities.” World Bank: Washington, DC. World Bank (2011d). “Rising Food and Energy Prices in Europe and Cen- World Bank (2013s). “GDP Growth (Annual %).” World Development tral Asia Rising Food and Energy Prices in Europe and Central Asia.” Indicators (accessed December 18, 2013). Available at http://data Washington, DC. .worldbank.org/indicator/NY.GDP.MKTP.KD.ZG. World Bank (2012a). Turn Down the Heat: Why a 4°C Warmer World Must World Bank (2013t). “World Bank Poverty and Inequality Database.” be Avoided. World Bank: Washington, DC. World DataBank (accessed March 25, 2014). Available at http:// World Bank (2012b). “Food Prices, Nutrition, and the Millennium databank.worldbank.org/data/views/variableselection/selectvariables Development Goals.” Global Monitoring Report 2012, World Bank: .aspx?source=poverty-and-inequality-database. Washington, DC. World Bank (2013u). “Tajikistan—Overview of Climate Change Activities.” World Bank (2013). Turn Down the Heat: Climate Extremes, Regional World Bank: Washington, DC. Impacts, and the Case for Resilience. World Bank: Washington, DC. World Bank (2013v). “Kazakhstan—Overview of Climate Change Activi- World Bank (2013a). “Inclusion Matters. The Foundation for Shared ties.” World Bank: Washington, DC. Prosperity.” World Bank: Washington, DC. World Bank (2013w). “Turkmenistan—Overview of Climate Change Activi- World Bank (2013b). “World Development Indicators.” World Develop- ties.” World Bank: Washington, DC. ment Indicators (accessed June 6, 2014). Available at http://data World Bank (2013x). “Uzbekistan—Overview of Climate Change Activi- .worldbank.org/. ties.” World Bank: Washington, DC. World Bank (2013c). “Urban Population (% of Total).” World Develop- World Bank (2013y). “Tajikistan: Autonomous Adaptation to Climate ment Indicators (accessed March 25, 2014). Available at http://data Change: Economic Opportunities and Institutional Constraints for .worldbank.org/indicator/SP.URB.TOTL.IN.ZS. Farming Households.” World Bank: Washington, DC. World Bank (2013d). “Access to Electricity (% of Population).” World World Bank (2013z). “Sustainable Energy for All: Global Tracking Frame- Development Indicators (accessed March 25, 2014). Available at http:// work Report.” World Bank, United Nations and the International data.worldbank.org/indicator/EG.ELC.ACCS.ZS. Energy Agency: Washington DC World Bank (2013e). “Oil Rents (% of GDP).” World Development Indica- World Bank (2014). “Indigenous Peoples in Latin America. Political, Economic, tors (accessed May 19, 2014). Available at http://data.worldbank.org/ and Social Overview at the End of the Second International Decade of indicator/NY.GDP.PETR.RT.ZS. the World’s Indigenous Peoples.” Draft submitted for discussion at the World Bank (2013f). “Natural Gas Rents (% of GDP).” World Develop- World Conference on Indigenous Peoples, World Bank: Washington DC ment Indicators (accessed May 19, 2014). Available at http://data World Bank Data (2010). “Population Projection Tables by Country and .worldbank.org/indicator/NY.GDP.NGAS.RT.ZS. Group.” HNP Stats (accessed March 24, 2014). Available at http:// World Bank (2013g). “Employment in Agriculture (% of Total Employ- web.worldbank.org/WBSITE/EXTERNAL/TOPICS/EXTHEALTHNU- ment).” World Development Indicators, Washington, DC (accessed TRITIONANDPOPULATION/EXTDATASTATISTICSHNP/EXTHNPST December 19, 2013). Available at http://data.worldbank.org/indicator/ ATS/0,,contentMDK:21737699~menuPK:3385623~pagePK:641684 SL.AGR.EMPL.ZS?display=map. 45~piPK:64168309~theSitePK:3237118,00.html. World Bank (2013h). “Population Growth (Annual %).” World Development World Bank and GFDRR (2013). “Building Resilience. Integrating Climate Indicators, Washington, DC (accessed December 19, 2013). Available and Disaster Risk into Development.” World Bank and GFDRR: at http://data.worldbank.org/indicator/SP.POP.GROW. Washington, DC. World Bank (2013i). “Unemployment, Total (% of Total Labor Force).” World Bank, UNISDR and CAREC (2012). “Central Asia and Caucasus World Development Indicators, Washington, DC (accessed Decem- Disaster Risk Management Initiative (CAC DRMI). Risk Assessment ber 19, 2013). Available at http://data.worldbank.org/indicator/ for Central Asia and Caucasus—Desk Study Review.” The World Bank, SL.UEM.TOTL.ZS?display=map. The United Nations International Strategy for Disatarer Reduction World Bank (2013j). “Poverty Headcount Ratio at $2 a Day (PPP) (% (UNISDR) and Central Asia Regional Economic Cooperation (CAREC): of Population).” World Development Indicators, Washington, DC Washington DC, Geneva and Manilla (accessed December 19, 2013). Available at http://data.worldbank World Travel and Tourism Council (2013). “Travel & Tourism. Economic .org/indicator/SI.POV.2DAY. Impact 2012: Caribbean.” World Travel and Tourism Council: London. World Bank (2013k). “Access to Electricity (% of Population).” World WorleyParsons (2012). “Climate Vulnerability, Risk and Adaptation Development Indicators (accessed December 11, 2013). Available at Assessments—Helping Countries Prepare an Effective Power Sector http://data.worldbank.org/indicator/EG.ELC.ACCS.ZS]. Response: Focus on Turkmenistan.” International Fund for Saving World Bank (2013l). “Electric Power Consumption (kWh per capita).” the Aral Sea (IFAS): Tashkent, Uzbekistan. World Development Indicators (accessed December 11, 2013). Avail- Wright B.D. (2011). “The Economics of Grain Price Volatility.” Applied able at http://data.worldbank.org/indicator/EG.USE.ELEC.KH.PC. Economic Perspectives and Policy, 33(1), 32–58. World Bank (2013m). “Electricity Production from Natural Gas Sources (% Wu, J., He, B., Lü, A., Zhou, L., Liu, M., and Zhao, L. (2010). “Quantita- of Total).” World Development Indicators (accessed March 25, 2014). tive Assessment and Spatial Characteristics Analysis of Agricultural Available at http://data.worldbank.org/indicator/EG.ELC.NGAS.ZS. Drought Vulnerability in China.” Natural Hazards, 56(3), 785–801. 274 B I BLIOGRA P HY Xu, D., Li, C., Zhuang, D., and Pan, J. (2011). “Assessment of the Relative Zhang, H., Ouyang, Z., Zheng, H., and Wang, X. (2009a). “Recent Climate Role of Climate Change and Human Activities in Desertification: A Trends on the Northern Slopes of the Tianshan Mountains, Xinjiang, Review.” Journal of Geographical Sciences, 21(5), 926–36. China.” Journal of Mountain Science, 6(3), 255–65. Yechieli, Y., Shalev, E., Wollman, S., Kiro, Y., and Kafri, U. (2010), “Response Zhang, N., Shugart, H.H., and Yan, X. (2009b). “Simulating the Effects of the Mediterranean and Dead Sea Coastal Aquifers to Sea Level of Climate Changes on Eastern Eurasia Forests.” Climatic Change, Variations.” Water Resources Research, 46(12), 1–11. 95(3–4), 341–61. Yin, J. (2012). “Century to Multi-century Sea-level Rise Projections from Zhang, N., Yasunari, T., and Ohta, T. (2011a). “Dynamics of the Larch CMIP5 Models.” Geophysical Research Letters, 39(17). Taiga–Permafrost Coupled System in Siberia under Climate Change.” Yin, J., and Goddard, P.B. (2013). “Oceanic Control of Sea-level Rise Pat- Environmental Research Letters, 6(2), 024003. terns along the East Coast of the United States.” Geophysical Research Zhang, X., Alexander, L., Hegerl, G.C., Jones, P., Tank, A.K., Peterson, Letters, 40(20), 5514–20. T.C., Trewin, B., and Zwiers, F.W. (2011b), “Indices for Monitoring Yin, J., Schlesinger, M.E., and Stouffer, R.J. (2009). “Model Projections Changes in Extremes based on Daily Temperature and Precipitation of Rapid Sea-level Rise on the Northeast Coast of the United States.” Data.” Wiley Interdisciplinary Reviews: Climate Change, 2(6), 851–70. Nature Geoscience, 2(4), 262–66. Zhao, M., and Running, S.W. (2010). “Drought-induced Reduction in Yu, M., Li, Q., Hayes, M.J., Svoboda, M.D., and Heim, R.R. (2013). “Are Global Terrestrial Net Primary Production from 2000 through 2009.” Droughts Becoming More Frequent or Severe in China based on the Science, 329(5994), 940–43. Standardized Precipitation Evapotranspiration Index: 1951–2010?” Ziervogel, G., and Ericksen, P.J. (2010). “Adapting to Climate Change International Journal of Climatology. to Sustain Food Security.” Wiley Interdisciplinary Reviews: Climate Zarhloule, Y., Fekkoul, A., Boughriba, M., Carneiro, J.F., Correia, A., Change, 1(4), 525–40. Kabbabi, A., Rimi, A., and El Houadi, B. (2009). “Climate Change Zomer, R., Trabucco, A., Bossioc, D.A., and Verchota, L.V. (2008). “Climate and Human Activities Impact on the Groundwater of the Eastern Change Mitigation: A Spatial Analysis of Global Land Suitability for Morocco: Case of Triffa Plain and Shallow Coastal Mediterranean Clean Development Mechanism Afforestation and Reforestation.” Aquifer at Saidia.” Innovations in Groundwater Governance in the Agriculture Ecosystems & Environment, 126(1–2), 67–80. MENA Region, Middle East North Africa Seminar Report from World Zullo, J., Pinto, H.S., Assad, E.D., and Ávila, A.M.H. (2011). “Potential for Water Week 2008, Stockholm. Growing Arabica Coffee in the Extreme South of Brazil in a Warmer Zelazowski, P., Malhi, Y., Huntingford, C., Sitch, S., and Fisher, J.B. (2011). World.” Climatic Change, 109(3-4), 535–48. “Changes in the Potential Distribution of Humid Tropical Forests on a Zwiers, F.W., Zhang, X., and Feng, Y. (2011). “Anthropogenic Influence on Warmer Planet.” Philosophical Transactions. Series A, Mathematical, Long Return Period Daily Temperature Extremes at Regional Scales.” Physical, and Engineering Sciences, 369(1934), 137–60. Journal of Climate, 24(3), 881–92. Zeng, N., Yoon, J.H., Marengo, J.A., Subramaniam, A., Nobre, C.A., Mariotti, A., and Neelin, J.D. (2008). “Causes and Impacts of the 2005 Amazon Drought.” Environmental Research Letters, 3(1), 014002. 275