d i s c u s s i o n pa p e r n u m B e r 7 august 2010 d e v e l o p m e n t a n d c l i m at e c h a n g e d i s c u s s i o n pa p e r s 1 56665 d e v e l o p m e n t a n d c l i m a t e c h a n g e Economics of Adaptation to Climate Change -- Ecosystem Services D I S C u S S I O N PA P E r N u M B E r 7 AuGuST 2010 D E V E L O P M E N T A N D C L I M A T E C H A N G E Economics of Adaptation to Climate Change ­ Ecosystem Services By Glenn-Marie Lange with Susmita Dasgupta, Timothy Thomas, Siobhan Murray, Brian Blankespoor, Klas Sander, and Timothy Essam Papers in this series are not formal publications of the World Bank. They are circulated to encourage thought and discussion. The use and citation of this paper should take this into account. The views expressed are those of the authors and should not be attributed to the World Bank. Copies are available from the Environment Department of the World Bank by calling 202-473-3641. © 2010 The International Bank for Reconstruction and Development / THE WORLD BANK 1818 H Street, NW Washington, DC 20433, U.S.A. Telephone: 202-473-1000 Internet: www.worldbank.org/climatechange E-mail: feedback@worldbank.org All rights reserved. August 2010 This volume is a product of the staff of the International Bank for Reconstruction and Development / The World Bank. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgement on the part of the World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. RIGHTS AND PERMISSIONS The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applica- ble law. The International Bank for Reconstruction and Development / The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; telephone 978-750-8400; fax 978-750-4470; Internet: www.copyright.com. Cover image: Eastern Thailand mangroves. © Winnond / Shutterstock Images, LLC. All dollars are U.S. dollars unless otherwise indicated. III Table of ConTenTs 1. Background 1 1.1 cosystemServicesinPreviousGlobalStudiesoftheCostsofAdaptation E 2 1.2 cosystemServicesAddressedinthisReport E 3 2. Wood Fuels and Non-Wood Forest Products 4 2.1 urrentandProjectedUseofWoodFuel C 5 2.2 mpactofClimateChangeonForestNetPrimaryProductivity I 7 2.3 daptationMeasuresforForestry A 8 3. Mangrove Forests and Coastal Protection Services 11 3.1 ssessingtheImpactofClimateChangeonMangroves A 12 3.2 LimitationsoftheApproach 14 3.3 daptationMeasuresforMangroves A 15 4. Conclusion 17 references 18 Tables 1 People relying on traditional biomass for energy, 2004 to 2030 (millions) 4 2 use of wood fuel in 2006 and projections to 2030 6 3 Percentage change in regional timber production to the year 2050 8 4 Area with mangroves by region, 1980 to 2005 (thousandhectares) 11 5 Potential for mangroves in developing countries to migrate in response to sea-level rise 14 IV E COSy STEM SErV I C E S Figures 1 Wood fuels, 2006, Asia (CubicMetersperCapita) 6 2 Wood fuels, 2006, Africa (CubicMetersperCapita) 7 3 Wood fuels, 2006, Latin America and the Caribbean (cubicmeterspercapita) 8 4 Wood-fuel surplus and deficit in East Africa 9 5 Potential for migration of mangroves in response to sea-level rise 15 Boxes 1 Forest income and rural livelihoods 5 2 Estimating storm-surge zones and human resources at risk 13 3 Mangrove afforestation and coastal protection in Bangladesh: The importance of siting 16 1 1. baCkground aspects of the study, while the Netherlands, UK, and Switzerland have agreed to fund the analysis. Climate change is expected to have major negative The overall study has two objectives. The first is to help impacts on developing countries, but current estimates decision makers in developing countries to better under- of the cost of measures to adapt to climate change are stand and assess the risks posed by climate change and incomplete, crude, or entirely absent. A few global stud- to better design strategies to adapt to climate change. ies have been carried out in recent years (Oxfam 2007; This will be achieved by carrying out case studies at the Stern 2006; UNDP 2007; UNFCCC 2007; World country level that identify costs, priorities, and sequenc- Bank 2006). A program to develop national adaptation ing to integrate adaptation measures into national devel- plans--established by UNFCCC--has been imple- opment plans and budgets. The second objective is to mented in about 23 least-developed countries so far. develop a global estimate of adaptation costs to inform Many other studies have been carried out within coun- the international community's efforts, including tries, and for specific sectors such as agriculture or UNFCCC and the Bali Action Plan; to provide access water. But the economics of adaptation to climate to adequate, predictable, and sustainable support; and to change is a new research area, and no agreed methodol- provide new, additional resources to help the most ogy to assess overall costs has yet emerged. In addition, vulnerable developing countries meet adaptation costs. the effects of climate change depend importantly on The study will proceed simultaneously with two parallel national and subnational climate parameters--such as tracks: one at the country level and another at the global temperature variation over a day and by season, and the level based on global and regional modeling. volume and timing of precipitation. These parameters have been developed at the global or broad regional The global track will ensure the availability of an esti- levels, but there is much less confidence in estimates at mate of developing country and regional adaptation the national level. costs to contribute to the discussion on climate change leading up to the Copenhagen conference in late 2009. An understanding of the full array of adaptation options, But the global track cannot identify the country-specific including institutional and policy changes, is crucial to plans and actions that will be needed by each country; prioritize the most effective adaptation strategies. that is the purpose of the country case studies. The Countries need better estimates of the overall budget second objective of the global track is to develop the implications of implementing "climate resilient develop- procedures that will be needed to generate aggregate ment," both to implement their national strategies and adaptation costs based on the country case studies, once to inform discussions concerning possible international those results are available. assistance. In order to develop estimates of the cost of adaptation at the national, regional, and global levels, a The overall study attempts to make a distinction partnership has been formed between the World Bank between development and adaptation. Many develop- and the governments of the Netherlands, the UK, and ing countries are far behind in meeting the Switzerland. The World Bank is leading the technical Millennium Development Goals and need a great deal 2 E COSy STEM SErV I C E S of investment to meet the MDGs even if climate · Biodiversity, though not explicitly identified as a change were not occurring. This has been called the separate ecosystem service in the initial report, has "development deficit." Adaptation costs are defined as been identified as a separate service in subsequent the incremental costs imposed by climate change, in reports that provide more detailed assessments of addition to, but not including, the costs of develop- specific ecosystems, e.g., marine and coastal ecosys- ment. In practice, it is often hard to distinguish the tems (UNEP 2006). two. The best adaptation measures may not always be "add-ons" to conventional development projects and The very broad scope of ecosystem services make them programs. Many practitioners have said that good a part of everything humans use or do, either directly or development is the best form of adaptation (Schelling indirectly. In the ecosystem services approach of the 1992, echoed by the Stern report in 2006). MA, ecosystems are not so much a unique sector like Nevertheless, the study will keep to the distinction agriculture or infrastructure as they are a way of analyz- wherever possible. ing human activities in the context of the global envi- ronment, which makes life possible. The entire earth This paper reports on a study for the global track on and all human activities are part of ecosystems of one ecosystem services. The scope of this report is more sort or another, including human-dominated ecosystems narrow than the other sector reports. As will be such as urban areas. Provisioning services and cultural/ described below, not all ecosystem services are recreational services are used directly, both as private addressed, and it was not possible to identify at the goods and public goods. Regulating and supporting global level a cost of adaptation. Nevertheless, the threat services are used indirectly, as inputs to the production to certain critical ecosystem services is identified, quan- of goods and services consumed by humans. tified to the extent possible, and examples of costs for potential adaptation measures are described. The first study on the global costs of adaptation by the World Bank (2006) was based largely on incremental 1.1 eC osys Tem serviC es in pre v i o u s costs to "climate-proof " investments in infrastructure and global sT udies of T he C osTs o f did not explicitly address ecosystems. UNFCCC (2007) adapTaT ion took a more disaggregated approach with separate sector studies to estimate adaptation costs for infrastructure; The Millennium Ecosystem Assessment (2005) estab- water; agriculture, forestry and fisheries; coastal zones; lished a classification of ecosystems and ecosystem and human health. Tourism and recreation were not services that is now widely used. The overarching cate- explicitly addressed, but part of their adaptation costs gories of ecosystem services include were implicitly included in the infrastructure and coastal zone assessments. Some regional and country studies · Provisioning services are goods that people obtain have considered tourism; for example, OECD (2008) from ecosystems such as food, fuel, fiber, fresh addressed problems and possible adaptation measures for water, and genetic resources. winter tourism and fishing tourism in Europe. · Regulating services include benefits people obtain from the regulation of ecosystem processes, includ- In the background documents for UNFCCC, a separate ing air quality maintenance, protection from natural case study was done for the costs of adaptation for hazards (floods, storms), climate regulation, erosion ecosystems, although these costs were not included in control, regulation of human diseases, and water the final estimate. Ecosystems were not treated in the purification. Millennium Ecosystem Assessment sense, but were · Cultural services include spiritual enrichment, recre- defined very narrowly--the only ecosystem service ation, education, and aesthetic experiences. considered was biodiversity. Adaptation was to be · Supporting services are services necessary for the achieved by increasing the terrestrial and marine area production of all other ecosystem services such as under protection status to the target area established by primary production, production of oxygen, and soil IUCN, 10 percent of global land area. The target area formation. for protection status is the amount considered necessary D E V E L O P M E N T A N D C L I M AT E C H A N G E D I S C u S S I O N PA P E r S 3 under current conditions, before climate change, not the · Provisioning services: ecosystems provide a range of additional area needed for adaptation to climate change. wild products (food, medicines, building materials, The incremental cost needed to adapt to climate change etc.) that are critical for livelihoods. Among all the was not identified. Since the area currently under ecosystems, forests and woodlands are the most protection does not meet the minimum necessary for important, providing wood fuels and non-wood for- biodiversity protection, this estimate combines a "devel- est products (NWFP). The use of wood fuels and opment deficit" as well as an adaptation cost. NWFP, critical for the livelihoods of over 2 billion people in developing countries, has not been Furthermore, it did not define "adequate" adaptation for addressed. The sector study for forestry considers conserving biodiversity under climate change, and it only the largest economic forest product--industrial appears not to be possible to determine what the impact timber--and does not include all the non-industrial of this measure would be on biodiversity. Because of products. such concerns, it was omitted from the global costs of · Regulating services: a number of regulating services adaptation reported in UNFCCC (2007). are not addressed, or only partly addressed, in the sector studies, such as the flood and storm protec- 1.2 eC osys Tem serviC es addre s s e d tion services of wetlands. The sector study for in T his reporT coastal protection from sea-level rise and increased storms and flooding considered built-up and beach The present study of the economics of adaptation to areas and measures to protect them through infra- climate change follows the sector approach, with separate structure construction (e.g., seawalls, dikes) or estimates of adaptation measures for infrastructure, water, beach nourishment. Protection of coastal areas by agriculture, industrial forestry, fisheries, coastal zones, wetlands and coastal forests, which were very human health, and ecosystems. However, using the defi- important in certain areas in the 2004 tsunami in nition of ecosystem services developed by the Asia (Bratz et al. 2006), was not addressed. Millennium Ecosystem Assessment (MA 2005), it is Generally, all the global studies have focused more clear that most of the services are included in the sector on "hard" adaptation measures rather than "soft" studies either implicitly or explicitly. The sectors adaptation measures; for example, building dams addressed in each study take place within ecosystems, and seawalls rather than rehabilitating natural sys- and ecosystem services are critical to their functioning tems like wetlands. even if these services are not identified explicitly. · Cultural services, tourism and recreation: the costs of Provisioning services are included explicitly in water, maintaining coastal tourism infrastructure is implic- agriculture, industrial forestry, and fisheries sector studies. itly part of the infrastructure study and the coastal Regulating and supporting services are used indirectly for study, but there are other aspects of tourism that the production of goods and services in the sector stud- have not been addressed; for example, the impact of ies; such as pollination as an input to agriculture, or coral bleaching and resulting death of coral on dive watershed protection for water supply, hydropower, and and snorkeling tourism. agriculture. Although not addressed explicitly in sector · Biodiversity has not been discussed in any of the studies, projections about the future of these sectors sector studies. implicitly assume some level of these indirect services. Within the limited scope of this report, it is not possi- Because most ecosystem services have already been ble to address all these missing ecosystem services. addressed, explicitly or implicitly, in the other sector However, we will consider two services: studies for this report, they cannot be counted again here without double-counting. There are, however, a · Wood fuels and NWFP number of ecosystem services that have not been · Coastal protection services provided by mangroves addressed, either explicitly or implicitly. 4 2. Wood fuels and non-Wood Forests also provide non-wood forest products (NWFP), which are very important for many house- foresT produCTs holds in developing countries (See Box 1). NWFP comprise a highly diverse group of products, varying by At the global level, more than 2 billion people rely on country, that includes food products (wild game and biomass for energy, mostly in developing countries. This hides, fruits, nuts, honey, vegetables), materials for figure is expected to grow in absolute terms in all construction and household implements (thatch for regions except China (Table 1).1, 2 For many countries roofs, grasses for mats and baskets, wood for cooking in Sub-Saharan Africa, Latin America, and Asia, utensils, etc.), medicines, gums, fodder for livestock, biomass, mainly wood fuel, constitutes a major source of ornamental foliage, and many other products. Some of household energy, mainly used for cooking and heating. these products are used for subsistence and some are In Africa, where less than 10 percent of the population sold in markets, including international markets. in rural areas has access to electricity, 80 percent of households rely on biomass to satisfy energy needs. But What is striking is the paucity of global data for non- even in urban areas where access to electricity is much industrial forest products, especially subsistence use further advanced, many households use wood fuels-- (Seppala et al. 2009; Sunderlin et al. 2005; Arnold et al. mainly charcoal--for cooking. 2003). As recently as 2009, the International Union of Forest Research Organizations published a very detailed report--Adaptation of Forests and People to Climate Change (Seppala et al. 2009)--that had no global figures for NWFP or wood fuels. Table 1. people relying on TradiTional FAO publishes annual estimates of the volume of wood biomass for energy, 2004 To 2030 (millions) fuel and charcoal used in all countries, but no estimate 2004 2030 of the value. There is no global estimate of NWFP. Many of these products are used for subsistence and are Sub-Saharan Africa 575 720 often poorly accounted for, if at all, in national statistics. North Africa 4 5 To the extent that such products are traded internation- India 740 782 ally, the volume and value of such products are recorded. China 480 394 FAO reported over US$2 billion in global trade in Indonesia 156 180 NWFPs in 2002, mainly exports from industrialized rest of Asia 489 561 Latin America & Caribbean 83 85 1 While people in developed countries also use wood fuel and harvest Total 2528 2727 non-timber forest products, they are far less significant for household livelihoods. Given the scarcity of data, we focus on developing coun- Source: iea 2006. tries where populations are much more vulnerable. Note: Traditional biomass includes wood fuel, charcoal, other wood-derived 2 Recent global price developments for energy have demonstrated that these products, crop residues, and animal dung. forecasts are at best indicative and that future global energy price develop- ments may change this picture quite drastically. D E V E L O P M E N T A N D C L I M AT E C H A N G E D I S C u S S I O N PA P E r S 5 box 1. foresT inCome and rural livelihoods A meta-analysis (Vedeld et al. 2004), synthesizing 54 case studies, found that forest environmental income (wood fuel, NWFP, livestock graz- ing/fodder) contributed an average of uS$678 (PPP-adjusted) to rural household income, accounting for 22 percent of household incomes. Dependence on forests was even greater for poorer households; forest incomes accounted for an average of 32 percent of income for poor and very poor households. Wood fuel and wild foods accounted for nearly 75 percent of the value. A more recent study found that in India for- est resources accounted for 40 percent of income for households in the bottom two income quartiles (Narain et al. 2008). It is clear that for poor households, access to forest products is critical for survival. Forest products contribute in several ways to rural livelihoods: · Current consumption: forest products are a regular part of household consumption, either on a daily basis (e.g., wood fuel for cooking) or seasonal (e.g., fruits and vegetables). For many households in rural areas, forest products are essential to keep them from falling into poverty or more extreme poverty. · Safety net: forest products are used to overcome shortfalls or loss of income, e.g., in a time of seasonal unemployment, before crop har- vesting, or household emergencies. · Pathway out of poverty: if income from forest products is sufficient, households may be able to escape poverty. While the income from forest products accrues mostly to rural communities living in or adjacent to forests, they are not the only ones who rely on forests. Many urban households, who purchase these products, rely on wood fuel and charcoal as a relatively inexpensive source of ener- gy. They can be purchased in very small amounts and can be used with simple, inexpensive equipment, as little as a few large stones. Charcoal is especially important because it is easier than wood to transport and to use. In Sub-Saharan African countries, several tens or even hundreds of thousands of mostly poor people rely on the charcoal trade for their livelihoods--either as charcoal producers, traders, transporters, or producers of charcoal stoves. Many urban households also rely on traditional medicines for healthcare. For poor urban households, the loss of these affordable forest products would be a serious hardship. countries (FAO 2005, quoted in (Seppala et al. 2009)). Table 2 shows the amount of wood fuel used by region in But international trade in NWFPs is only a small part 2006, the most recent year for which comprehensive data of total use. Because of the lack of data about NWFP, are available, with projections by FAO to 2030.3 There is the rest of this section is concerned with wood fuels, the considerable variation in the use of wood fuel across impact of climate change, and adaptation measures. regions. Africa has the highest wood-fuel consumption, Some observations about the implications for NWFPs both in absolute numbers and per capita, using 0.63 cubic will be included at the end of this section. meters of wood fuel per capita annually. Even though East Asia and South Asia have the lowest use per capita, 2.1 CurrenT and projeCT ed us e o f this should not disguise the fact that--especially in South W ood fuel Asia--a large share of the population--especially the urban and rural poor--rely on wood fuel to satisfy energy Wood fuel includes charcoal and other wood-derived needs, notably for cooking and heating. Therefore, within fuels. The FAO database includes consumption of each region, use of wood fuel can vary enormously; each wood fuel and charcoal, reported separately, for each region contains countries with very high and very low per country; other wood-derived fuels are not included. capita wood-fuel use (see Figures 1­3). Wood is consumed in the production of charcoal, so to estimate total wood-fuel use, charcoal must be On a global level, demand and supply of wood fuel is converted to its wood-fuel equivalent. The conversion balanced (Seppala et al. 2009), but the global figure hides factor used in this report is 13.9 cubic meters of cut considerable imbalances at the country and subnational wood per ton of charcoal. This represents an average levels. For example, a study of wood-fuel availability in obtained from data for several Sub-Saharan African East Africa (Drigo 2005) shows vast areas with abun- countries. It has to be acknowledged that conversion dance, but also large areas with serious shortfalls in efficiencies can vary significantly by country and region wood fuel (Figure 4). and are probably lowest for Sub-Saharan African countries. 3 Projections for 2050, the target year for the study, were not available. 6 E COSy STEM SErV I C E S Table 2. use of Wood fuel in 2006 and projeCTions To 2030 2006 2030 Woodfuel Population Percapitawood Woodfuel Population Percapitawood Region (millionm3) (millions) fuel(m3/person) (millionm3) (millions) fuel(m3/person) South Asia 383 1,516 0.25 373 2,027 0.18 Southeast Asia 186 564 0.33 113 708 0.16 East Asia 213 1,531 0.14 152 1,654 0.09 Africa 589 940 0.63 1,185 1,513 0.78 South America 241 453 0.53 400 577 0.69 rest of the World 258 1,556 0.17 328 1,788 0.18 Total 1,869 6,559 0.28 2,552 8,266 0.31 Source: fao 2009 provided country-level data on use of fuel wood and charcoal in 2006. broadhead et al. (2001) as quoted in arnold et al. (2003), provided regional projections for charcoal and fuel wood for 2030. World bank (2009) provided population in 2030. Note: Wood-fuel use reported separately for fuel wood and charcoal. authors estimated total wood fuel by converting charcoal to fuel-wood equivalent. Typically, as income grows, the use of wood fuel will continue to rely heavily on wood fuel (FAO 2008; declines, as households substitute fossil fuels, which are International Energy Agency 2006). On a per capita often easier and cleaner to use, for wood fuel. However, basis, wood-fuel use in Africa and Latin America is even with income growth, most developing countries projected to increase substantially by 2030. figure 1. Wood fuels, 2006, asia (cubic meterS per capita) D E V E L O P M E N T A N D C L I M AT E C H A N G E D I S C u S S I O N PA P E r S 7 figure 2. Wood fuels, 2006, afriCa (cubic meterS per capita) 2.2 i mpa CT of C limaTe Change o n Sohngen/Sedjo projections. Sohngen et al. (2001) used fores T neT primary produC Ti v i Ty the BIOME3 ecological model of Haxeltine and Prentice (1996) together with two general circulation There are many factors that will affect the availability of models--UIUC and Hamburg. The projections are wood fuel and NWFP in the future. This case study concerned with future production of industrial timber considers only factors related to climate change that and are based mainly on projections of changes in directly affect forest productivity--temperature, precipi- forest net primary productivity, a measure of the rate of tation and increased atmospheric CO2 concentration-- plant growth. The changes in NPP apply to all forests, and draws on the available literature to project the so these projections provide a reasonable indication, at impact. We use the same projections used by Sedjo the regional level, of the direct impact of climate (2010) for the industrial forest case study--i.e., those change on natural forests that supply wood fuels and developed by Sohngen et al. (2001)--and will not NWFP. repeat the description of the model, other than to note that this model is widely accepted and cited by, for The Sohngen/Sedjo projections indicate NPP will example, Easterling (2007) and Seppala (2009). increase under climate change in all developing regions (low- to mid-latitudes), at least through 2050 (Table 3). Sohngen et al. used a time horizon of 2145 for their Forest productivity is projected to increase from a low of estimates; this was adjusted by Sedjo for 2050, the 4­5 percent in Africa and Asia-Pacific to as much as 22 time horizon of this study. The resulting estimates, percent in India, depending on the model. Little is known reported in Sedjo (2010), are hereafter referred to as about the potential impact of climate change on NWFP 8 E COSy STEM SErV I C E S figure 3. Wood fuels, 2006, laTin ameriCa and The Caribbean (cubic meterS per capita) (Easterling 2007; Seppala 2009). It is not clear that Table 3. perCenTage Change in regional circumstances that increase NPP for trees will also Timber produCTion To The year 2050 increase NWFP. Impacts are likely to be highly site-specific. Region Hamburg UIUC 1995­2050 1995­2050 2.3 a dap TaT ion measures for High-LatitudeForests fores Try North America (1) (2) Europe 6 11 Based on projections of increased forest NPP, it would Former Soviet union 7 3 appear that no adaptation measures are needed for the China 12 11 provision of wood fuels in developing countries. Oceania (3) 13 However, there are ways in which climate change may Low-toMid-LatitudeForests indirectly affect forests that have not been quantified South America 19 10 and work in the opposite direction: in some regions, India 22 14 climate change may result in increased fires, pest Asia-Pacific 10 4 infestation, and disease. Furthermore, climate change Africa 14 5 may trigger population migration and increased pres- Total All Forests 6 5 sure for conversion of forests to agricultural land. None of these factors have been considered in this Source: adapted from sohngen et al.( 2001) by sedjo (2010). The results for the period 1995­2045 from sohngen et al. were straight-line extended to 2050. assessment. D E V E L O P M E N T A N D C L I M AT E C H A N G E D I S C u S S I O N PA P E r S 9 figure 4. Wood-fuel surplus and defiCiT in easT afriCa Supply-demand balance within 5 arcminute cells High deficit Light deficit Light surplus High surplus Source: reproduced from drigo (2005). Within each region, there is also likely to be significant with heavy rainfall and dense forests; others are very variation in the impact of and response to climate arid and covered mostly by savannah woodlands. change. Sub-Saharan Africa, for example, includes Averaging over the region, forest NPP in Africa is vastly different countries. Some have tropical climates projected to increase, but in semi-arid and arid regions, 10 E COSy STEM SErV I C E S productivity could decline substantially (Seppala et al. have little capacity to adapt to changes. Decreases in 2009). rainfall, and increasing occurrence and severity of droughts can have a devastating impact on these Furthermore, the dependence of rural communities on communities and their forests. Adaptation measures for forests also varies among countries and within a country. such communities have not been considered in this While the average contribution to household income report. A recent report on adaptation of forests (IUFRO from forests was estimated at 22 percent (Box 1), 2009) stressed the importance of social rather than communities who make their homes within forests are technical adaptation measures--flexible management entirely dependent on forests for their livelihoods and and new modes of governance. 11 3. mangrove foresTs and have resulted in significant loss of mangroves over recent decades in all regions (Table 4). CoasTal proTeCTion serviCes Mangroves are salt-tolerant forests at the interface Adaptation to climate change will require living with between terrestrial and marine communities in tropical sea-level rise and increased storm surges in many and subtropical regions. They receive water from the coastal areas. Coastal protection from storm surge and ocean with the tides as well as freshwater, which carries flooding is partly provided by built infrastructure, sediment and nutrients from upland rivers. Their aerial addressed by the Coastal sector study (Nicholls e root system must be exposed to air for part of the day. 2010). Mangrove forests are a form of natural infra- Climate change will result in the loss of mangroves if structure that also provide coastal protection in tropical sea-level rise is sufficient to cut off the flow of fresh regions. This section describes the extent of mangrove water and nutrients and drown the roots. forests, how mangroves will be affected by climate change, the human resources at risk due to loss of Historically, mangroves are very dynamic; they have coastal protection from mangroves, and the potential been subject to near continuous disturbance over the for adaptation. past few thousand years, showing considerable capacity to adapt to fluctuations in sea-level rise (Alongi 2008; Mangroves provide many ecosystem services. They Erwin 2009; Gilman et al.). Adaptation depends on the provide goods that contribute to livelihoods (food, availability of suitable land for natural migration, a timber, wood fuel, medicine) and habitat and nurseries for fish and other wildlife. They maintain water quality and protect coral reefs by trapping sediment, nutrients and contaminants; sediment trapping also protects Table 4. area WiTh mangroves by region, coasts from erosion. 1980 To 2005 (thouSaNd hectareS) Mangroves also provide storm protection by reducing Change,1980 the flow of water and absorbing wave energy. The 1980 1990 2000 2005 to2005(%) importance of mangroves was demonstrated by the Africa 3,670 3,428 3,218 3,160 ­14 Asian tsunami in 2004. Coastal areas with good Asia 7,769 6,741 6,163 5,858 ­25 mangrove forests suffered far less damage and loss of Oceania 2,181 2,090 2,012 1,972 ­10 life than adjacent areas without mangroves (Bratz et al. North & 2,951 2,592 2,352 2,263 ­23 Central 2006; Das and Vincent 2009; Forbes and Broadhead America 2007; Kathiseran and Rajendran 2005; Vermaat and South 2,222 2,073 1,996 1,978 ­11 Thampanya 2006). A meta-analysis estimated the mean America annual value of services provided by mangrove forests at Total 18,794 16,925 15,740 15,231 ­19 roughly $400/hectare (Brander et al. 2006). However, Source: fao 2007. high population pressure and economic development 12 E COSy STEM SErV I C E S continued supply of sediment and nutrients from fresh- on the inundation area resulting from storm surges. The water inflows, and a rate of sea-level rise that is not inundation zone for a 1-meter sea level rise and increase greater than the rate at which mangroves can migrate. in frequency and intensity of storms was calculated using data and methods described in Dasgupta et al. 3.1 a ssessing T he impaCT of Cl i m aTe (2009) and summarized in Box 2.4 C hange on mangroves Response of mangroves to SLR. To estimate the impact of There have been a few studies of the impact of climate SLR on mangroves and the potential for adaptation, we change and sea-level rise on mangroves (discussed in use another characteristic in the DIVA database, wetland Alongi 2008). These range from a high of 30 percent of migratory potential (WMP). WMP indicates the poten- all wetlands (Solomon et al. 2007) to more moderate tial for wetlands, including mangroves, to migrate land- figures for specific areas, such as a 10­20 percent loss of ward in response to a 1-meter rise in sea level. The mangroves in Pacific Islands (Gilman et al. 2008; migratory potential is based on a few geophysical char- Gilman et al. 2006). These studies did not quantify the acteristics of the coastline: coastal type, topography, tidal human resources at risk from loss of mangroves, or costs range, and other information when available (e.g., of adaptation measures. whether mangroves are associated with an island or mainland coast), as described in Hoozemans et al. To assess the current level of coastal protection provided (1993). The migratory potential of mangroves depends by mangroves and the potential for mangroves to adapt on a wide range of additional factors that are site- to climate change by 2050, we use information from two specific and highly variable; for mangroves, this includes global GIS databases, DIVA and a database of world such factors as the continued flow of sediment and mangrove cover provided by UNEP-WCMC. The DIVA nutrients from inland stream. Such detailed information database contains information about the length of the was not available on a global scale. world's coastline and various characteristics associated with each coastal segment. For this report, the character- Five possible responses to SLR, or categories of wetland istics of interest include the presence of mangrove migratory potential, were defined for the DIVA wetlands, wetland migratory potential (explained below), database: and the population and GDP for 2000. 1. No, or hardly any change Mangrove coastline. The DIVA database includes infor- 2. A retreat of the coastline, combined with inland mation about the total area of mangroves in each coastal migration of coastal ecosystems segment, but does not indicate the length of the 3. A retreat of the coastline without the possibility of mangroves along the coast. The mangroves may stretch inland migration due to topography (e.g., coastlines along the entire length of the coastal segment or only a with relatively high relief ) small portion. The WCMC database includes polygons 4. A possible retreat of the coastline but increase of showing the exact location, size, and shape of the flooding area behind the coastline ("ponding") mangroves. The WCMC and DIVA databases were 5. Total loss of the coastal ecosystem (Hoozmans et al. combined to estimate the actual length of mangrove 1993). coastline within each DIVA coastal segment. In the DIVA database, no mangroves occur in areas with Human resources at risk. Information from DIVA indi- the most extreme responses, WMP 1 or WMP 5 (other cates the human resources in coastal areas--population wetlands may fall in these categories). If mangroves can and GDP for 2000. No projections were made of popu- migrate, WMP category 2, then they can continue to lation or GDP for 2050 in coastal zones; analysis of provide coastal protection services. Mangroves in these human resources protected by mangroves uses baseline areas may survive in their current location to the extent data. Within a coastal zone, the number of people and GDP that are at risk under climate change and SLR, 4 Sea-level rise may not reach 1 meter by 2050, but this was the smallest where mangroves may provide some protection, depend unit of SLR for which the database can provide figures. D E V E L O P M E N T A N D C L I M AT E C H A N G E D I S C u S S I O N PA P E r S 13 box 2. esTimaTing sTorm-surge zones and human resourCes aT risk Storm surge zones are locations that would be inundated by a given wave height, assuming the SrTM value represents ground elevation and there are no coastal protection measures. In the calculation of storm surges (wave heights or extreme sea levels), we follow the method outlined by Nicholls (2008) where future storm surges are calculated as follows: Future storm surge = S100 + SLr + (uPLIFT * 100 yr ) / 1000 + SuB + S100 * x where: S100 = 1-in-100-year surge height (m) SLr = sea-level rise (1 m) uPLIFT = continental uplift/subsidence in mm/yr SuB = 0.5 m (applies to deltas only) x = 0.1, or increase of 10%, applied only in coastal areas currently prone to cyclone/hurricane. We apply the wave height calculated for the coastline segment closest to a drainage basin outlet to inland areas within that basin. We also included a distance decay factor of 0.3 m per 1 km distance from the coastline, in estimating wave height for inland cells. Source: based on dasgupta et al. (2009). that natural migration or sediment accretion keeps pace there is a potential to maintain coastal protection by with sea-level rise (Alongi 2008). Mangroves in WMP migration of mangroves with a 1-meter sea level rise category 3 cannot migrate, and the human resources (Table 5, Figure 5). These mangrove coasts support a associated with them will lose their protection. significant share of population and economic activity in Mangroves in WMP category 4 are at great risk, but coastal areas, 97 million people (55 percent) and GDP may survive, depending on the effect of flooding behind of US$189 billion (45 percent). Another 31 percent of the coastline. If the flooding is severe enough and mangroves falls into categories 3 and 4, in which persists long enough to seriously disrupt the flow of climate change will seriously compromise the viability freshwater and nutrients to mangroves, the mangroves of mangroves and the coastal protection services they will be severely degraded and may die, putting at risk the provide. population currently protected by them. Category 3 mangroves account for the only 9 percent of Mangroves extend for nearly 37,000 km, accounting for mangrove coastline, but 28 percent of the population 13 percent of coastline in developing regions, ranging and 41 percent of GDP associated with mangroves. from a low of 4 percent of the coastline in the Middle Survival of mangroves and continued coastal protection East and North Africa to a high of 21 percent in is possible, but at risk depending on local conditions. Sub-Saharan Africa (Table 5). The longest extent of Category 2 mangroves account for the remaining 22 mangrove coastline occurs in East Asia and the Pacific percent of mangrove coastline. These mangroves are the and Latin America, which, together, account for about most vulnerable to SLR and are likely to be lost, two-thirds of mangroves worldwide. Under SLR, the increasing exposure to damage from storm surge to 29 inundation zone for mangrove coastal areas is inhabited million people and GDP of US$56 billion. by 176 million people, generating GDP of US$418 billion, with an average per capita income of US$2,381. The vulnerability of mangroves varies a great deal by These human resources, partially protected by region and by country. In the Middle East and North mangroves, are potentially at risk from a 1-meter SLR Africa, the region with the least mangrove coastline, and increased storm surge. only 15 percent of mangroves have the potential to migrate and survive, while 85 percent will be lost or at Across all developing regions, 69 percent of the serious risk. By contrast, in Sub-Saharan Africa, 78 mangrove coastline falls under WMP category 2, where percent of mangroves are in WMP category 2, with 14 E COSy STEM SErV I C E S Table 5. poTenTial for mangroves in developing CounTries To migraTe in response To sea-level rise Resourcesatrisk Totalcoast- Coastlinewith Coastline Mangrove line mangrove with coastlineby Population GDP GDPper WMP (km) (km) mangrove WMP ('000) ('000USD) capita East Asia and 2 53,832 9,240 17% 72% 50,282 138,115,190 2,747 Pacific 3 41,238 3,028 7% 24% 16,369 32,676,600 1,996 4 7,664 538 7% 4% 27,309 107,175,838 3,925 Total 102,734 12,806 12% 100% 93,961 277,967,628 2,958 Latin America 2 61,643 8,671 14% 64% 6,979 27,527,516 3,944 and the Caribbean 3 25,187 2,889 11% 21% 1,631 7,284,711 4,468 4 11,411 1,981 17% 15% 2,998 16,856,432 5,622 Total 98,240 13,541 14% 100% 11,608 51,668,659 4,451 Middle East 2 1,398 131 9% 15% 80 136,035 1,705 and North Africa 3 13,329 717 5% 83% 1,447 9,220,504 6,373 4 7,764 19 0.2% 2% 8,756 41,887,176 4,784 Total 22,491 866 4% 100% 10,283 51,243,715 4,983 South Asia 2 6,885 1,681 24% 69% 33,686 20,764,983 616 3 3,085 390 13% 16% 6,915 4,353,492 630 4 3,832 349 9% 14% 6,542 5,261,091 804 Total 13,803 2,421 18% 100% 47,142 30,379,566 644 Sub-Saharan 2 17,490 5,379 31% 78% 6,181 2,818,601 456 Africa 3 10,252 1,095 11% 16% 2,511 1,970,184 785 4 5,676 416 7% 6% 3,829 1,934,048 505 Total 33,419 6,890 21% 100% 12,520 6,722,833 537 World 2 141,249 25,102 18% 69% 97,207 189,362,325 1,948 3 93,092 8,120 9% 22% 28,872 55,505,491 1,922 4 36,346 3,303 9% 9% 49,435 173,114,585 3,502 Total 270,687 36,525 13% 100% 175,514 417,982,401 2,381 Source: Calculations described in the text. Wmp: Wetland migratory potential, described in the text. great potential for migration. In many regions, the per Factors that overestimate coastal protection: capita GDP is higher in areas where mangroves are likely to be lost, compared to areas where mangroves · The mangrove database shows the extent and shape have the capacity to migrate. of the mangrove area, but does not indicate the sta- tus of the mangroves, e.g., patchiness, health, size 3.2 limi TaTions of T he approaC h of trees. Studies have shown that specific character- istics of mangroves are important for protection There are several major limitations to this approach; from storm surge. For example, if stands are not some lead to an overestimation of the coastal protection dense enough, they provide insufficient resistance to service, while others lead to an underestimation. wave energy, but if the stand is too dense, waves may simply pass over (Bratz et al. 2006). D E V E L O P M E N T A N D C L I M AT E C H A N G E D I S C u S S I O N PA P E r S 15 Many mangrove forests may not survive to 2050, regard- figure 5. poTenTial for migraTion of less of the impact of climate change. For those forests mangroves in response To sea-level that do survive, demographic, economic and other rise factors may block migration, even where the ecological 100% conditions would make it possible. Coastal areas are the 90% most densely populated parts of the globe, with many 80% 70% large, rapidly expanding urban areas; competition for 60% space is fierce. Preserving and cultivating mangroves as a 50% 40% source of coastal defense will require addressing compet- 30% ing land uses, which is not possible in this report. 20% 10% 0% 3 . 3 d a pTaT i o n m e a s u r e s f o r a E. Asia, Pacific Lat. America, Mid. East, Carib. N. Africa South Asia Sub-Saharan Africa World mangroves WMP 4 WMP 3 WMP 2 Source: Table 5. Adaptation measures to maintain the protection services Note: Wmp = wetland migratory potential as defined in the text. of mangroves from the impact of climate change can Wmp 2: inland migration is possible take the form of (a) measures to support migration of Wmp 3: inland migration of mangroves is not possible mangroves where feasible, including afforestation, Wmp 4: inland migration may be possible, but with substantial increased flooding which threatens the viability of mangroves replanting, and rehabilitation of mangroves in appropri- ate areas; (b) substitution of "hard" infrastructure for mangroves; (c) combinations of mangrove forests and coastal embankments; or (d) migration of population and economic activities out of areas subject to storm · The mangrove database from WCMC is for 1997; surge and flooding. We will discuss the opportunities data for 2007 will only be available in 2010. In some for supporting mangrove forests, either through migra- countries there has been significant loss of man- tion or afforestation for use in combination with coastal groves since 1997, so use of 1997 data will tend to embankments. overestimate the current levels of coastal defense. FAO (2007) indicates that globally only 3 percent of Mangroves in areas with the potential for migration may mangroves were lost between 2000 and 2005, so the migrate naturally in response to sea-level rise if they are 1997 data may be reasonably accurate at the regional not blocked by other land uses, and sea-level rise is not level. But the loss of mangroves may be much larger faster than the natural migration rate. In other areas, in some countries. intervention may be necessary. There have been many successful attempts to plant or rehabilitate mangroves in Factors that underestimate coastal protection and Asia and East Africa, including a large-scale effort in resources at risk: many countries affected by the 2004 tsunami (UNEP- WCMC 2006). Reviewing efforts in the Philippines, · Due to lack of data, the analysis does not include Primavera and Esteban (2008) found mixed results. Many some small-island nations in Africa, Asia and the afforestation or restoration and rehabilitation efforts Pacific, and Latin America. failed because of inappropriate species and poor site selec- tion. Mangroves were often planted in lower intertidal or The other major limitation of this approach is that the subtidal zones, where mangroves do not naturally occur, potential for migration is only the first step toward because more suitable land was not available. With understanding whether mangroves will actually migrate competing uses for coastal land, it will be difficult to find or not. Mangroves are already under severe pressure suitable land for mangroves as sea-level rises. from conversion for aquaculture and tourism, overcut- ting, pollution, and other factors. Mangroves have been The costs of afforestation and replanting mangroves can lost in many areas and are severely degraded in others. vary enormously: Primavera and Esteban (2008) report 16 E COSy STEM SErV I C E S average planting costs in the Philippines of over $500/ Another approach to coastal protection combines affor- hectare. This does not include the costs of purchasing estation with mangroves and "hard" infrastructure, land, where necessary. The Ramsar Secretariat, quoted planting mangroves in front of an embankment. in Gilman and Ellison (2007), reported a range of costs Mangroves can be relatively inexpensive to plant, per hectare from US$225 to US$216,000, depending on provide additional benefits, and reduce the necessary the amount of rehabilitation needed. height of the embankment as well as its maintenance costs (Tri et al. 1998). However, it is important to One opportunity is the rehabilitation of abandoned design the forest correctly for specific sites because there aquaculture sites (if they are in areas identified as WMP can be great variation in the protection provided by 2). Over the past few decades, large areas of mangrove forests with different characteristics, e.g., width (Forbes forests, especially in Asia, were converted for aquacul- and Broadhead 2008) and location (Box 3). ture, mainly shrimp farming. Many of these farming operations were abandoned after about five years due to Even in areas with the natural potential for mangrove disease and loss of profitability; the operators moved migration, there will be enormous variation across coun- onto new sites (Barbie, 2009). Abandoned shrimp tries and within countries in the efficiency of using ponds are good sites for restoring mangroves because mangrove forests as "natural infrastructure" to protect they are natural mangrove habitat with all the condi- coastal communities against storms and flooding. Where tions necessary for mangroves to thrive. However, aban- densely settled urban areas compete with mangroves for doned shrimp ponds are highly degraded and cannot be coastal land, it may be more efficient to use built infra- used for any other purpose because the soil has become structure. Mangrove areas that meet both natural condi- very acidic, compacted, and of poor quality (Wolanski tions as well as socioeconomic conditions for migration 2006). Mangroves will not naturally re-colonize these of mangroves, or use in combination with built infra- areas until the land is rehabilitated, and rehabilitation structure can only be identified at the country level. can be expensive. Barbier (2009) reported costs of US$8,812­$9,318 per hectare for rehabilitation, For mangrove areas without the potential for migration, replanting, and maintaining mangrove seedlings. alternative measures will be needed to deal with However, the benefits from restoration were signifi- increased vulnerability to storms and flooding, either cantly higher than the costs. built infrastructure or migration of population. The costs of these measures are beyond the scope of this study. box 3. mangrove afforesTaTion and CoasTal proTeCTion in bangladesh: The imporTanCe of siTing A study to design the optimal combination of mangrove forest size and polder height was carried out for Hatia island in Bangladesh, an island that is often hit by cyclones, including the record 1970 cyclone. The study used simulation modeling to (a) identify the relationship between storm-surge height and forest parameters such as species, density, tree girth and forest width; and (b) based on this information, to determine the necessary forest area for a given height of embankment. The authors derived the function showing the relationship between surge height and forest width up to 600 meters wide for different parts of the island. The study found that storm-surge attenuation varied not only by forest width, but also by location on the island. At the southern end of Hatia island, a mangrove forest 600m wide reduced the surge height by 0.45m, from about 6.20 m to 5.75m. For forest width of 133 m, surge height was reduced by 0.18m. However, no appreciable (>0.1 m) reduction in surge height was obtained by mangroves at the south- eastern or southwestern sides of the island. The results indicate that mangroves be used in combination with "hard" infrastructure, but that for- est site as well as width must be planned carefully because site-specific characteristics greatly influence the extent of storm protection. Source: institute for Water management 2000. 17 4. ConClusion forests from overharvesting, pollution, and clearing for development. Even where mangroves survive the impacts of climate change, they do not provide complete protec- Regarding wood fuels and NWFP, the impact of tion from storms and flooding. The residual impact of climate change at the regional level does not appear to climate change has not been estimated. reduce forest productivity directly, implying that there are no serious adaptation costs. However, as explained Serious gaps in the coverage of ecosystem services in section 2, this conclusion must be viewed with remain, notably with regard to biodiversity. However, it caution because there is great variation among countries is not at all clear from the biological perspective what and within countries in the supply-demand balance of adaptation measures are necessary and effective for wood fuels, and in the degree of dependence on forest biodiversity. Some ecosystems and their unique flora products by different households. It is likely that the and fauna will be lost; others may survive if species can relative surplus that may occur in some countries or migrate. It is possible that designing barrier-free corri- some areas of a country will not compensate for the dors to allow natural migration can help promote adap- declining forest resources in other areas. This is espe- tation for some species.5 The studies of adaptation costs cially true for subsistence use, for which proximity to at the country level can carry this work forward, going forests is essential. There are many case studies of into more detail for the ecosystem services covered in household dependence on forest products, but there is this study (coastal protection from natural infrastructure insufficient information to estimate the dependence at and non-industrial forest products) and addressing the the regional and global levels. This is further compli- missing ecosystem services. cated by the uncertainty regarding the impact of climate change at the subnational level. The oceans have been particularly neglected in adapta- tion studies. The fisheries sector addresses some aspects Regarding coastal protection provided by mangrove of ocean ecosystems, but only those directly related to forests, there is great potential for adaptation by natural fisheries. Oceans play an important and complex role in or assisted migration, and geological evidence shows that regulating many fundamental processes on land and sea. mangroves have responded this way to sea-level rise in Mangrove forests were discussed, but coral reefs, sea- the past. Only about 20 percent of mangroves are likely grass beds, and kelp forests are also important. There is to be lost due to sea-level rise; the rest may migrate if relatively little information about the impact of climate there are no other land uses blocking them. The costs of change on ocean ecosystems and appropriate adaptation intervening to assist this process are not substantial, but measures. Country case studies cannot solve this prob- can vary a great deal. There has been a lot of experience lem, as ocean ecosystems cut across national boundaries with replanting and rehabilitating mangroves, so and vast areas are common property. mistakes that result in failure, common in the past, can be avoided. The major obstacle will be with competing land uses, as well as existing pressure on mangrove 5 Such an approach is being considered in Germany (OECD 2008). 18 referenCes Brander, L., R. Florax, and J. Vermaat. 2006. "The empirics of wetland valuation: A comprehensive summary and meta-analysis of the literature." Agarwala, S., and S. Fankhauser, eds. 2008. Economic Environmental and Resource Economics 33: 223­250. Aspects of Adaptation to Climate Change. Costs, Benefits and Policy Instruments. Paris: OECD. Broadhead, J., J. Bahdon, and A. Whiteman. 2001. "Woodfuel consumption modeling and results." Annex 2 Alongi, D. 2008. "Mangrove forests: Resilience, protec- in "Past Trends and Future Prospects for the Utilization of tion from tsunamis and responses to global climate Wood for Energy." Working Paper No. GFPOS/WP/05, change." Estuarine, Coastal and Shelf Science 76:1­13. Global Forest Products Outlook Study. Rome: FAO. Arnold, Michael, Gunnar Köhlin, Reidar Persson, and Das, S., and J. Vincent. 2009. "Mangroves protected Gillian Shepherd. 2003. "Fuelwood Revisited: What villages and reduced death toll during Indian super Has Changed in the Last Decade?" CIFOR Occasional cyclone." PNAS 106(18):7357­7360. Paper No. 39. Jakarta: Center for International Forestry Research (CIFOR). Downloaded from http://www. Dasgupta, S., B. Laplante, C. Meisner, D. Wheeler, and cifor.cgiar.org/Publications/ J. Yan. 2009. "The impact of sea level rise on developing countries: a comparative analysis." Climatic Change Barbier, E. 2007. "Valuing ecosystem services." Economic 93:379­388. Policy Issue 49:177­229. Drigo, Rudi. 2005. "WISDOM ­ East Africa: Barbier, E., E. Koch, B. Silliman, et al. 2008. "Coastal Woodfuel Integrated Supply/Demand Overview ecosystem-based management with nonlinear ecological Mapping (WISDOM) Methodology." Rome: FAO. functions and values." Science 319(18): 321­323. Easterling, W.E., P.K. Aggarwal, P. Batima, et al. 2007. Barbier, E. 2009. "Valuation of ecosystem services." In "Food, fibre and forest products." In M.L. Parry, O.F. K. McLeod and H. Leslie, eds. Ecosystem-based Canziani, J.P. Palutikof, P.J. van der Linden, and C.E. Management for the Oceans: Applying Resilience Thinking. Hanson, eds. Climate Change 2007: Impacts, Adaptation Washington, DC: Island Press. and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Berry, P. 2007. "Adaptation options on natural ecosys- Panel on Climate Change. Cambridge, UK: Cambridge tems." Unpublished report to the UNFCCC Secretariat. University Press. Braatz, S., S. Fortuna, J. Broadhead, and R. Leslie, eds. Erwin, K. 2009. "Wetlands and global climate change: 2006. "Coastal protection in the aftermath of the Indian the role of wetland restoration in a changing world." Ocean tsunami: What role for forests and trees?" Wetlands Ecology and Management 17:71­84. Proceedings of the Regional Technical Workshop, Khao Lak, Thailand, August 2831, 2006. D E V E L O P M E N T A N D C L I M AT E C H A N G E D I S C u S S I O N PA P E r S 19 FAO. 2007. "The World's Mangroves 1980-2005." FAO Kathiseran, K. and N. Rajendran. 2005. "Coastal Forestry Paper 153. Rome: FAO. mangrove forests mitigated tsunami." Estuarine, Coastal and Shelf Science 65:601­601. FAO. 2008. State of the World's Forests 2008. Rome: FAO. Matthews, E., R. Payne, M. Rohweder, and S. Murray. 2000. Pilot Analysis of Global Ecosystems: Forest FAO. 2009. State of the World's Forests 2009. Society, Ecosystems. Washington: World Resources Institute. Forests and Forestry: Adapting for the Future. Rome: FAO. Millennium Ecosystem Assessment. 2005. Ecosystems and Human Well-Being. Washington, DC: Island Press. Forbes, K., and J. Broadhead. 2007. The role of coastal forests in the mitigation of tsunami impacts. Rome: FAO. Narain, U., S. Gupta, and K. van t'Veld. 2008. "Poverty and resource dependence in India." Ecological Economics Gilman, E., H. Van Lavieren, J. Ellison, et al. 2006. 66: 161­176. "Pacific island mangroves in a changing climate and rising sea." UNEP Regional Seas Report and Studies Nicholls. 2010. "Costs of adaptation to climate change No. 179. Nairobi: UNEP. in coastal zones." Environment Department Paper. Washington, D.C.: World Bank. Gilman, E., and J. Ellison. 2007. "Efficacy of alternative low-cost approaches to mangrove restoration, American OECD. 2008. Economic Aspects of Adaptation to Climate Samoa." Estuaries and Coasts 33(4):641­651. Change. Paris: OECD Gilman, E., J. Ellison, N. Duke, and C. Filed. 2008. Oxfam 2007. "Adapting to Climate Change. What is "Threat to mangroves from climate change and adapta- Needed in Poor Countries and Who Should Pay?" tion options." Aquatic Biology vol 89, no. 2: 237-250. Oxfam Briefing Paper 104. New York, Brussels, Geneva and Washington: Oxfam International. Haxeltine, A., and I.C. Prentice. 1996. "BIOME3: An equilibrium terrestrial biosphere model based on Parry, M.L., O.F. Canziani, J.P. Palutikof, P.J. van der ecophysicological constraints, resource availability, and Linden, and C.E. Hanson, eds. 2007. Climate Change competition among plant functional types." Global 2007: Impacts, Adaptation and Vulnerability. Contribution Biogeochemical Cycle 10(4): 693­709. of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Hoozemans, F.M.J., M. Marchand, and H,A. Cambridge, UK: Cambridge University Press. Pennekamp. 1993. Sea level rise: A global vulnerability assessment. 2nd revised edition. The Hague: Delft Primavera, J., and J. Esteban. 2008. "A review of Hydraulics and Tidal Waters Division, Ministry of mangrove rehabilitation in the Philippines: successes, Transport, Public Works and Water Management. failures and future prospects." Wetlands Ecology and Management 16:345­358. Institute for Water Management (IWM). 2000. Effect of afforestation on storm surge propagation for coastal embank- Schelling, T. 1992. "Some economics of global warm- ment rehabilitation project. Dhaka: IWM. ing." American Economic Review 82(1) 1­14. International Energy Agency (IEA). 2006. World Energy Sedjo, R. 20010 "Adaptation of forest to climate Outlook 2006. Paris: IEA. change." Environment Department Paper, Washington, DC: World Bank. IUFRO. 2009. Adaptation of forests and people to climate change. IUFRO World Series Vol. 22. Vienna, Seppala, R., A. Buck, and P. Katila. 2009. Adaptation of Austria: IUFRO Forests and People to Climate change--A Global Assessment 20 E COSy STEM SErV I C E S Report. IUFRO World Series, volume 22. Vienna: UNEP. 2006. Marine and coastal ecosystems and human International Union of Forest Research Organizations wellbeing: A synthesis report based on the findings of the (IUFRO). Millennium Ecosystem Assessment. Nairobi: UNEP. Sohngen, B., R. Mendelsohn, and R. Sedjo. 2001. "A UNDP. 2007. Human Development Report 2007/08. global model of climate change impacts on timber New York: Palgrave McMillan. markets." J. Agr. Resour. Econ 26(2): 326­343. Available at http://ageconsearch.umn.edu/ UNEP WCMC. 2006. In the front line: shoreline protec- bitstream/31044/1/26020326.pdf. tion and other ecosystem services from mangroves and coral reefs. Cambridge, UK: UNEP-WCMC. Solomon, S., D. Quin, M. Manning, Z. Chen, M. Marquis, K.B. Averyl, M. Tignor, and H. Miller, eds. UNFCCC. 2007. Investment and Financial Flows to 2007. Climate change 2007: The Physical Science Basis. Address Climate Change. Bonn, Germany: Climate Contribution of working Group I to the Fourth Change Secretariat. Assessment Report of the IPCC. Cambridge: Cambridge University Press. Veldeld, P., A. Angelsen, E. Sjaastad, and G. Berg. 2004. "Counting on the environment: forest incomes and the Sunderlin, W., A. Angelson, B. Belcher, P. Burgers, R. rural poor." Environment Department Paper No. 98. Nasi, L. Santoso, and S. Wunder. 2005. "Livelihoods, Washington, DC: World Bank. forests, and conservation in developing countries: An overview." World Development 33(9) 1382­1402. Vermaat, J., and U. Thampanya. 2006. "Coastal mangrove forests mitigated tsunami: A further Stern, N. 2006. Stern Review: Economics of Climate response." Estuarine, Coastal and Shelf Science 69:1­3 Change. Cambridge: Cambridge University Press. Wolanski, E. 2006. "Synthesis of the protective func- Taylor, M., C. Ravilious, and E. Green. 2003. Mangroves tions of coastal forests and trees against natural of East Africa. Cambridge, UK: UNEP World hazards." In Braatz, S., S. Fortuna, J. Broadhead, and R. Conservation Monitoring Center. Leslie, eds. "Coastal protection in the aftermath of the Indian Ocean tsunami: What role for forests and trees?" Tri, N., W. Adger, and P. Kelly. 1998. "Natural resource Proceedings of the Regional Technical Workshop, Khao management in mitigating climate impacts: the example Lak, Thailand, August 2831, 2006. of mangrove restoration in Vietnam." Global Environmental Change 1(8) 49­61. World Bank. 2006. Investment Framework for Clean Energy and Development. Washington DC.: World Bank. World Bank. 2009. World Development Indicators. Washington, DC: World Bank. The World Bank Group 1818 H Street, NW Washington, D.C. 20433 USA Tel: 202-473-1000 Fax: 202-477-6391 Internet: www.worldbank.org/climatechange