THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN LEVERAGING EARTH OBSERVATION DATA TO IDENTIFY INVESTMENT OPPORTUNITIES IN NBS FOR CLIMATE RESILIENCE IN CITIES AND COASTS ACROSS THE WORLD THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN LEVERAGING EARTH OBSERVATION DATA TO IDENTIFY INVESTMENT OPPORTUNITIES IN NBS FOR CLIMATE RESILIENCE IN CITIES AND COASTS ACROSS THE WORLD 2024 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington, DC 20433 TELEPHONE: +1-202-473-1000; INTERNET: www.worldbank.org Some rights reserved. This work is a product of the staff of The World Bank and the Global Facility for Disaster Reduction and Recovery (GFDRR). 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, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. 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 construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. This work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Please cite this work as follows: World Bank. 2024. The Nature-Based Solutions Opportunity Scan: Leveraging Earth Observation Data to Identify Investment Opportunities in NBS for Climate Resilience in Cities and Coasts across the World. © World Bank, Washington, DC. License: CC BY 3.0 IGO. If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. 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 infringements rests solely with you. If you wish to reuse a component of the work, it is your responsibility to determine whether permission is needed for that reuse 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, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org Cover design, layout, and illustrations: Estudio Relativo TABLE OF CONTENTS ACKNOWLEDGMENTS 7 ABBREVIATIONS AND ACRONYMS 8 EXECUTIVE SUMMARY 9 The challenge 9 The objective 10 The approach 11 The impact 12 The way forward 13 1 BACKGROUND 15 2 WHAT IS THE NBSOS? 18 3 HOW IS THE NBSOS APPLIED? 20 3.1 The urban NBS opportunity scan 21 3.2 The coastal NBS opportunity scan 28 4 HOW IS THE NBSOS INFORMING DEVELOPMENT FINANCE? 39 5 SUSTAINABILITY AND OPERATIONAL MODEL 43 6 NEXT STEPS AND EVOLUTION OF THE NBSOS 45 6.1 The development of analytical capabilities 45 6.2 Toward more capacity for delivery 46 REFERENCES 47 APPENDIX A: DETAILED METHODOLOGY OF THE URBAN NBSOS 49 APPENDIX B: DETAILED METHODOLOGY OF THE COASTAL NBSOS REPORT 87 THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN TABLE OF CONTENTS MAPS Map ES-1: Locations where investment opportunity mapping using the NBSOS was used to inform World Bank–financed operations and core diagnostics.......................................................................12 Map 1: Mapping of priority areas: Example maps of NBS priority area identification across the urban area of Dakar, Senegal ........................................................................................................................22 Map B1: Recommended areas for building solutions in Dakar, senegal...............................................................23 Map 2: Mapping of NBS suitability: Dakar, Senegal.............................................................................................24 Map 3: Mapping of NBS suitability, Dakar, Senegal.............................................................................................24 Map 4: Mapping of open green space benefits, Dakar, Senegal.......................................................................... 25 Map 5: Optimization using multicriteria analysis, Dakar, Senegal....................................................................... 27 Map 6: Mapping coastal flooding, Viti Levu, Fiji..................................................................................................30 Map 7: Extent of coastal ecosystems, Viti Levu, Fiji............................................................................................31 MAP 8: Opportunities for mangroves, Viti Levu, Fiji............................................................................................ 33 MAP 9: Coastal flood risk reduction of NBS, Viti Levu and Vanua Levu, Fiji........................................................ 35 MAP 9: Benefits and costs of NBS, Viti Levu, Fiji................................................................................................ 37 FIGURES FIGURE ES-1: Nature-Based Solutions Opportunity Scan Methodology.............................................................. 11 FIGURE 1: World Bank projects with nature-based components between fiscal years 2012 and 2023................16 FIGURE 2: Nature-Based Solutions Opportunity Scan Methodology...................................................................19 FIGURE 3. Families of Nature-Based Solutions...................................................................................................20 FIGURE 4. Components, data, processes, and results comprising the four steps for applying the NBSOS in urban areas.....................................................................................................................................21 FIGURE 5: Benefits provided by open green spaces............................................................................................ 26 FIGURE 6: Components, data, processes, and results comprising the four steps for applying the NBSOS in coastal areas.................................................................................................................................29 FIGURE 7: NBS types and interventions considered in the coastal NBSOS......................................................... 32 FIGURE 8: Description of NBS scenarios used in the benefits assessment........................................................34 FIGURE 9: Benefits and costs of mangrove interventions per hectare (present value) in Viti Levu.................... 38 BOXES BOX 1: Mapping potential opportunities for building solutions ..........................................................................23 TABLES TABLE 1: Details of NBSOS delivered between mid-2022 and April 1, 2024.........................................................41 THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN 7 ACKNOWLEDGMENTS The Nature-Based Solutions Opportunity Scan (NBSOS) report was developed by the World Bank’s Global Program on Nature-based Solutions for Climate Resilience (GPNBS) with support from the World Bank’s Global Facility for Disaster Reduction and Recovery (GFDRR). Further support and assistance was received from PROBLUE, the Global Water and Sanitation Partnership (GWSP), NBSInvest, and the European Space Agency’s Global Development Assistance program. This report was written by a dedicated team led by Boris van Zanten and Brenden Jongman and composed of Alida Alves Beloqui, Luke Brander, Fabio Cian, Alejandra Gijon Mancheno, Marie-Flore Michel, William Ouellette, Matteo Parodi, Borja G. Reguero, Ian Andrew Smith, and Koen F. Tieskens. The team extends special thanks to the peer reviewers Sajid Anwar, Suranga Kahandawa, Nancy Lozano Gracia, Diego Rodriguez, and chair Niels Holm-Nielsen, whose insights and feedback were critical to enhancing the quality and rigor of this report. The report was designed by Estudio Relativo and edited by Hope Steele. Appreciation is extended to all contributors and supporters for their dedication and collaboration, which have been instrumental in the successful completion of this project. 8 THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ABBREVIATIONS AND ACRONYMS API Application Programming Interface GPURL Urban, Disaster Risk Management, AOI area of interest Resilience and Land ASAs advisory services and analytics IBAT Integrated Biodiversity Assessment CCDRs Country Climate and Development Tool Reports IDA International Development Association CIF Climate Investment Funds IPF investment project financing CO2/ha/year carbon dioxide per hectare per year LAC SD Latin America and the Caribbean CSOs civil society organizations Sustainable Development Practice DEM Digital Elevation Model Group ENB Environment, Natural Resources and MDBs multilateral development banks the Blue Economy NBS nature-based solutions EO Earth observation NBSOS Nature-Based Solutions Opportunity GCS Google Cloud Storage Scan GEE Google Earth Engine NDVI normalized difference vegetation index GFDRR Global Facility for Disaster Reduction PG Practice Group and Recovery RETF recipient-executed trust fund GIS geographic information systems SCC social cost of carbon GP Global Practice SD Sustainable Development Practice GPNBS Global Program on Nature-Based Group Solutions for Climate Resilience SFINCS Super-Fast INundation of CoastS All dollar amounts are US dollars unless otherwise indicated. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: EXECUTIVE SUMMARY 9 ES 1 2 3 ES 4 5 6 A1 EXECUTIVE A2 SUMMARY THE CHALLENGE Countries are facing rising Mangrove 2 climate-related challenges that ©Joel Vodell on Unsplash are making them more vulnerable to climate-related disasters. These problems are especially relevant in cities and coastal areas, where climate risks emerge because of sea-level rise, increasing intensity of rainfall, and extreme heat. Investing in nature-based solutions (NBS) for climate resilience can be effective in reducing climate risks while also bringing other important benefits for communities and the environment. These investments can utilize a variety of natural features by, for instance, creating urban green spaces and corridors, restoring watercourses and coastlines, and preserving natural wetlands and mangroves. Often, the optimization of benefits is achieved through the integration of natural and gray infrastructure, thereby minimizing life-cycle costs and enhancing environmental outcomes. Even though the World Bank’s NBS for climate resilience lending portfolio is steadily growing, governments and Task Teams face challenges in the identification, design, and implementation of investment projects. 10 CHAPTER: EXECUTIVE SUMMARY THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES The integration of NBS for climate resilience into the World Bank portfolio has increased 1 substantially. Between fiscal years 2012 and 2023, the World Bank approved over 2 200 projects with NBS components. The financing committed to NBS components in these 3 4 projects combined exceeds $10 billion. A significant share of these lending operations has 5 received technical support from the Global Program on Nature-Based Solutions for Climate 6 Resilience (GPNBS),1 which is housed at the Global Facility for Disaster Reduction and A1 Recovery (GFDRR). While interest in NBS for climate resilience has increased at the World A2 Bank and among clients, technical barriers persist and investments in NBS remain a modest share of total financing going toward infrastructure and climate resilience. Actionable A common challenge is that identifying potential NBS information is investments is hindered by a lack of data and technical expertise on NBS. Having the right information to needed at the right understand opportunities for NBS investment at the time to support the identification stage is critical for enabling further scaling up of NBS greening of infrastructure and climate-resilience projects. Without rapid and accurate information on NBS investments. opportunities, and without the right technical expertise throughout the project cycle, projects often fall back on more traditional gray infrastructure investments. THE OBJECTIVE The Nature-Based Solutions Opportunity costs and benefits. The NBSOS is therefore Scan (NBSOS) supports the World Bank, its primarily developed for use in the early phase of clients, and its development partners in project investment planning, where it provides clear identifying NBS investment opportunities, indications of potential NBS interventions and their benefits for a city or coastline. In addition, the NBSOS understanding the benefits that these may is applied for strategic diagnosis—including Country bring for communities and the environment, Climate and Development Reports (CCDRs) and and integrating these NBS interventions into climate adaptation planning—where cross-sectoral investment programming. investment prioritization is important. The development of the NBSOS responds directly NBSOS is a geospatial analysis and participatory to the demand from project teams and government process offered by GFDRR as an on-demand service decision-makers who are interested in integrating in cities and coastal areas worldwide. The NBSOS NBS in investment planning but do not have the is applied at the request of and in collaboration with necessary information to understand investment World Bank project teams and their clients, and it is typologies, geographic priorities, and the expected tailored to each specific case in order to provide the 1 For more information about GPNBS, see https://naturebasedsolutions.org/. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: EXECUTIVE SUMMARY 11 ES most useful advice. NBSOS relies on an array of openly available global geospatial data sets 1 that are complemented with local data where available. The tool provides a starting point 2 in understanding NBS investment opportunities. Following the NBSOS, the GPNBS team 3 provides additional capacity building and technical support to help ensure that its outcomes 4 inform the development of more detailed analyses, such as pre-feasibility and design 5 studies and, eventually, investments. 6 A1 A2 THE APPROACH The NBSOS uses 10- to 30-meter The methodology and the software are designed in resolution global geospatial data such a way that its deliverables—including results, interpretation, recommendations, and a full package and a sophisticated methodology of geospatial data—can be prepared in approximately to map the potential benefits four to six weeks. The NBSOS is tailored to the local of NBS and identify investment context prior to implementation in an inception meeting with the Task Team, in some cases also opportunities in cities and along involving the client. coastlines anywhere in the world. The NBSOS analysis consists of four FIGURE ES-1: NATURE-BASED SOLUTIONS OPPORTUNITY SCAN METHODOLOGY methodological steps (see figure ES-1): problem analysis, suitability mapping, benefit modeling, and decision support. The first step entails understanding the STEP UNDERSTANDING magnitude and spatial variation of climate THE PROBLEM resilience challenges and natural hazards in the area of interest. The second step consists of mapping suitable areas for STEP MAPPING NBS protection and creation of the NBS types SUITABILITY considered. The third step models and estimates the positive impact of NBS in addressing the identified climate resilience STEP MODELING NBS challenges and natural hazards. Finally, BENEFITS the fourth step consists of finding the optimal distribution of NBS investments to maximize benefits, through multicriteria DECISION STEP or cost-benefit analysis, providing relevant SUPPORT information to decision-makers. SOURCE: Original figure for this publication. 12 CHAPTER: EXECUTIVE SUMMARY THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES 1 2 3 THE IMPACT 4 5 6 A1 The NBSOS was successfully Most projects informed are investment project financing (IPF) A2 implemented in 20 countries engagements that use the NBSOS to identify potential NBS between mid-2022 and early 2024, interventions as part of the project. In addition, the NBSOS has been including 8 coastal landscapes used to identify NBS for adaptation investment needs for CCDRs. The NBSOS has been applied in different regions (map ES-1)—such as and 51 cities, informing an Africa, South Asia, and Latin America and the Caribbean—and across estimated $2.3 billion in Global Practices such as the Urban, Disaster Risk Management, development financing as well Resilience and Land (GPURL); Environment, Natural Resources and as key strategic assessments the Blue Economy (ENB); Water; and Transport Global Practices. (see map ES-1). Experience shows that the NBSOS has a higher project-level impact if GPNBS supports Task Teams by presenting and interpreting results, and by integrating results in (pre-)feasibility and design studies. Through early identification of potential NBS investment locations and by estimating the potential benefits of NBS reducing climate risk and providing ecosystem services, the NBSOS has been instrumental for project identification and design. Experience also shows the need to provide guidance to Task Teams, thus ensuring the correct interpretation of results and clarifying limitations of the methodology. The NBSOS is a first and rapid assessment that can provide useful indication of areas and types of NBS investments and their benefits (and costs), but it cannot replace a full feasibility and design study, which is often a next step. The role of an NBSOS assessment in the investment process should be clear to the task team and government clients. MAP ES-1: LOCATIONS WHERE INVESTMENT OPPORTUNITY MAPPING USING THE NBSOS WAS USED TO INFORM WORLD BANK–FINANCED OPERATIONS AND CORE DIAGNOSTICS SOURCE: Original map for this publication based on coordinates of NBSOS study sites. Locations of NBSOS NOTE: NBSOS = Nature-Based implementation Solutions Opportunity Scan. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: EXECUTIVE SUMMARY 13 ES 1 THE WAY FORWARD 2 3 4 5 6 Most projects financing climate In addition to highlighting specific opportunities A1 resilience, water, or other public for NBS investments, NBSOS results demonstrate the advantages of having a more holistic approach A2 infrastructure in urban and to NBS, integrating multiple benefits, and planning coastal settings can benefit from multifunctional solutions. This is, for example, applying the NBSOS at an early relevant for strategic engagements such as CCDRs, especially for countries with vulnerable coastlines, stage to identify opportunities for big cities, or rapid urban expansion. In this type of NBS investments. engagement, the NBSOS has proven itself to be an effective tool to inform development finance. Analytical capabilities of the NBSOS Within the current operational model, it are—and will be—continuously updated. Incremental science-based improvements is possible to further increase capacity are continually tested and implemented to deliver NBSOS to Task Teams by during applications. More fundamental about 25–50 percent. analytical improvements will focus on expanding the capability of the NBSOS Increased efficiency as a result of standardization, along with estimating biodiversity impact and a larger pool of expert consultants to run the analyses, can quantifying pluvial flood reduction by NBS expand its capacity up to about 70–80 cities and 15 coastal in urban areas. In addition, a growing body landscapes per year, but if the demand exceeds this capacity, of research and practice will increase it will lead to a longer response time to the requests of the ability to estimate unit costs of NBS Task Teams. Additional capacity could be created either adjusted to the landscape or country context by developing a webtool with graphical user interface for as part of the NBSOS. As more NBS for nonexpert users, or by stimulating industry (that is, World Bank climate resilience projects are studied and vendors) to adopt the NBSOS methods by opening the source implemented globally, more data points on code and enabling knowledge transfer. The latter option seems costs become available. In collaboration favorable, since 20 external partners—including multilateral with GFDRR and PROGREEN,2 GPNBS is development banks (MDBs), other development partners, developing rapid costing tools to inform private sector firms, and leading civil society organizations and improve unit cost estimations used in (CSOs)—have requested access to the NBSOS. Hence, the the NBSOS. source code of the urban NBSOS will become available on request upon publication of this report for a selected number of external partners (see https://naturebasedsolutions.org/ opportunity-scan). 2 Details about PROGREEN, which is “a global partnership for sustainable and resilient landscapes,” can be found at https://www. progreen.info/. 14 THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN Urban streams ©MJ Haru_Unsplash THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: BACKGROUND 15 ES 1 2 3 1 4 5 6 A1 BACKGROUND A2 ©Mojtaba Hoseini_Unsplash Cities and coastal From the 1970s to the period 2010–20, the frequency of extreme heat areas are increasingly and dry events increased across cities globally, and the frequency of extreme wet events has increased since the 1990s. Global sea-level rise exposed to climate of about 0.125 millimeters per year is also increasing the risk of flooding change impacts. for coastal cities. More than half of the world’s population lives in cities, while about 10 percent live along low elevation coastal zones. Urban areas are becoming more crowded, with the resultant loss of greenspace affecting biodiversity and with climate and disaster impacts such as extreme heat and flooding having a greater effect on a greater number of people. By protecting natural systems and investing in nature-based solutions (NBS), infrastructure projects can build resilience and protect development gains for future generations (see Mukim and Roberts 2023). NBS for climate resilience are NBS can provide a range of coastal wetlands to reduce multifunctional solutions to benefits such as reducing disaster erosion, and planned urban meet the rising challenge of risks, restoring biodiversity, afforestation initiatives to alleviate climate resilience. creating opportunities for and mitigate urban heat stress. recreation, improving human NBS are often integrated into health, ensuring water and larger infrastructure investment food security, and supporting projects and can complement gray community well-being and infrastructure to reduce disaster livelihoods (van Zanten et al. 2023). risk and build resilience, while Examples include the strategic bringing additional benefits. design of urban parks to combat flooding, the rehabilitation of 16 CHAPTER: BACKGROUND THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES Despite the multiple benefits of NBS, it remains 1 2 difficult to bring these solutions to scale as part 3 of development, environmental, and climate 4 5 resilience projects. 6 A1 A2 Even though the integration of NBS for climate committed to NBS components. GPNBS-supported resilience investment has been growing, investments projects benefitted over 24 million people, afforested/ in NBS are still a minor share of total financing of reforested or restored more than 2.8 million hectares the World Bank’s sustainable development and of degraded ecosystems, protected nearly 35,000 infrastructure portfolios. Between fiscal years 2012 kilometers of coastal areas, and brought 14 million and 2023, the Bank approved 200 projects that include hectares of land under sustainable land management an NBS component. The total committed financing or enhanced biodiversity protection. Despite these for project components that include NBS exceeds results, several implementation barriers persist, and $10 billion (figure 1) (World Bank 2023; the numbers there is an opportunity to increase the integration from this report have been complemented with World of NBS as part of projects investing in urban Bank fiscal year 2022 and 2023 committed financing). development, coastal management and the blue Since 2021, the Global Program on Nature-Based economy, water management, transport, landscape Solutions for Climate Resilience (GPNBS) supported restoration, and climate mitigation. 31 of these projects, with $2.4 billion financing FIGURE 1: WORLD BANK PROJECTS WITH NATURE-BASED COMPONENTS BETWEEN FISCAL YEARS 2012 AND 2023 3500 40 35 3000 30 FUNDING (US$ MILLIONS) 2500 NUMBER OF PROJECTS 25 2000 20 1500 15 1000 10 500 5 0 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 APPROVED NBS COMMITMENT TOTAL NBS PROJECTS SOURCE: Original figure for this publication based on NBSOS data. NOTE: NBS = Natured-Based Solutions. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: BACKGROUND 17 ES 1 2 3 4 5 6 A1 A2 One of the challenges to integrating NBS in development finance is the early identification of investment opportunity areas, particularly in data-poor environments. In the project identification phase, with limited time and resources at hand, there is a need for a quick and robust analytical approach that can support the initial identification of ©Alec Douglas opportunity areas for investment in NBS, relying on globally available geospatial data. Such Unsplash investment opportunity mapping can inform projects in different ways: It supports project It provides a baseline for local And it informs (pre-) identification and is a starting stakeholder and community feasibility studies, design, and point of an investment plan. dialogue. implementation. Responding to this need, the Global Program on Nature-Based Solutions for Climate Resilience (GPNBS) and the World Bank’s Global Facility for Disaster Reduction and Recovery (GFDRR) have developed the NBS Opportunity Scan (NBSOS).3 3 For details about the GPNBS see https://www.naturebasedsolutions.org/; see also GFDRR 2023. 18 CHAPTER: WHAT IS THE NBSOS? THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES 1 2 3 4 5 2 6 A1 A2 WHAT IS THE NBSOS? ©Francesco Ungaro_Unsplash The primary objective of This ensures that its outcomes can employing the NBSOS is inform the development of more detailed analyses, such as pre-feasibility studies. to identify priorities for By identifying opportunities for NBS potential NBS investments during preliminary assessments and in projects that are in project inception, it becomes more feasible to seamlessly integrate nature- preparation or in the early based infrastructure into final solutions implementation stage. and investment strategies. NBSOS is a standardized geospatial It aims to support World Bank teams, governments, methodology and a participatory process and other investors to understand which NBS types offered as an on-demand service to Task have most potential in a particular city, what potential Teams for NBS investment opportunity project sites are, what their potential benefits are, and how NBS can complement gray infrastructure. mapping in cities and coastal areas worldwide. The NBSOS methodology is designed in a way that its deliverables—including results, interpretation, and recommendations, and geospatial data package—can be prepared in approximately four to six weeks. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: WHAT IS THE NBSOS? 19 ES NBSOS relies on an array of openly available 10- to 30-meter resolution Earth observation data and other 1 geospatial data sets. The analysis consists of four methodological steps, as shown in figure 2. 2 3 4 5 6 A1 A2 First, understanding The second step The third step models Fourth, the decision the problem and consists of mapping and estimates support step consists mapping hazards: suitable areas for the positive of finding the optimal what are the implementing the NBS socioeconomic distribution of NBS to spatial distribution types considered. impact (that is, the maximize benefits, and magnitude To achieve this, the benefits) NBS has which is accomplished of resilience and physical conditions in addressing the through multicriteria sustainability (for example, soil identified resilience and cost-benefit challenges and what type, slope) required and sustainability analysis. are the solutions for each NBS type are challenges. considered to cope ascertained. with these challenges? FIGURE 2: NATURE-BASED SOLUTIONS OPPORTUNITY SCAN METHODOLOGY Visualize spatial variability in STEP UNDERSTANDING hazard exposure & highlight regions THE PROBLEM of impactful NBS intervention Map locations to protect, enhance, STEP MAPPING NBS and create green infrastructure for SUITABILITY each NBS family STEP MODELING NBS Estimate potential benefits from each BENEFITS NBS family for all suitable locations Identify opportunity areas for NBS DECISION STEP investment through multi criteria/cost- SUPPORT benefit analysis SOURCE: Original figure for this publication. NOTE: An NBS family is a group of nature-based solutions, such as urban farming or mangrove forests. See figure 3 and World Bank 2021. 20 CHAPTER: HOW IS THE NBSOS APPLIED? THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES 1 2 3 4 5 3 6 A1 A2 HOW IS THE NBSOS APPLIED? ©Beira_WB In this process, the area of interest is determined: in the urban case, this can vary from a small settlement area to a large catchment of several hundred kilometers squared; in the coastal case, it can vary from few Before the hundred kilometers to the entire coast of a small country. Moreover, analysis starts, through consultation with the Task Team, the most important NBS benefits linked to the climate resilience and sustainability challenges in the NBSOS is the area are identified and relevant NBS types are selected. The NBS tailored to the are selected from the 14 different NBS families that are described in the needs of the Catalogue of Nature-Based Solutions for Urban Resilience (World Bank 2021) (figure 3). Task Team. FIGURE 3: FAMILIES OF NATURE-BASED SOLUTIONS URBAN AND UPLAND RIVER AND STREAM TERRACES AND SLOPES BUILDING SOLUTIONS OPEN GREEN SPACES GREEN CORRIDORS FORESTS RENATURATION NATURAL INLAND CONSTRUCTED INLAND URBAN FARMING BIORETENTION AREAS RIVER FLOODPLAINS MANGROVE FORESTS WETLANDS WETLANDS PONDS, LAKES, AND SUBMERGED AQUATIC OTHER COASTAL SALT MARSHES SANDY SHORES REEF ECOSYSTEMS SMALL WATER BODIES VEGETATION WETLANDS SOURCE: World Bank 2021. NOTE: NBS = nature-based solutions. Next, each of the four steps followed to implement the NBSOS is described separately for the urban and coastal analyses. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: HOW IS THE NBSOS APPLIED? 21 ES 3.1 1 2 THE URBAN NBS OPPORTUNITY SCAN 3 4 5 The different components, data, processes, and results of the four methodological steps for applying the 6 NBSOS in urban areas are summarized in figure 4. Each of these steps is explained in the sections below. A1 A detailed description of the methodology is provided in appendix A. The urban NBSOS is described in this A2 section using the example of Dakar, Senegal. FIGURE 4: COMPONENTS, DATA, PROCESSES, AND RESULTS COMPRISING THE FOUR STEPS FOR APPLYING THE NBSOS IN URBAN AREAS STEP 1 UNDERSTANDING THE PROBLEM HAZARDS INPUT DATA GIS PROCESSING OUTPUT RESULT • DEM Flood Exposure & Reduction Pluvial Flood Exposure & PLUVIAL FLOODING • Flood Model • Population Density Potential Mapping Priority Area Maps HEAT STRESS • Air temperature Model Air Temperature Exposure Heat Stress Priority Area Maps • Population Density Mapping LACK OF GREEN SPACE • Existing Green Space Map Health/Recreation • Population Density Recreation Opportunity Mapping Priority Area Maps STEP 2 MAPPING NBS SUITABILITY NBS FAMILIES INPUT DATA GIS PROCESSING OUTPUT RESULT • Soil Properties • Climate Trends • Terrain • Land Cover NBS Suitability Mapping NBS Suitability Maps • Land Use • Greenness • Bare Soil Frequency STEP 3 MODELING NBS BENEFITS BENEFITS INPUT DATA GIS PROCESSING OUTPUT RESULT Pluvial Flood Reduction • Pluvial Flood Priority Maps Runoff Reduction & Storage Pluvial Flood Benefits Maps • NBS Suitability Maps Model Heat Stress Reduction • Heat Stress Priority Maps Air Temperature Reduction Heat Stress Benefits Maps • NBS Suitability Maps Model Improved Access to • Heath/Recreation Priority Maps Distance from Green Space Health/ Green Space • NBS Suitability Maps Model Recreation Benefits Maps STEP 3 DECISION SUPPORT BENEFITS INPUT DATA GIS PROCESSING OUTPUT RESULT Pluvial Flood Importance • Pluvial Flood Benefits Maps Heat Stress Importance • Heat Stress Benefits Maps Combined Weighted Benefits Optimal NBS Maps Mapping Access to Green Space • Health/Recreation Benefits Importance Maps SOURCE: Original figure for this publication. NOTE: DEM = Digital Elevation Model; GIS = geographic information systems; NBS = nature-based solutions; NBSOS = Nature-Based Solutions Opportunity Scan. 22 CHAPTER: HOW IS THE NBSOS APPLIED? THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES 1 3.1.1 MAP 1: MAPPING OF PRIORITY AREAS: EXAMPLE MAPS OF NBS PRIORITY AREA IDENTIFICATION ACROSS THE URBAN AREA OF DAKAR, SENEGAL 2 3 STEP 1: 4 UNDERSTANDING A 5 URBAN RESILIENCE PRIORITY 6 CHALLENGES AREAS FOR PLUVIAL FLOOD A1 Priority areas for investment EXPOSURE A2 REDUCTION within the city are identified by looking at where the -- Roads resilience and sustainability challenges are most Very Low prevalent. Low Medium The urban NBSOS assesses where High to implement NBS to reduce Very High pluvial flooding; where to mitigate extreme heat; and where there is B a need to increase access to green PRIORITY space, which is assumed to provide AREAS FOR health, recreation, and social HEAT EXPOSURE REDUCTION cohesion benefits to communities (map 1). -- Roads Exposure to flooding is calculated as a product of population Very Low density4 and annual flood Low probability based on the Fathom Medium global pluvial flood model High (Sampson et al. 2015). Using Very High a 30-meter resolution Digital Elevation Model (FABDEM, no date), C a downstream catchment area PRIORITY is determined for each potential AREAS FOR THE NBS location to prioritize areas CREATION OF GREEN SPACE with the highest flood exposure PROVIDING in its downstream catchment. HEALTH AND RECREATION In the case of heat stress, heat BENEFITS exposure is calculated as the product of population and average Very Low annual air temperature based Low on a global temperature model Medium High Very High -- Roads 4 Population density is calculated using Meta’s Resources and tools for advancing AI, available at https:// SOURCE: Original maps for this publication based on NBSOS data. ai.meta.com/resources/. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: HOW IS THE NBSOS APPLIED? 23 ES (Hooker, Duveiller, and Cescatti. 2018). Finally, to define areas lacking green spaces that provide health and 1 recreation benefits, the NBSOS estimates the number of people within each sub-neighborhood that are not 2 within 300 meters of a green space larger than 1 hectare (Konijnendijk 2023), based on the remote sensing 3 derived normalized difference vegetation index (NDVI). 4 5 3.1.2 6 A1 STEP 2: A2 For example, the NBSOS can identify areas with degraded vegetation MAPPING URBAN NBS and determine to what extent that area can be restored by implementing SUITABILITY NBS such as green corridors, urban forests, or open green spaces. Areas suitable for creating new NBS are mapped for each NBS type, Suitability maps are created to allowing visualization of each of the selected NBS types at the city identify areas for protecting level (map 3). The NBSOS looks for opportunities in non–built-up areas. existing greenspaces and However, box 1 shows an example of a customized NBSOS application areas suitable for creating or in which recommendations of preferred areas for building solutions constructing new NBS (map 2) implementation were provided. BOX ❶ MAP B1: RECOMMENDED AREAS FOR BUILDING SOLUTIONS IN DAKAR, SENEGAL MAPPING POTENTIAL OPPORTUNITIES FOR BUILDING SOLUTIONS A Because of very high urban density SOLUTIONS and low space availability for other FOR FLOOD NBS, opportunities for solutions in REDUCTION buildings were studied for the city of Dakar. This is an example of the -- Roads possibilities of customization when using the NBSOS. Recommended areas for building Very Low solutions are identified combining Low three criteria: areas with high building Medium density, areas with low opportunities High for creating other NBS, and areas with high priority for applying NBS for Very High either flood or heat reduction. Green roofs and rain barrels are effective to reduce runoff and B mitigate flooding downstream. In the SOLUTIONS “very high” flood reduction potential FOR HEAT hexagons (East Dakar), buildings REDUCTION cover, on average, 79 percent of the area. Green roofs and green walls -- Roads are effective to mitigate heat stress, impacting mutually in the public space and inside buildings. In the Very Low “very high” heat potential hexagons Low (central urban area, buildings cover Medium on average 89 percent of the area. High This information can aid decision -makers to decide where to focus Very High incentive programs and engagement campaigns to encourage the SOURCE: Original maps for this publication based on NBSOS data. development of this type of NBS. 24 CHAPTER: HOW IS THE NBSOS APPLIED? THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES MAP 2: MAPPING OF NBS SUITABILITY: DAKAR, SENEGAL NBS suitability is determined by 1 2 NBS-type specific rule sets using a diverse 3 set of Earth observation indicators that 4 rely on publicly available data sets with 5 global coverage. 6 A1 The rule sets are based on indicators that describe A2 soil properties (Earth Engine Data Catalog, no date-f, no date-g); bare soil frequency (Earth Engine Data Catalog, no date-e); surface water frequency (Earth Engine Data Catalog, no date-d); precipitation (Earth Engine Data Catalog, no date-a); slope (FABDEM, no date); land capability (Sentinel-2 timeseries) (Earth Engine Data Catalog, no date-e, no date-c); land cover (Earth Engine Data Catalog, no date-b, no Protection of Existing Greenspace date-c) and distance to roads; buildings, and water NBS Creation Opportunities bodies (OpenStreetMap contributors 2024). Settlement Extent SOURCE: Original map for this publication based on NBSOS data. NOTE: This is an example spatial delineation of currently vegetated regions to protect and suitable areas for the creation of NBS within the urban area of Dakar, Senegal. MAP 3: MAPPING OF NBS SUITABILITY, DAKAR, SENEGAL A B OPEN GREEN SPACES GREEN CORRIDORS Protection of Existing Greenspace NBS Creation Opportunities NBS Creation Opportunities Settlement Extent Settlement Extent SOURCE: Original map for this publication based on NBSOS data. NOTE: Example of suitable areas for creating new NBS are mapped for open green spaces and green corridors within the urban area of Dakar. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: HOW IS THE NBSOS APPLIED? 25 3.1.3 ES 1 STEP 3: 2 3 MODELING NBS BENEFITS IN CITIES 4 The NBSOS applies index-based valuation To demonstrate the benefits in a spatially explicit 5 6 of flood reduction benefits, heat reduction way, the models described in Step 1 are utilized to A1 benefits, and health and recreation estimate the three main benefits. For instance, the A2 benefits that are expected when the NBS spatial flood mitigation model is used to measure the is implemented. area in the city where NBS have the highest potential to reduce the exposure of people to stormwater Optionally, these benefits can be complemented flooding. The heat model estimates the effect of NBS with a qualitative score (high-medium-low) for other families on air temperature and how many people benefits as defined in the Catalogue of Nature-Based near the NBS benefit from this cooling effect. Finally, Solutions for Urban Resilience (World Bank 2021). the health, recreation, and social interaction benefits These benefits describe the positive impact of the of the NBS are estimated by counting the additional selected types of NBS interventions addressing population that has access to green space within 300 climate resilience and sustainability challenges in the meters of their homes. The results are presented specific city being considered. The NBSOS estimates per NBS as the number of hectares providing high, these benefits for all the suitable areas and for each medium, and low levels of benefit, and are displayed NBS family, as identified in Step 2. on the map (map 4). This helps to demonstrate which NBS families will yield the most benefits and where to implement them to maximize those benefits. MAP 4: MAPPING OF OPEN GREEN SPACE BENEFITS, DAKAR, SENEGAL A B REDUCING PLUVIAL FLOOD EXPOSURE IMPROVING RECREATION/HEALTH VIA ACCESS TO GREEN SPACES Low Low Medium Medium High High Settlement Extent Settlement Extent 26 CHAPTER: HOW IS THE NBSOS APPLIED? THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES C D 1 REDUCING HEAT STRESS EXPOSURE OPEN GREEN SPACE POTENTIAL BENEFITS 2 3 4 OPEN GREEN SPACE POTENTIAL BENEFITS 5 FLOOD Low 6 Medium A1 High A2 HEALTH / Low RECREATION Medium High HEAT Low Medium Low High Medium High 0 50 100 150 200 250 300 350 Settlement Extent AREA (HA) SOURCE: Original maps and figure for this publication based on NBSOS data. NOTE: This is an example of spatial variability in the potential benefits from the creation of open green spaces for reducing pluvial flood exposure, improving recreation/health via access to open green spaces, and reducing heat stress across the city. Other benefits of NBS — such FIGURE 5: BENEFITS PROVIDED BY OPEN GREEN SPACES as resources production and biodiversity (figure 5)—can be Pluvial flood risk Social Interaction valued using a land use matrix- reduction based valuation approach (Burkhard et al. 2012), linking Heat stress risk NBS families defined in the reduction Catalogue of Nature-Based Solutions for Urban Resilience to a qualitative score of 1 (low) Biodiversity to 3 (high). These benefits are added to the three spatially OPEN GREEN modeled benefits (flood Tourism and SPACES recreation reduction, heat reduction, and health and recreation/social interaction) on a case-by- case basis, depending on their Carbon Human Health local relevance. storage and sequestration Stimulate local economies and job creation SOURCE: World Bank 2021. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: HOW IS THE NBSOS APPLIED? 27 3.1.4 ES 1 STEP 4: 2 3 DECISION SUPPORT FOR URBAN 4 NBS 5 MAP 5: OPTIMIZATION USING MULTICRITERIA ANALYSIS, DAKAR, SENEGAL 6 As a final step, NBS investment opportunity A1 areas, based on a spatial multicriteria A2 analysis, are suggested. In practice, areas suitable for creating or constructing NBS interventions—such as urban forests, open green spaces, and urban agriculture—overlap. However, each of these NBS interventions will provide different benefits. Urban forests have a high canopy density and will be effective in reducing heat, while open green spaces can be designed for water storage and reduce flooding. By assigning weights to the different benefits (for example, assigning a OPTIMAL NBS higher weight to the flood reduction potential Bioretention Areas when that is more important than reducing Green Corridors heat), NBS that compete for space can be Urban Forests prioritized in areas that are exposed to multiple Open Green Spaces hazards and resilience challenges. The NBSOS Settlement Extent prioritizes those areas that are most effective in delivering the selected benefits (map 5). Potential investment scenarios can be developed from this SOURCE: Original map for this publication based on data from the NBSOS. NOTE: This is an example of scan outputs describing the optimization exercise—for instance, by choosing spatial variability in optimal NBS solutions. Optimal solutions the NBS delivering the highest 20 or 10 percent of were determined by combining weighted normalized benefits combined benefits. from pluvial flood exposure reduction (60 percent), heat stress reduction (20 percent), and improved health/recreation (20 percent). The optimal solution describes the NBS type providing the highest level of combined benefits for each pixel. 3.1.5 URBAN NBSOS VALIDATION A validation exercise was performed to validate NBS implementation for flood reduction and on the inputs and outputs of the NBSOS pluvial flood assessment of the level of benefits obtained from assessment and related recommendations. the proposed solutions. Therefore, it was essential The NBSOS uses flood hazard from a global model to understand the agreement between the Fathom (Sampson et al. 2015) to define where urban areas get flood maps (Fathom 2019, 2023; Sampson et al. flooded and, based on this, where to implement NBS 2014) and flood hazard maps prepared using locally to reduce pluvial flooding. Flood hazard data have sourced data. In addition, a key issue for the NBSOS a high impact on the definition of priority areas for was to verify the priority areas for flood reduction 28 CHAPTER: HOW IS THE NBSOS APPLIED? THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES NBS through comparison with a local high-resolution of more expensive and time-consuming studies, 1 flood modeling study. Finally, the NBSOS estimated which could deliver more accurate results. 2 degradation levels of urban greenspaces to classify 3 4 areas as protection areas (low degradation) and 5 creation areas (high degradation). The last validation 6 objective was to compare NBSOS results on A1 protection and creation with solutions proposed by A2 the local study. The results from the validation exercise showed that, despite some difference in accuracy between the global model and the local one, the main hazard characteristics are captured and factored in the NBSOS. Moreover, the use of an updated version of this model (Fathom v3; see Fathom 2023) shows great improvement in resolution and accuracy, which will be reflected in the accuracy of future NBSOSs. Regarding NBSOS outputs, despite the huge difference in terms of resolution and accuracy between the inputs used by the NBSOS (global and low-resolution data) and those used by the local study (in-situ and high-resolution data), the NBSOS can correctly identify the main hotspots for implementing NBS for flood risk reduction, and it is also able to make recommendations for NBS creation and protection River renaturation similar to those of the local study. The NBSOS is able ©Arun antony on Unsplash to provide a direction early on, before the application 3.2 THE COASTAL NBS OPPORTUNITY SCAN The different components, data, Each of these steps is explained in the sections below. A detailed description of the methodology is provided processes, and results of the four in appendix B. The coastal NBSOS is described in this methodological steps for applying section uses the example of Viti Levu, the largest the NBSOS in coastal areas are island in Fiji. The main methodological differences between the urban and the coastal NBSOSs are that summarized in figure 6. the coastal version assesses coastal NBS over a 30-year period, considering ecosystem degradation, climate change and socioeconomic development, while the urban NBSOS does not consider the THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: HOW IS THE NBSOS APPLIED? 29 ES evolution of benefits over time. Also, in contrast to the urban NBSOS, the coastal version presents a cost- 1 benefit analysis for the interventions, while the urban NBSOS assesses trade-offs between NBS types using 2 multicriteria analysis. 3 4 5 FIGURE 6: COMPONENTS, DATA, PROCESSES, AND RESULTS COMPRISING THE FOUR STEPS FOR APPLYING THE NBSOS IN COASTAL AREAS 6 A1 STEP 1 UNDERSTANDING THE PROBLEM A2 HAZARDS INPUT DATA GIS PROCESSING OUTPUT RESULT • Bathymetry • Topography • Land Use COASTAL FLOODING • Sea Level Rise SFINCS Coastal Flood Modeling Coastal Flood Exposure Maps • Waves • ESLs • Population Density • Building Footprint STEP 2 MAPPING NBS SUITABILITY NBS FAMILIES INPUT DATA GIS PROCESSING OUTPUT RESULT • Currents NBS Extent • Bathymetry • Climate Trends • Land Cover NBS Suitability Mapping NBS Suitability Maps • Land Use • Greenness • Bare Soil Frequency STEP 3 MODELING NBS BENEFITS BENEFITS INPUT DATA GIS PROCESSING OUTPUT RESULT Coastal Flood Reduction • Coastal Flood Exposure Maps Flood Exposure Reduction Model Avoided Damage from NBS Maps • NBS Suitability Maps Carbon Sequestration and • Mangrove C Emission Factor Mangrove Extent Carbon Cost Carbon Storage & Sequestration Storage • Mangrove Suitability Maps Model Benefits Maps Tourism and Recreation • Annual Tourism Data Economic Beach Valuation Model Tourism Benefits Maps • Beach and Hotel Location Fisheries and Raw Materials • Ecosystem Service Database Ecosystem Extent Valuation Fisheries/Materials Benefits • NBS Suitability Maps Model Maps STEP 3 DECISION SUPPORT BENEFITS INPUT DATA GIS PROCESSING OUTPUT RESULT Restoration Costs • Cost Databases NBS Suitability Maps Calculation of Benefit-to-Cost Benefit-to-Cost Ratio Maps ratio for Each Project Ecosystem Benefits • Benefits from Step 3 NBS Suitability Maps SOURCE: Original figure for this publication. NOTE: GIS = geographic information systems; NBS = nature-based solutions; NBSOS = Nature-Based Solutions Opportunity Scan; SFINCS = Super-Fast INundation of CoastS. 30 CHAPTER: HOW IS THE NBSOS APPLIED? THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES 1 3.2.1 2 3 STEP 1: 4 UNDERSTANDING COASTAL RESILIENCE CHALLENGES 5 6 Using the hydraulic model Super-Fast Inundation of Structural flood damages to buildings are estimated A1 CoastS (SFINCS; Leijnse et al. 2021), coastal flood overlaying flood extent and depth overlaying flood A2 extent and flood depth is modeled for the baseline maps with datasets of population (FCL and CIESIN scenario and for a 2050 scenario, which considers 2016) and building footprints (Google Research, sea-level rise and increasing storm intensity due to no date; Sirko et al. 2021) for each return period. climate change. The hydraulic model accounts for This step returns maps with flooded population and several effects: extreme surge levels, wave set-up, assets during coastal storm events, now and in future run-up for different return periods (10, 50, 100 and scenarios, considering sea-level rise. To illustrate 500 years), and the increase in water level due to areas most exposed to coastal flooding, the NBSOS sea-level rise (by 2050). The model uses as input a shows the exposure of coastal buildings to a 100- digital elevation model (Hawker et al. 2022);5 land use, year return period coastal flood event under 2050 extreme waves, and water levels (Hersbach et al. 2018); conditions. An example for Viti Levu, Fiji, is displayed and the rates of sea-level rise for future scenarios in map 6. (IPCC 2021). The model output is a flood map showing inundated locations and the local water depth with a spatial resolution of 30 meters. MAP 6: MAPPING COASTAL FLOODING, VITI LEVU, FIJI WATER DEPTH (in meters) 0-1m 2m 3m 4m >4m Flooded Buildings SOURCE: Original map for this publication based on data from the NBSOS. NOTE: This is an example of coastal flood maps showing the flooded buildings in Viti Levu, Fiji, during to a 100-year return period storm by 2050 (including sea-level rise). 5 The database used for the Digital Elevation Model is available at https://data.bris.ac.uk/data/dataset/s5hqmjcdj8yo2ibzi9b4ew3sn. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: HOW IS THE NBSOS APPLIED? 31 3.2.2 ES 1 STEP 2: 2 3 MAPPING COASTAL NBS SUITABILITY 4 Globally available geospatial data sets are The result shows the current extent of key coastal 5 6 used to map the baseline extent of coastal ecosystems that function as nature-based solutions A1 ecosystems, such as mangroves (Bunting for climate resilience (the example of Viti Levu is A2 et al. 2022), coral reefs (Lyons et al. 2020), and shown in map 7). The coastal NBSOS uses a specific beaches (Google Earth 2023). subset of coastal NBS types derived from the long list provided in figure 3. MAP 7: EXTENT OF COASTAL ECOSYSTEMS, VITI LEVU, FIJI NBS CURRENT EXTENT Mangroves Coral reefs Beaches SOURCE: Original map for this publication based on data from the NBSOS. NOTE: This is an example of the baseline extent of key coastal ecosystems that function as nature-based solutions for climate resilience as of 2022 in Viti Levu, Fiji. For mangroves, the NBSOS identifies two types their restoration. These interventions could occur at of interventions: (1) protection of mangroves sites within 100 meters of existing mangroves and at by protecting areas where they exist today and most 100 meters inland from the coastline, capturing remain healthy, and (2) enhancement of mangroves intertidal areas currently without mangrove cover. (figure 7). Protection is preventing degradation or loss of mangrove extent. Enhancement could occur For coral reefs, potential enhancement sites are at locations where mangroves exist at present but identified based on the presence of coral reefs in are sparse, or at locations where mangroves may shallow water areas close to shore—between the have existed in the past but do not at present, and shoreline and a water depth of 3 meters—since where environmental conditions may be favorable for these areas are most effective for flood reduction 32 CHAPTER: HOW IS THE NBSOS APPLIED? THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES purposes. Coral reef enhancement interventions in For the NBSOS, the GPNBS team manually interprets 1 NBSOS should have a measurable coastal protection and digitizes sandy beaches based on the Google 2 or adaptation outcome. They are understood as Earth imagery as a comprehensive globally available 3 4 interventions that reduce the water levels on the reef data set is not available. NBS interventions on 5 and attenuate waves, thereby protecting coastal beaches include the conservation of beaches so that 6 assets. In most cases, such coral reef interventions they can keep their relative height with respect to A1 would be hybrid solutions, combining artificial rising sea levels, thus protecting their function for A2 structures with reef restoration (Jongman et al. 2021). coastal protection and for tourism. In practice, this Coral reef presence is identified using global coral intervention will often require beach nourishments, reef extent data from the Allen Coral Atlas. which should be conducted in a sustainable way to avoid negative environmental impact (figure 7). FIGURE 7: NBS TYPES AND INTERVENTIONS CONSIDERED IN THE COASTAL NBSOS Area with Creation of healthy protected areas MANGROVES forest TYPE 1 (HEALTY) PROTECTION Area with Natural recolonization mangrove loss of planting MANGROVES TYPE 2 (DEGRADED) ENHANCEMENT Area with complete Hydrological LOST mangrove loss TYPE 2 restoration and natural colonization MANGROVES ENHANCEMENT of planting Artificial reefs and planting CORAL TYPE 2 REEFS ENHANCEMENT Areas with sandy TYPE 2 BEACHES beaches ENHANCEMENT Nourishing to keep with SLR SOURCE: Original figure for this publication. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: HOW IS THE NBSOS APPLIED? 33 ES MAP 8: OPPORTUNITIES FOR MANGROVES, VITI LEVU, FIJI 1 2 3 4 MANGROVES 5 Protection of exisiting 6 mangroves A1 Mangroves creation A2 opportunity SOURCE: Original map for this publication based on data from the NBSOS. NOTE: This is an example map of possible NBS interventions for mangrove ecosystems (mangrove areas to protect and potential areas for mangroves rehabilitation) in Viti Levu, Fiji. 3.2.3 STEP 3: MODELING NBS BENEFITS AND ESTIMATING COSTS IN COASTAL AREAS The benefits and costs of coastal NBS are valued by comparing the baseline situation to future scenarios in which the extent and condition of NBS is expected to change as a result either of degradation by anthropogenic action or of enhancement by NBS interventions (map 8). In those future scenarios, looking at the situation in 2050, the NBSOS accounts for projected sea-level rise, anticipated economic and demographic growth in coastal zones, expected ecosystem degradation, and potential effect of ecosystem enhancement by NBS. To value each coastal NBS individually, there is a comparison of scenarios where only one type of ©Cahe Viana_Unsplash ecosystem is enhanced while the others degrade. 34 CHAPTER: HOW IS THE NBSOS APPLIED? THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES FIGURE 8: DESCRIPTION OF NBS SCENARIOS USED IN THE BENEFITS ASSESSMENT 1 2 3 BUSINESS AS USUAL DEGRADATION OF NBS 4 BASELINE (- 20% hydraulic performance, reduction in area of reefs and 5 2020 CONDITIONS mangroves, and coastal erosion of beaches) 6 A1 PROTECTION PROTECTION OF NES A2 (same hydraulic performance and area cover for all ecosystems - as of 2020) BEACH ENHANCEMENT ENHANCEMENT OF BEACHES (same hydraulic performance for beaches - as of 2020, while other ecosystems degrade, no coastal erosion) REEF ENHANCEMENT ENHANCEMENT OF REEFS (+ 20% hydraulic performance for reefs, while other ecosystems degrade. Increase in cover of 20%) MANGROVE ENHANCEMENT ENHANCEMENT OF MANGROVES (+ 20% hydraulic performance for mangroves and increase in cover in suitable areas, while other ecosystems degrade) WHAT IS THE RELATIVE EFFECT OF 2050: Sea-level rise and extreme water levels DIFFERENT ECOSYSTEMS? projected (ERA-5, COAST-RP), demographic and Since there are uncertainties on how ecosystems economic growth in line with GDP growth projected in will evolve in the next decades, and to facilitate SSP2 (IPCC). comparability between different NBS, this assessment compares the effect of losing and gaining 20% of the performance of each ecosystem, due to degradation and NBS enhancement, respectively. At each coastal segment, it is thus possible to estimate which ecosystem units provide most ESS. SOURCE: Original figure for this publication. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: HOW IS THE NBSOS APPLIED? 35 ES FLOOD RISK REDUCTION BENEFITS 1 Annual expected damages from coastal flooding flooding. Avoided damages as a result of the presence 2 are estimated using a probabilistic flood risk of coastal NBS are estimated for the scenarios 3 assessment. Building on the inundation maps presented in figure 8, expected damages between the 4 developed with the SFINCS model and building on scenarios. Map 9 shows a graphical representation of 5 the flood exposure analysis described in Step 1, the modeled inundation of coastal areas for different 6 annual direct damages to buildings are calculated NBS scenarios. A1 (see appendix B for specifications) due to coastal A2 MAP 9: COASTAL FLOOD RISK REDUCTION OF NBS, VITI LEVU AND VANUA LEVU, FIJI Including enhancement of ecosystems No change in ecosystems Including degradation of ecosystems SOURCE: Original map for this publication based on data from the NBSOS. NOTE: This is an example map of a flood event caused by a 100-year storm in 2050 for different scenarios of landscape change between 2020 and 2050 in Viti Levu and Vanua Levu, Fiji. TOURISM, LIVELIHOODS, AND CLIMATE MITIGATION ESVD, no date) are used in regression analyses to BENEFITS estimate functions that relate the value of fisheries In addition to the flood and climate risk reduction (for mangroves and tourism) and tourism (only for benefits of coastal NBS, they provide other coral reefs) services to the characteristics and critical benefits for development, including solid context of the site. These functions are applied to contributions to nature-based tourism, food predict location-specific benefits while considering security, and climate mitigation. For coral reefs size of the ecosystem, the population density of the and mangroves, estimations from the Ecosystem area, and the income of the beneficiaries. Services Valuation Database (Brander et al. 2023; 36 CHAPTER: HOW IS THE NBSOS APPLIED? THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES The value of beach conservation for tourism is per unit area of 6.3 tonnes of carbon dioxide per 1 estimated using the sum of tourists’ willingness hectare per year (CO2 /ha/year). Carbon stored per 2 to pay for beach nourishment, plus a share of the year are valued economically using the social cost of 3 4 added value of hotel and accommodation revenues. carbon (SCC), which is the monetary value of damages 5 The NBSOS distributes the tourism value of beaches caused by emitting 1 more tonne of CO2 in a given year. 6 using the beach visitation function, based on the A1 distance of beaches to hotels. The hotel locations are Estimations of the per hectare costs of mangrove A2 obtained from Booking.com. For the NBSOS, beach restoration are obtained from Bayraktarov tourism values are quantified using local tourism et al. (2016) and Su, Friess, and Gasparatos (2021); data on number of tourists and average duration of estimations of the costs of coral reef restoration stay; these data usually come from the country’s are obtained from Bayraktarov et al. (2016). For both department of tourism. coral reef restoration and mangrove restoration, regression analysis is used to adjust cost estimations Blue carbon sequestration benefits are estimated of restoration per unit area to local PPP per capita for enhanced or restored mangrove areas. GDP. The cost of beach nourishment is obtained from Computation of additional carbon sequestration from Spencer, Strobl, and Campbell (2022). NBS multiplies the cumulative additional mangrove area by a representative carbon sequestration rate 3.2.4 STEP 4: DECISION SUPPORT FOR COASTAL NBS The cost-benefit analysis gives a first indication of the economic viability of investing in NBS for coastal climate resilience and its spatial variation across the area of interest. In the cost-benefit analysis, all benefits and costs are expressed as present value, discounted over a 28-year project lifetime. In the output of the NBSOS, the costs and benefits are expressed per hectare for Seychelles mangroves, per square meter for beach conservation, ©Hashimoto and per linear meter for coral reef interventions. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: HOW IS THE NBSOS APPLIED? 37 ES The NBSOS shows the results of the cost-benefit coastal section. A benefit-to-cost ratio of >1 indicates 1 analysis in a spatially explicit way to guide Task a positive return on investment. This, as shown in 2 Teams and clients what are potentially viable map 9, helps to identify the most viable interventions 3 locations to invest. Benefit-to-cost ratios, which within a country considering the integrated value of 4 are obtained by dividing the sum of all benefits by the flood reduction and other benefits such as increases 5 costs, are calculated and mapped per 2-kilometer to fisheries, carbon sequestration, and tourism. 6 A1 A2 MAP 9: BENEFITS AND COSTS OF NBS, VITI LEVU, FIJI SOURCE: Original map for this publication based on data from the NBSOS. BENEFIT-TO-COST RATIO NOTE: This is an example of the benefit-to-cost ratio of <1 mangrove interventions in Viti 1-3 Levu, Fiji. Benefits included are 3-5 flood reduction, fisheries and raw >5 material extraction, and blue carbon sequestration. 38 CHAPTER: HOW IS THE NBSOS APPLIED? THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES The contribution of flood protection, fisheries, Figure 9 panel b shows that, on average, mangrove 1 and carbon sequestration benefits to the restoration has a positive net present value, but that 2 3 overall benefit-cost ratio varies across coastal most of the flood protection value is concentrated in 4 sections (figure 9). 46 sites with a high coastal flood risk reduction value. 5 At these sites, most flood damage can potentially be 6 avoided, while other benefits or ecosystem services A1 are also provided. A2 FIGURE 9: BENEFITS AND COSTS OF MANGROVE INTERVENTIONS PER HECTARE (PRESENT VALUE) IN VITI LEVU A B AVERAGE FOR ALL COASTAL SECTIONS AVERAGE FOR SECTIONS WITH HIGH MANGROVE FLOOD PROTECTION BENEFITSa $61,000 $60k PRESENT VALUE PER HECTARE $40k (US$, THOUSANDS) $20k $6,100 $0k -$20k raw materials raw materials Fisheries and Fisheries and Blue carbon Blue carbon Flood Flood Cost Cost NPV NPV COST FLOOD BENEFITS CO-BENEFITS NPV SOURCE: Original figure for this publication. NOTE: NPV = net present value. a. These are the 46 sections in Viti Levu with 20 percent highest flood protection. This cost-benefit analysis As the NBSOS runs on global data sets, there is uncertainty represents the first screening in both cost and benefit estimations at different levels. Hence, further studies are needed to test concept designs to support strategic planning and to understand the feasibility of potential investments in and site prioritization. coastal NBS for climate resilience. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: HOW IS THE NBSOS INFORMING DEVELOPMENT FINANCE? 39 ES 1 2 3 4 4 5 6 A1 HOW IS THE A2 NBSOS INFORMING DEVELOPMENT FINANCE? ©David Cutler The NBSOS was successfully implemented in 20 countries between 2022 and April 1, 2024, including in coastal landscapes in 8 countries and In 14 of the applications, the NBSOS was used to identify investments in in 51 cities across NBS for climate resilience for an IPF operation. The combined financing 14 countries committed to these 14 operations amounts to $2.3 billion; a share of this (table 1). financing amount will likely be allocated to NBS for climate resilience. In several other countries—such as India, Sri Lanka, Thailand, and Timor- Leste—the NBSOS has been used to inform the dialogue on NBS with the government upstream through advisory services and analytics (ASAs). In addition, in Belize, the Eastern Caribbean countries, Senegal, Sri Lanka, and the West-Bank and Gaza, the NBSOS was used to estimate investment needs in NBS for adaptation for their respective Country Climate and Development Reports (CCDRs),6 integrating NBS in the economies’ long-term strategies for climate adaptation. 6 Information about the Country Climate and Development Reports can be found at https://www.worldbank.org/en/publication/country- climate-development-reports. 40 CHAPTER: HOW IS THE NBSOS INFORMING DEVELOPMENT FINANCE? THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES In May 2022, the urban NBSOS was used to inform forests would reduce exposure to extreme heat. 1 the N’Djamena Urban Resilience Project—a The findings of the NBSOS in N’Djamena have been 2 $150 million International Development Association integrated as investment options in (pre-)feasibility 3 4 (IDA) investment project—by identifying the areas studies for drainage infrastructure and solid waste 5 in which NBS contribute to climate resilience and investment. Furthermore, the project will aim to 6 reduce exposure of vulnerable populations to expand and consolidate a virtuous cycle for the A1 flooding and extreme heat. The NBSOS estimated resilience of green local initiatives already present A2 NBS investment opportunity areas for urban parks in the city, such as urban agriculture, local food and forests, urban agriculture, green corridors, markets, production organic composting, and tree and river floodplain measures. For urban parks and nurseries. Finally, following the NBSOS, the project forests, the NBSOS identified nearly 400 hectares is measuring its results in terms of successful NBS, of investment opportunity areas contributing tracking the percentage of the population living to a reduction of pluvial flood exposure, while within a 2-kilometer radius of an effective NBS for 285 hectares were identified where urban parks and climate resilience. In St. Lucia, an integrated coastal and urban NBSOS was used to assess investment needs in NBS for adaptation for the CCDR and to support the identification of NBS investments options for the forthcoming St. Lucia Flood Risk Project (table 1). The NBSOS identified potential investments in NBS to mitigate pluvial and fluvial food as well as extreme heat in the towns Castries and Anse La Raye, the area of interest of a new IPF $20 million operation. Forty hectares were found to be potentially suitable for a flood detention area or as an urban park with a water storage function, while 50 hectares of green corridors were mapped with the potential to highly reduce the exposure of people to extreme heat. After GPNBS held an NBS workshop in September 2023, the World Bank received a request to finance a new investment operation on flood risk reduction in St. Lucia, likely with a component on nature-based flood management interventions complementing gray infrastructure investment. The coastal NBS, which informed the CCDR, estimated spatially explicit benefit-cost ratios of potential investment in beach conservation and nourishment, mangroves, and coral reefs to protect St. Lucia against coastal flooding, sea-level rise, and © Simone Lee increasing extreme water levels. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: HOW IS THE NBSOS INFORMING DEVELOPMENT FINANCE? 41 ES TABLE 1: DETAILS OF NBSOS DELIVERED BETWEEN MID-2022 AND APRIL 1, 2024 1 2 COUNTRY PROJECT NBSOS APPLICATION GP/PG TYPE OF FINANCING FINANCING 3 CODE ENGAGEMENT OF IPF (US$, SOURCE IPF 4 MILLIONS) 5 6 Argentina P178534 NBSOS in 4 secondary Water IPF IBRD 200 A1 cities A2 Belize P181064 NBSOS in Belize City ENB, IPF IBRD/RETF 16.5 and country’s coastline GPURL P501812 NBSOS Belize coastline ENB CCDR Burkina Faso P177918 NBSOS in 3 secondary Transport, IPF IDA 200 cities GPURL Central African P178774 NBSOS in Bangui and 3 GPURL IPF IDA Republic secondary cities 70 Chad P177044 NBSOS in N’Djamena GPURL IPF IDA 150 P177163 NBSOS in 3 secondary GPURL IPF IDA 140 cities Côte d’Ivoire P168308, NBSOS in Abidjan GPURL IPF IDA 315 P177062 NBSOS and 8 GPURL IPF IDA 300 secondary cities Eastern P179742 NBSOS in coastal areas LAC SD CCDR n.a. Caribbean front n.a. Countriesa office Fiji P181433 NBSOS in coastal areas ENB IPF CIF/RETF/ in Viti Levu and Vanua 37 IDA Levu India P502683 NBSOS in 5 cities GPURL ASA n.a. n.a. Mali P171658 NBSOS in Bamako GPURL IPF 250 IDA Continues in the next page → SOURCE: Original table for this publication. NOTE: The pipeline beyond April 1, 2024, is not included in this table. ASA = advisory services and analytics; CCDR = Country Climate and Development Reports; CIF = Climate Investment Funds; ENB = Environment, Natural Resources and the Blue Economy Global Practice; GPURL = Urban, Disaster Risk Management, Resilience and Land Global Practice; IDA = International Development Association; IPF = investment project financing; LAC SD = Latin America and the Caribbean Sustainable Development Practice Group; n.a. = not applicable; NBSOS = Nature-Based Solutions Opportunity Scan; RETF = recipient-executed trust fund; SD = Sustainable Development Practice Group. a. Eastern Caribbean Countries in this analysis include Dominica, Grenada, St. Lucia, and St. Vincent and the Grenadines. 42 CHAPTER: HOW IS THE NBSOS INFORMING DEVELOPMENT FINANCE? THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES 1 COUNTRY PROJECT NBSOS APPLICATION GP/PG TYPE OF FINANCING FINANCING 2 CODE ENGAGEMENT OF IPF (US$, SOURCE IPF MILLIONS) 3 4 Nepal P163418 NBSOS in Itahari GPURL IPF 116 IDA 5 6 St. Lucia P503961 NBSOS in 2 cities and GPURL IPF IDA A1 in the island’s coastal 20 A2 areas Senegal P175830 NBSOS in Dakar and GPURL IPF IDA 290 Thies P180943 NBSOS in Dakar and ENB, SD CCDR n.a. n.a. Thies Somalia P170922 NBSOS in 4 cities GPURL IPF 154 IDA/RETF Sri Lanka P176456 NBSOS in coastal areas ENB ASA n.a. n.a. P500980 NBSOS in coastal areas ENB CCDR n.a. n.a. Thailand P178093 NBSOS in 4 cities Water ASA n.a. n.a. Timor-Leste P178790 NBSOS in coastal areas ENB ASA n.a. n.a. West Bank and P179452 NBSOS in 7 cities GPURL CCDR n.a. n.a. Gaza SOURCE: Original table for this publication. NOTE: The pipeline beyond April 1, 2024, is not included in this table. ASA = advisory services and analytics; CCDR = Country Climate and Development Reports; CIF = Climate Investment Funds; ENB = Environment, Natural Resources and the Blue Economy Global Practice; GPURL = Urban, Disaster Risk Management, Resilience and Land Global Practice; IDA = International Development Association; IPF = investment project financing; LAC SD = Latin America and the Caribbean Sustainable Development Practice Group; n.a. = not applicable; NBSOS = Nature-Based Solutions Opportunity Scan; RETF = recipient-executed trust fund; SD = Sustainable Development Practice Group. a. Eastern Caribbean Countries in this analysis include Dominica, Grenada, St. Lucia, and St. Vincent and the Grenadines. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: SUSTAINABILITY AND OPERATIONAL MODEL 43 ES 1 2 3 5 4 5 6 A1 SUSTAINABILITY A2 AND OPERATIONAL Urban farming 2 MODEL © Unsplash The NBSOS is a service offered to Task Teams to identify opportunities for NBS investments in the early stages of projects, using a small core team of staff and consultants in GFDRR. Currently, the NBSOS team Following a request from a Task Team, a two-page has the capacity to deliver the NBSOS in about 50 cities and 10 1 proposal is prepared. This includes tasks, deliverables, coastal landscapes per fiscal and level of effort/budget. year. The application follows a streamlined process: The proposal is fine-tuned in an inception meeting to determine the precise area of interest, climate resilience challenges of interest, the NBS types 2 of interest, and the potential integration of locally collected data. A short inception note is prepared based on the meeting. Once the inception note is cleared by the Task Team, 3 the NBSOS analysis is conducted by the GPNBS team and deliverables are prepared. Results are presented to the Task Team and final 4 deliverables—a PowerPoint deck and a geospatial data package that is available upon request—are shared. 44 CHAPTER: SUSTAINABILITY AND OPERATIONAL MODEL THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN The demand for the NBSOS ES Among these developments is the increased automatization of 1 2 is increasing rapidly, the process to reduce the number of input hours required for each application. Furthermore, sustainability risks are mitigated 3 4 and developments are by the good archiving of data, by documentation, and by having 5 continually being made to multiple consultants in the team who can perform the tasks 6 increase the efficiency and needed to obtain coastal and urban NBSOS results. The team also collaborates with a wider team of GPNBS experts who can quality of the service. A1 A2 support Task Teams on the use of these results for project design and implementation. Nonetheless, the capacity for deployment is not limitless, and NBSOS applications need to be planned carefully to ensure timely delivery for effective operational use of the outcomes. The NBSOS runs primarily in Python,7 with data and To ensure access to both recent and historical end products stored in Google Cloud Storage (GCS). Earth observation (EO) data, the EO data is obtained The NBSOS input data sets are characterized by with Google Earth Engine (GEE). GEE provides globally comprehensive spatial coverage, allowing the access to a multi-petabyte catalogue of EO data, NBSOS to be implemented in any region of the world. ranging from raw satellite observation bands to Global data sets are processed for an area of interest analysis-ready land cover products. GEE’s Python (AOI) representing the spatial extent of analysis, Application Programming Interface (API) allows which is defined with Task Teams for each case. As seamless integration of data acquisition in the NBSOS the availability, detail, and breadth of global data Python framework. Most of the analysis in NBSOS sets continue to increase, so will the functionality of is done in Python, except the heat module and map the NBSOS. visualization, which run in R software.8 The use of open data and open software makes it possible to turn the NBSOS into an open-source tool in the future. GPNBS has the ambition to make parts of the source code available on GitHub upon publication of the methodology in a peer-reviewed academic journal.9 7 Python is a programming language. It allows you to work quickly and lets you integrate systems effectively. For more information about Python, see https://www.python.org/. 8 R is a free software environment appropriate for statistical computing and graphics. More information can be found at https://www.r-project.org/. 9 More details about GitHub can be found at https://github.com/. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: NEXT STEPS AND EVOLUTION OF THE NBSOS 45 ES 1 2 3 6 4 5 6 A1 NEXT STEPS AND A2 EVOLUTION OF © Anita Denunzio_Unsplash THE NBSOS There are two main areas for next steps for the NBSOS. The first involves the improvement of NBSOS’ analytical capability; the second involves increasing its capacity. This section looks briefly at each. 6.1 THE DEVELOPMENT OF ANALYTICAL CAPABILITIES Analytical capabilities of the NBSOS are—and will pluvial flood reduction by NBS. Combining spatial continue to be—constantly improved and tested species diversity data from tools such as the during NBSOS applications. Such incremental Integrated Biodiversity Assessment Tool (IBAT) and ongoing improvements include remote sensing–based habitat connectivity models,10 the GPNBS team observations of coastal erosion, erosion modeling is working with partners to design a biodiversity along inland water bodies, and suitability mapping of module for the NBSOS. In addition, the GPNBS team green roofs in dense urban areas. Such improvements is developing a model that can provide a quantitative rely on existing scientific approaches and are typically estimate of flood hazard and exposure reduction tested in two or three applications before being as a result of NBS investment options. The current integrated into the main workflow. model identifies in which areas NBS have the highest potential of reducing flood exposure, whereas the More fundamental analytical improvements will updated model will calculate flood extent and flood focus on expanding the capability of the NBSOS depth both with and without NBS. estimating biodiversity impact and quantifying 10 More information about IBAT can be found at https://www.ibat-alliance.org/. 46 CHAPTER: NEXT STEPS AND EVOLUTION OF THE NBSOS THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES A growing body of research and As more NBS for climate resilience are studied and 1 practice will increase the ability to implemented globally, more data points on costs become 2 3 estimate unit costs of NBS adjusted to available. In collaboration with GFDRR and PROGREEN, GPNBS 4 the landscape or country context as is developing rapid costing tools to inform and improve unit cost estimations used in the NBSOS. 5 part of the NBSOS. 6 A1 A2 6.2 TOWARD MORE CAPACITY FOR DELIVERY Within the current operational model, it is possible to further increase capacity responding to additional demand from World Bank Task Teams and beyond. Increased efficiency due to standardization, along with a larger pool of expert consultants to run the analyses, can expand the capacity for coastal and urban NBSOS in fiscal year 2025 by 25–50 percent of current capacity, which might be sufficient to meet demand in the World Bank. If the demand exceeds this capacity, it will lead to the requests of Task Teams needing a longer response time. Several options can be considered to further expand capacity to deliver NBS investment opportunity recommendations. Capacity could be increased even The former option comes with risks, including the sustainability of maintaining such a webtool with more by developing a webtool graphical user interface as well as a potential lack of with graphical user interface expertise among users to interpret and use the results for nonexpert users. It could to identify investments. The latter option—stimulating industry—might be a more promising avenue. Since also be increased by stimulating a pre-publication on the NBSOS was released in development partners and July 2023 (GFDRR and ESA 2023), about 20 external industry (that is, World Bank partners—including multilateral development banks (MDBs), other development partners, private sector vendors) to adopt the NBSOS firms, and leading civil society organizations (CSOs)— methods by opening the code and have requested access to the NBSOS source code to enabling knowledge transfer. adopt the methodology for investment identification. Piloting external dissemination, the World Bank aims to start sharing the source code of the urban NBSOS on request with a limited number of external partners upon publication of this report. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN 47 REFERENCES Bayraktarov, Elisa, Megan I. Saunders, Sabah Abdullah, Morena Earth Engine Data Catalog. No date-f. iSDAsoil Organic Carbon Mills, Jutta Beher, Hugh P. Possingham, Peter J. Mumby, and (data set). Provider iSDA. https://developers.google.com/earth- Catherine E. Lovelock. 2016. “The Cost and Feasibility of Marine engine/datasets/catalog/ISDASOIL_Africa_v1_carbon_organic. Coastal Restoration.” Ecological Applications 26 (24): 1055–74. https://doi.org/10.1890/15-1077. Earth Engine Data Catalog. No date-g. OpenLandMap Soil Texture Class (USDA System) (data set). Provider EnvironmetriX Ltd. Brander, L. M., R. de Groot, V. Guisado Goñi, V. van ‘t Hoff, P. https://developers.google.com/earth-engine/datasets/catalog/ Schägner, S. Solomonides, A. McVittie, F. Eppink, M. Sposato, OpenLandMap_SOL_SOL_TEXTURE-CLASS_USDA-TT_M_v02. L. Do, A. Ghermandi, and M. Sinclair. 2023. Ecosystem Services Valuation Database (ESVD). Foundation for Sustainable ESVD. No date. Ecosystem Services Valuation Database Development and Brander Environmental Economics. (ESVD). Foundation for Sustainable Development and Brander Environmental Economics. https://www.esvd.net/. Bunting, Pete, Ake Rosenqvist, Lammert Hilarides, Richard M. Lucas, Nathan Thomas, Takeo Tadono, Thomas A. Worthington, FABDEM (Forest And Buildings removed Copernicus 30m DEM). Mark Spalding, Nicholas J. Murray, and Lisa-Maria Rebelo. 2022. No date. Awesome-gee-community-catalog. https://gee- “Global Mangrove Extent Change 1996–2020: Global Mangrove community-catalog.org/projects/fabdem/. Watch Version 3.0.” Remote Sensing 14 (15): 3657. https://doi. org/10.3390/rs14153657. FCL and CIESIN (Facebook Connectivity Lab and Center for International Earth Science Information Network), Columbia Burkhard, Benjamin, Franziska Kroll, Stoyan Nedkov, and Felix University. 2016. High Resolution Settlement Layer (HRSL). Müller, 2012. “Mapping Ecosystem Service Supply, Demand and Source imagery for HRSL, copyright DigitalGlobe. https:// Budgets,” Ecological Indicators 21: 17–29, https://doi.org/10.1016/j. dataforgood.facebook.com/dfg/tools/high-resolution- ecolind.2011.06.019. settlement-layer. Earth Engine Data Catalog. No date-a. CHIRPS Pentad: Climate Fathom. 2019. Fathom-global 2.0. https://www.fathom.global/ Hazards Group InfraRed Precipitation With Station Data (Version product/global-flood-map/. 2.0 Final) (data set). Provider USCB/CHG. https://developers. google.com/earth-engine/datasets/catalog/UCSB-CHG_ Fathom. 2023. Fathom-global 3.0. https://www.fathom.global/ CHIRPS_PENTAD. product/global-flood-map/. Earth Engine Data Catalog. No date-b. Dynamic World V1 (data GFDRR (Global Facility for Disaster Reduction and Recovery). set). Provider World Resources Institute Google. https:// 2023. GFDRR Annual Report 2023: Bringing Resilience to Scale. developers.google.com/earth-engine/datasets/catalog/ Washington, DC: World Bank. GOOGLE_DYNAMICWORLD_V1. GFDRR and ESA (Global Facility for Disaster Reduction and Earth Engine Data Catalog. No date-c. ESA WorldCover 10m v100 Recovery and European Space Agency). 2023. “The Nature-Based (data set). Provider ESA/VITO/Brockmann Consult/CS/GAMMA Solutions Opportunity Scan: Leveraging Earth Observation Data Remote Sensing/IIASA/WUR. https://developers.google.com/ to Identify Investment Opportunities in NBS for Climate Resilience earth-engine/datasets/catalog/ESA_WorldCover_v100. around the Globe.” GFDRR and ESA. https://naturebasedsolutions. org/opportunity-scan. Earth Engine Data Catalog. No date-d. JRC Global Surface Water Mapping Layers, v1.4 (data set). Provider EC JRC/Google. https:// Google Earth. 2023. Google Earth V 10.49.0.0. https://earth. developers.google.com/earth-engine/datasets/catalog/JRC_ google.com/. GSW1_4_GlobalSurfaceWater. Google Research. No Date. Dataset: Open Buildings: A Dataset of Earth Engine Data Catalog. No date-e. Harmonized Sentinel-2 Building Footprints to Support Social Good Applications. https:// MSI: MultiSpectral Instrument, Level-2A (data set). Provider sites.research.google/open-buildings/. European Union/ESA/Copernicus. https://developers.google. com/earth-engine/datasets/catalog/COPERNICUS_S2_SR_ HARMONIZED. 48 THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN Hawker, Laurence, Peter Huh, Luntadila Paulo, Jeison Sosa, Mukim, Megha, and Mark Roberts, editors. 2023. Thriving: Making James Savage, Christopher Sampson, and Jeffrey Neal. 2022. Cities Green, Resilient, and Inclusive in a Changing Climate. “A 30 m Global Map of Elevation with Forests and Buildings Washington, DC: World Bank. https://doi.org/10.1596/978-1-4648- Removed.” Environmental Research Letters 17 (2): 024016. https:// 1935-3. doi.org/10.1088/1748-9326/ac4d4f. OpenStreetMap contributors. 2024. Planet dump [2024]. https:// Hersbach, H., B. Bell, P. Berrisford, G. Biavati, A. Horányi, J. Muñoz planet.openstreetmap.org. Sabater, J. Nicolas, C. Peubey, R. Radu, I. Rozum, D. Schepers, A. Simmons, C. Soci, D. Dee, and J.-N. Thépaut. 2018. ERA5 hourly Sampson, Christopher C., Andrew M. Smith, Paul D. Bates, Jeffery data on single levels from 1940 to present. Copernicus Climate C. Neal, Lorenzo Alfieri, and Jim E. Freer. 2015. “A High-Resolution Change Service (C3S) Climate Data Store (CDS). https://doi. Global Flood Hazard Model.” Water Resources Research 51 (8)” org/10.24381/cds.adbb2d47. 7358–81. https://doi.org/10.1002/2015WR016954. Hooker, Josh, Gregory Duveiller, and Alessandro Cescatti. 2018. “A Sirko, Wojciech, Sergii Kashubin, Marvin Ritter, Abigail Annkah, Global Dataset of Air Temperature Derived from Satellite Remote Yasser Salah Eddine Bouchareb, Yann Dauphin, Daniel Keysers, Sensing and Weather Stations.” Scientific Data 5: 180246. https:// Maxim Neumann, Moustapha Cisse, and John Quinn. 2021. doi.org/10.1038/sdata.2018.246. “Continental-Scale Building Detection from High Resolution Satellite Imagery.” arXiv, Cornell University. https://arxiv.org/ IPCC (Intergovernmental Panel on Climate Change). 2021. abs/2107.12283. “Summary for Policymakers.” In Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Spencer, Nekeisha, Eric Strobl, and Alrick Campbell. 2022. Sixth Assessment Report of the Intergovernmental Panel on “Sea Level Rise under Climate Change: Implications for Beach Climate Change, edited by V. Masson-Delmotte, P. Zhai, A. Pirani, Tourism in the Caribbean.” Ocean & Coastal Management 225 (1): S.L. Connors, C. Péan, S. Berger, N. Caud, Y. Chen, L. Goldfarb, 106207. https://doi.org/10.1016/j.ocecoaman.2022.106207 . M.I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R. Matthews, T.K. Maycock, T. Waterfield, O. Yelekçi, R. Yu, and B. Zhou. In Press. Su, Jie, Daniel A. Friess, and Alexandros Gasparatos. 2021. https://www.ipcc.ch/report/ar6/wg1/downloads/report/IPCC_ “A Meta-Analysis of the Ecological and Economic Outcomes of AR6_WGI_SPM_final.pdf. Mangrove Restoration.” Nature Communications 12: 50. https:// doi.org/10.1038/s41467-021-25349-1. Jongman, Brenden, Gonzalo Gutiérrez Goizueta, Boris van Zanten, and Borja González Reguero. 2021. “Blue Barriers: A Nature-Based van Zanten, Boris, Gonzalo Gutierrez Goizueta, Luke Brander, Solution to Build Resilience.” Development and a Changing Climate Borja Gonzalez Reguero, Robert Griffin, Kavita Kapur Macleod, blog. https://blogs.worldbank.org/en/climatechange/blue- Alida Alves, Amelia Midgley, Luis Diego Herrera, and Brenden barriers-nature-based-solution-build-resilience . Jongman. 2023. Assessing the Benefits and Costs of Nature-Based Solutions for Climate Resilience: A Guideline for Project Developers. Konijnendijk, Cecil C. 2023. “Evidence-Based Guidelines for World Bank, Washington, DC. Greener, Healthier, More Resilient Neighbourhoods: Introducing the 3–30–300 Rule.” Journal of Forestry Research 34: 821–30. World Bank. 2021. A Catalogue of Nature-Based Solutions for https://doi.org/10.1007/s11676-022-01523-z . Urban Resilience. Washington, DC: World Bank. http://hdl.handle. net/10986/36507. Leijnse, Tim, Maarten van Ormondt, Kees Nederhoff, and Ap van Dongeren. 2021. “Modeling Compound Flooding in Coastal World Bank. 2023. Nature-Based Solutions for Climate Systems Using a Computationally Efficient Reduced-Physics Resilience in the World Bank Portfolio. World Bank. https:// Solver: Including Fluvial, Pluvial, Tidal, Wind- and Wave-Driven naturebasedsolutions.org/sites/default/files/2023-03/ Processes,” Coastal Engineering 163: 103796. https://doi. P1768250d0db6f0c80bb5b08c648e4d0f18.pdf. org/10.1016/j.coastaleng.2020.103796. Lyons, Mitchell B., Chris M. Roelfsema, Emma V. Kennedy, Eva M. Kovacs, Rodney Borrego-Acevedo, Kathryn Markey, Meredith Roe, Doddy M. Yuwono, Daniel L. Harris, Stuart R. Phinn, Gregory P. Asner, Jiwei Li, David E. Knapp, Nicholas S. Fabina, Kirk Larsen, Dimosthenis Traganos, and Nicholas J. Murray. 2020. “Mapping the World’s Coral Reefs Using a Global Multiscale Earth Observation Framework.” Remote Sensing in Ecology and Conservation 6 (4): 557–68. https://doi.org/10.1002/rse2.157. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN APPENDIX A: DETAILED METHODOLOGY OF THE URBAN NBSOS 50 THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN TABLE OF CONTENTS ABBREVIATIONS AND ACRONYMS 52 A.1 OVERVIEW OF PRODUCT, APPLICATION, AND METHODOLOGY 53 A.1.1 Overview 53 A.2 METHODOLOGY 54 A.2.1 Software/coding scheme 54 A.2.2 Mapping priority areas 55 A.2.3 Mapping NBS suitability 60 A.2.4 Mapping benefits 65 A.2.5 Mapping optimal allocation of NBS 73 A.3 APPLICATION OF THE NBSOS AS A SERVICE TO WORLD BANK TASK TEAMS 76 A.3.1 Task Team consultation 76 A.3.2 Interpretation, presentation, and recommendations 77 A.4 NBSOS VALIDATION: NBS FOR FLOOD REDUCTION 79 A.4.1 Introduction 79 A.4.2 Objectives 80 A.4.3 Description of local study used for comparison 80 A.4.4 Results 81 A.4.5 Conclusions 84 REFERENCES 85 THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN 51 TABLE OF CONTENTS MAPS MAP A.1: Flooded areas in Banjul, The Gambia, by Fathom v2 (upper maps), Fathom v3 (middle maps), and local hydrodynamic model (lower maps), for the cases of 20 years (left) and 100 years (right) return periods.....................................................................................................................82 MAP A.2: Comparison between areas in Banjul, The Gambia, identified by the NBSOS as high priority for NBS implementation (green) and areas for SUDS implementation recommended by the local study........... 83 MAP A.3: NBS protection areas in Banjul, The Gambia, defined by the NBSOS (purple), and in yellow, the green park suggested by the local study.......................................................................................................84 FIGURES FIGURE A.1: Identifying priority areas................................................................................................................. 55 FIGURE A.2: A framework to support the first identification of potential investments in NBS..........................60 FIGURE A.3: A hierarchy of approaches under the nature-based solutions umbrella......................................... 63 FIGURE A.4: NBSOS benefits and impacts.......................................................................................................... 65 BOXES BOX A.1:Decision support for NBS investments in urban areas.......................................................................... 75 TABLES TABLE A.1: Input data for pluvial flood mitigation modeling............................................................................... 56 TABLE A.2: Input data for extreme heat mitigation modeling............................................................................ 57 TABLE A.3: Input data for health, recreation, and social cohesion modeling..................................................... 58 TABLE A.4: NBS suitability ruleset......................................................................................................................61 TABLE A.5: Input data for NBS suitability modeling........................................................................................... 63 TABLE A.6: Input data for NBS opportunity classification.................................................................................64 TABLE A.7: Input data for flood mitigation benefit modeling............................................................................. 66 TABLE A.8: Land cover conversion and bare soil percentage change associated with each NBS family .......... 67 Table A.9: Example jobs and resources production weights for a subset of NBS families..................................71 Table A.1.1: Breakdown of NBS investment in Greater Dakar.............................................................................. 75 52 THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ABBREVIATIONS AND ACRONYMS ACP Africa Caribbean Pacific JRC Joint Research Centre AOI area of interest LCC Land Capability Classification API Application Programming Interface LST land surface temperature CAPEX capital expenditure MODIS Moderate Resolution Imaging COG Cloud Optimized GeoTIFF Spectroradiometer CN Curve Number NBS nature-based solutions CPI Consumer Price Index NBSOS Nature-Based Solutions Opportunity DEM Digital Elevation Model Scan EO Earth observation NDRR Natural Disaster Risk Reduction ESA European Space Agency NDVI Normalized Difference Vegetation Index GBA Greater Banjul Area, The Gambia OSM OpenStreetMap GCS Google Clouse Storage PPT PowerPoint GEE Google Earth Engine RP return period GFDRR Global Facility for Disaster Reduction RUSLE Revised Universal Soil Loss Equation and Recovery SUDS sustainable urban drainage systems IPCC Intergovernmental Panel on Climate TA technical assistance Change UFRMM Urban Flood Risk Mitigation Model THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX A 53 ES 1 2 3 A.1 4 5 6 A1 OVERVIEW OF PRODUCT, A2 APPLICATION, AND METHODOLOGY Urban forest ©Vickry Alvian_Unsplash A.1.1 OVERVIEW The Nature-Based Solutions Opportunity Scan (NBSOS) is a geospatial analysis and a participatory process that is offered as an on-demand service for nature-based solutions (NBS) investment opportunity mapping in cities around the world. It aims to support World Bank Task Teams, governments, and other investors to understand which NBS families have the most potential in a city, what potential project sites are, what their potential benefits are, and how NBS can complement gray infrastructure. The NBSOS methodology is designed in such a way that its deliverables—including results, interpretation, recommendations, and a geospatial data package— can be prepared in approximately six weeks. The NBSOS utilizes an array of openly available medium resolution (10- to 30-meter) Earth observation (EO) data and other geospatial data sets as inputs into an analytical workflow consisting of four methodological steps (see figure 4 in the main text for an overview of the NBSOS methodology). The first step is to understand the problem: what is the spatial distribution and magnitude of urban resilience and sustainability challenges and what are the solutions considered? The second step consists of mapping suitable areas for the NBS families considered. The third step models and estimates the positive impact of NBS to address the identified resilience and sustainability challenges. In the fourth step, optimal solutions are identified using multicriteria analysis to highlight solutions that maximize NBS co-benefits. The next section of this appendix describes the geospatial data sets and models that constitute the NBSOS (section A.2). It then outlines the participatory process and deliverables of the NBSOS as a service to World Bank Task Teams (section A.3). The last section describes a validation exercise for the NBSOS flood module (section A.4). 54 CHAPTER: APPENDIX A THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES 1 2 3 A.2 4 5 6 A1 A2 METHODOLOGY ©artiom-vallat_unsplash A.2.1 SOFTWARE/CODING SCHEME The NBSOS runs primarily in Python, with data and end products stored in Google Cloud Storage (GCS). In summary, the geospatial modeling component of the NBSOS compiles environmental indicators from global, publicly available EO and geospatial data sets to (1) prioritize regions of NBS implementation as a function of population exposure to pluvial flooding, extreme heat, and lack of access to green spaces; (2) identify NBS creation opportunities for the NBS families described in the World Bank’s Catalogue of Nature-Based Solutions for Urban Resilience by applying conditional suitability rulesets on environmental indicators; (3) estimate the potential for NBS creation to provide flood mitigation, extreme heat mitigation, health/recreation benefits, and other benefits; and (4) provide decision support by identifying optimal NBS families through a multicriteria analysis. The NBSOS input data sets are characterized by globally comprehensive spatial coverage, allowing the urban NBSOS to be implemented in any urban region of the world. Global data sets are subset and processed for an area of interest (AOI) defining the spatial extent of analysis. A.2.1.1 GOOGLE EARTH ENGINE The NBSOS highlights priority intervention areas, identifies land suitable to support NBS creation, and quantifies potential benefits of NBS in cities by synthesizing publicly available EO data. To ensure access to both recent and historical EO data, EO data were obtained with Google Earth Engine (GEE). GEE provides access to a multi-petabyte catalogue of EO data, ranging from raw satellite observation bands to analysis-ready land cover products. GEE’s Python Application Programming Interface (API) allows seamless integration of data acquisition in the NBSOS Python framework. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX A 55 A.2.1.2 ES DATA PROCESSING 1 2 3 The EO data derived from GEE are used in the development of spatial raster layers describing environmental 4 indicators that form the basis of NBSOS’ subsequent modules. Examples of environmental indicators are the 5 Land Capability Classification Index (Quant et al. 2020), tree canopy cover (Brandt et al. 2023), and bare soil 6 frequency (Demattê et al. 2020) (further described in section A.2.3 on mapping NBS suitability). Analysis-ready A1 indicators are centrally stored as Cloud Optimized GeoTIFF (COG) spatial raster format files in GCS, where data A2 can be accessed via Python by each benefit module of NBSOS. All further analysis in NBSOS is done in Python, except for the heat module and map visualization, which run in R software. A.2.2 MAPPING PRIORITY AREAS Priority area mapping consists of FIGURE A.1: IDENTIFYING PRIORITY AREAS identifying high-impact areas for NBS investment within the AOI guided by two questions: (1) Where are hazard Determine where Determine where to exposures highest? And (2) Where can people have the apply NBS to solve it NBS be implemented to reduce the highest exposure to more effectively. exposure to the hazard most effectively? the hazard. Therefore, this first step focuses on identifying where the problems are (figure A.1). Identifying priority areas helps to understand where to act first to have the highest impact, helping to make a strategic long-term plan. SOURCE: Original figure for this publication. NOTE: NBS = nature-based solutions. A.2.2.1 PLUVIAL FLOODING To identify pluvial flood exposure, the annual flood probability per neighborhood, derived from Fathom Global Flood Model (Fathom, no date), is multiplied by the number of people living in each neighborhood of the AOI. Subsequently, the NBSOS implements a basic hydrological model to identify neighborhoods where NBS investment could have the highest impact in reducing pluvial flood exposure. Pluvial flood mitigation by NBS is operationalized in the NBSOS as the combination of runoff reduction through surface cover changes and storage potential of NBS. 56 CHAPTER: APPENDIX A THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES A.2.2.1.1 INPUT DATA AND PROCESSING TABLE A.1: INPUT DATA FOR PLUVIAL FLOOD MITIGATION MODELING 1 2 Table A.1 describes elements of 3 the input data used to model pluvial DATA TYPE DATA SET NAME AND RESOLUTION 4 SOURCE 5 flood mitigation. 6 Flood hazard data Fathom Pluvial Flood Risk 30 meters (Fathom, no date) A1 A2 Population data High Resolution Settlement 30 meters Layer (FCL and CIESIN 2016) Digital Elevation Model FABDEM (Hawker et al. 2022) 30 meters SOURCE: Original table for this publication. The NBSOS uses Fathom Global Flood Maps v2 to estimate annual flood probability per neighborhood in the target city. Fathom Global Flood Maps V2 is a proprietary global flood model that provides spatially explicit flood simulations for 10 different return periods at a 90-meter resolution. Fathom modeled flood outputs of flood depth in millimeters are simplified into a binary outcome per pixel of either flooded (> 0) or not flooded (0). Annual flood risk per pixel (Fa,i) is calculated as where R is the return period in years and (FR,i) is the binary flood outcome for pixel i for return period R. The mean annual flood probability for each neighborhood is then multiplied by the total population of each neighborhood to estimate a neighborhood annual flood exposure value. Neighborhood population is determined by summing the rasterized population using High Resolution Settlement Layer population data per neighborhood (FCL and CIESIN 2016). Unless relevant neighborhood administrative boundaries are provided by Task Teams, neighborhood subregions of the AOI are created using the Hexagonal Hierarchical Geospatial Indexing System (Sahr 2020), which produces hexagonal grids of approximately 0.75 square kilometers per subregion. A.2.2.1.2 ESTIMATION OF FLOOD PRIORITY AREAS To account for the downstream effect of runoff reduction, the NBSOS uses a series of watershed delineations. Using the Python library Pysheds (Bartos 2020), first the flow accumulation per pixel is determined, describing the number of additional pixels directly upstream of that pixel. For each pixel with an accumulation flow greater than 500 pixels, the watershed (that is, the area that accumulates to that pixel) is calculated. To account for the inaccuracies of flood hazard data in urban areas, watersheds are generalized per neighborhood by combining the watersheds of each pixel per neighborhood to create a spatially explicit neighborhood watershed that gets assigned the flood hazard exposure value of its original neighborhood. Finally, the flood hazard exposure values of all neighborhood watersheds are summed up to obtain a raster that shows a relative distribution of population that can be affected per pixel accounting for flood probability. This raster is averaged over neighborhoods to estimate the flood priority areas, representing high-impact regions to apply NBS to reduce downstream hazard exposure (see map 1 of the main text for an example). THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX A 57 A.2.2.2 ES HEAT STRESS 1 2 3 A.2.2.2.1 INPUT DATA DESCRIPTION AND PROCESSING 4 5 Table A.2 describes the data sets used for modeling extreme heat mitigation. 6 A1 A2 TABLE A.2: INPUT DATA FOR EXTREME HEAT MITIGATION MODELING DATA TYPE DATA SET NAME AND SOURCE RESOLUTION Land surface MODIS Land Surface Temperature (Wan, Hook, and Hulley 2021) 1 kilometer temperature data Ambient air A global data set of air temperature derived from remote sensing and 1 kilometer temperature data weather stations (Hooker, Duveiller, and Cescatti 2018) Population data High Resolution Settlement Layer (FCL and CIESIN 2016) 30 meters SOURCE: Original table for this publication. To estimate spatial variability in heat exposure within the AOI, global spatial data sets on average annual air temperature and population density are combined. Average annual air temperature at 10-meter spatial resolution is estimated using the methodologies presented in Hooker, Duveiller, and Cescatti (2018), who describe a statistical model derived by comparing remotely sensed land surface temperature (LST) and ambient air temperature observed at weather stations at two meters above the land surface that incorporates information on geographic and climatic similarity. First, the five most recent years of daily daytime and nighttime LST rasters at a 1-kilometer spatial resolution from the Moderate Resolution Imaging Spectroradiometer (MODIS) (Wan, Hook, and Hulley 2021) are acquired and aggregated to a monthly timestep for a 5-kilometer buffered region around the AOI using GEE. The 1-kilometer monthly LST rasters are used as inputs in the geographically weighted and climate space weighted regression models described in Hooker, Duveiller, and Cescatti (2018), resulting in two distinct data sets of 1-kilometer resolution monthly air temperature. To obtain more accurate air temperature predictions, the two regression model outputs are combined using the stacked generalization weighting coefficients presented in Table 1 of Hooker, Duveiller, and Cescatti (2018), resulting in 12 rasters of estimated average monthly air temperature. Average annual air temperature is estimated by taking the mean value for each pixel across the 12 monthly air temperature rasters. To match the spatial resolution of the NBSOS, the average annual air temperature raster is resampled to 10-meter spatial resolution using cubic interpolation and cropped to the AOI. Population density is estimated using the High Resolution Settlement Layer (FCL and CIESIN 2016), which provides estimates of global human population distribution at approximately 30-meter spatial resolution by combining recent census data with high-resolution satellite imagery. After resampling the population data to the 10-meter native resolution of the NBSOS, a moving window Gaussian spatial weighting function is applied to each pixel within the AOI such that each pixel value represents the weighted population count within 1 kilometer 58 CHAPTER: APPENDIX A THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES of the pixel, where per pixel population counts closer to each focal pixel are weighed more heavily than per pixel 1 population counts further from each focal pixel. 2 3 4 To capture spatial variability within the AOI, mean annual air temperature and population density rasters are 5 aggregated to neighborhood subregions within the AOI. For each subregion, the mean annual air temperature 6 and sum of the population count for all raster pixels within the subregion are used to determine priority areas. A1 A2 A.2.2.2.2 ESTIMATION OF HEAT PRIORITY AREAS Heat priority areas are estimated at the subregion scale using an index calculated as: where AirTi represents the mean annual air temperature within each subregion i and Populationi represents the spatially weighted population count within each subregion i (see map 1 in the main text). A.2.2.3 HEALTH, RECREATION, AND SOCIAL COHESION A.2.2.3.1 INPUT DATA AND PROCESSING Table A.3 describes the data sets used as inputs for modeling health, recreation, and social cohesion benefits. TABLE A.3: INPUT DATA FOR HEALTH, RECREATION, AND SOCIAL COHESION MODELING DATA TYPE DATA SET NAME AND SOURCE RESOLUTION Existing green space Derived from bare soil frequency (Demattê et al. 2020) and productivity 10 meters data performance indicator (Quandt et al. 2020) (see sections 2.3.2.1 and 2.3.3.1) Population data High Resolution Settlement Layer (FCL and CIESIN 2016) 30 meters SOURCE: Original table for this publication. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX A 59 ES The health, recreation, and social cohesion benefits of urban nature relate to the proximity of nature to 1 residents. Novel, evidence-based guidelines for greener, healthier cities suggest that all urban residents 2 should be within 300 meters of an open green space (Konijnendijk 2023). Thus, neighborhoods with less green 3 space and a high population of residents greater than 300 meters from the nearest open green space represent 4 high-priority regions for NBS implementation. To identify spatial variability in priority areas for health, 5 recreation, and social cohesion benefits from NBS creation, data on population density are combined with the 6 size and location of existing green spaces. A1 A2 Population density is estimated using the High Resolution Settlement Layer (FCL and CIESIN 2016), which provides estimates of global human population distribution at approximately 30-meter spatial resolution by combining recent census data with high-resolution satellite imagery. Data on the size and location of existing green spaces come from the “protection” layer described in sections 2.3.2.1 and 2.3.3.1, where existing green spaces are identified based on recent trends in bare soil frequency and Earth observation greenness indexes. Existing green spaces providing health and recreation benefits are defined as contiguous green spaces greater than 1 hectare in area. Existing green spaces providing social cohesion benefits are defined as contiguous green spaces greater than 0.5 hectare in area (WHO 2016). To estimate the number of residents farther than 300 meters from existing green spaces, the existing green spaces greater than 1 hectare and those greater 0.5 hectares are vectorized and buffered by 300 meters. The sum of the population counts falling outside of the 300-meter buffered existing green space vectors is calculated for each subregion within the AOI to identify priority areas. A.2.2.3.2 ESTIMATION OF HEALTH, RECREATION, AND SOCIAL COHESION PRIORITY AREAS The health and recreation priority areas correspond to the number of residents who are more than 300 meters from contiguous green spaces greater than 1 hectare in area. The social cohesion priority areas correspond to the number of residents who are more than 300 meters from contiguous green spaces with an area greater than 0.5 hectares. In both cases, higher population counts that are more than 300 meters from an existing green space correspond to a higher-priority area value (see map 1 in the main text). A.2.2.4 VISUALIZATION AND MAPPING Priority areas for each hazard are provided in a series of maps created using R. Priority index values for each hazard of interest and subregion i are normalized by the equation: such that the index describes values from 0 to 1, where higher values correspond to neighborhood subregions with a higher-priority index value. The AOI subregions are then categorized into five groups (Very Low, Low, Medium, High, Very High) using the Jenks natural breaks clustering algorithm to minimize variability within priority categories and maximize variability across priority categories. The priority map for each benefit depicts the categorical index values ranging from Very Low to Very High for each subregion within the AOI (see map 1 of the main text for an example of mapping priority areas). 60 CHAPTER: APPENDIX A THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES 1 2 A.2.3 3 4 MAPPING NBS SUITABILITY 5 6 The NBSOS suitability mapping is a two-stage analysis to isolate regions within the AOI where NBS could A1 be implemented. First, areas within the AOI deemed suitable for supporting NBS based on environmental A2 properties (for example, slope, soil properties, land cover) are identified (section A.2.3.1) (figure A.2). Second, the areas with the capacity to support NBS are further analyzed using EO data on bare soil frequency and greenness to categorize areas as existing greenery to protect or degraded land representing opportunities for the creation of new NBS (see section A.2.3.2). FIGURE A.2: A FRAMEWORK TO SUPPORT THE FIRST IDENTIFICATION OF POTENTIAL INVESTMENTS IN NBS SOURCE: Original figure for this publication. NOTE: NBS = nature-based solutions. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX A 61 A.2.3.1 ES SUITABILITY RULESETS 1 2 3 The spatial identification of NBS suitability is done through a ruleset of Earth observation biophysical 4 indicators, which represent the state of perviousness and topology of a given location. For each type of NBS, 5 a specific ruleset to identify suitability was developed, as shown in table A.4. 6 A1 A2 TABLE A.4: NBS SUITABILITY RULESET DENSITY MAX HEAVY PREC. HYDRO SOIL NBS FAMILY SOIL DEPTH SLOPE MAX ELEV. MAX SOIL BULK TEXTURE SURFACE SOIL PH WATER ROADS BUILD. CLASS DAYS SOIL LCC MIN River No 1,500 35 100 >4 1.6 n.a. n.a. 50m No n.a. n.a. Renaturation Green No 1,500 35 150 4–7 1.6 n.a. n.a. No 10m n.a. n.a. Corridors Urban Forest No 1,500 35 150 4–7 1.6 n.a. n.a. No No n.a. n.a. Bioretention No 1,500 5 100 n.a. n.a. n.a. ≤3 No No n.a. n.a. Area Urban No 1,500 20 n.a. n.a. n.a. not 1 n.a. No No ≤7 n.a. Farming or 12 Open Green No 1,500 35 150 4-7 1.6 n.a. n.a. No No n.a. ≥2 Space SOURCE: Original table for this publication. NOTE: Build. = building footprint; Elev. max = maximum elevation; Slope max = maximum slope; Soil min = minimum soil depth; Soil bulk density max = maximum soil bulk density; Heavy prec. days = Heavy precipitation days; LCC = Land Capability Classification; n.a. = not applicable. 62 CHAPTER: APPENDIX A THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES A description of how each indicator of table A.4 was generated follows. All input data layers used to calculate 1 the indicators of table A.4 are listed in table A.5. 2 3 4 Building footprint (build.): To distinguish built-up area Land Capability Classification (LCC): LCC is a global 5 from pervious area, we merged the World Settlement land evaluation ranking that groups soils based on 6 Footprint (Marconcini et al. 2020) with Google Open their potential for agricultural and other uses. LCC can A1 buildings (Sirko et al. 2021). Every location identified help determine whether land is suitable for certain A2 as a settlement or building in either data set is uses and whether there are risks for degradation. classified as a building. LCC is calculated using the LCC method described in Quandt et al. (2020). Input data for the LCC algorithm Elevation (Elev. max): To map global elevation, are the soil properties and slope described above. FABDEM (Hawker et al. 2022) was used: a 30-meter resolution global digital elevation model (DEM) with Surface water: To identify surface water, a surface removed buildings and tree cover. water layer prepared by the EU Joint Research Centre (JRC) (Pekel et al. 2016) was merged with water bodies Slope (Slope max.): The elevation data described in OpenStreetMap (OSM) (OpenStreetMap Contributors above were used to calculate a slope in degrees using 2023) and ESA World Cover (Earth Engine Data the slope function of GEE. Catalog, no date-b). Soil data (Soil depth min., Soil PH, Soil bulk density Roads: Roads were identified using OSM data max., Soil texture, Hydro soil class): For soil data (OpenStreetMap Contributors 2023). For green in Africa, ISDA estimated soil properties were corridors, suitability was only identified within 10 used, mapped at 30-meter resolution for the entire meters of roads classified as “motorway,” “primary,” continent of Africa (Hengl et al. 2021). For other “primary_link,” “road,” “secondary,” “secondary_link,” regions, the global soil texture classes at a 250-meter “tertiary,” “tertiary_link,” “trunk,” and “trunk_link.” resolution were used (Hengl 2018), and SoilGrids soil properties for other soil characteristics (Poggio et al. Heavy precipitation (Heavy prec. days): To identify 2021). Hydrological soil groups were calculated using the number of days with heavy precipitation, daily the method of Ross et al. (2018) using soil properties precipitation values from the CHIRPS data set described above. (Earth Engine Catalog, no date-a.) of the five most recent years were used and the number of days with precipitation over 10 millimeters were counted. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX A 63 ES TABLE A.5: INPUT DATA FOR NBS SUITABILITY MODELING 1 2 DATA LAYER SOURCE 3 4 Land cover ESA World Cover (Earth Engine Data Catalog, no date-b) 5 6 Elevation FABDEM (Hawker et al. 2022) A1 A2 ISDA Soil Properties (Africa) (Hengl et al. 2021) Soil properties Open Land Soil Texture (Hengl 2018) SoilGrids (Poggio et al. 2021) Precipitation CHIRPS precipitation data set (Earth Engine Catalog, no date-a) Google Open Buildings (Sirko et al. 2021) Settlements World Settlement Footprint (Marconcini et al. 2020) Global Water Surface Layer (Pekel et al. 2016) Water OpenStreetMap Water (OpenStreetMap Contributors 2023) Roads OpenStreetMap Roads (OpenStreetMap Contributors 2023). SOURCE: Original table for this publication. A.2.3.2 FIGURE A.3: A HIERARCHY OF APPROACHES UNDER CLASSIFICATION OF CREATION AND PROTECTION THE NATURE-BASED SOLUTIONS UMBRELLA OPPORTUNITIES The World Bank Catalogue of Nature-Based Solutions for Urban Resilience describes a hierarchical approach to NBS implementation in urban ecosystems (World Bank 2021) (figure A.3). The first element in the hierarchy, referred to here as “protection,” describes the sustainable management of existing NBS to sustain benefits and biodiversity. The second element, referred to as “enhancement,” represents the restoration and rehabilitation of degraded NBS. The third element, referred to as “creation,” points to the implementation of new natural or green infrastructure in cities. Following the NBSOS suitability analysis to identify land that can support NBS within cities, the suitable land identified for each NBS family is further classified as areas to protect existing green space (combining the elements of protection and enhancement) and SOURCE: World Bank 2021. areas that represent opportunities for the creation of new NBS. 64 CHAPTER: APPENDIX A THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES A.2.3.2.1 CLUSTERING AND CLASSIFICATION METHOD FOR EARTH OBSERVATION INDICATORS 1 2 To categorize suitable land into areas for protection or creation, the NBSOS used two EO indicators: (1) the bare 3 4 soil frequency and (2) the productivity performance indicator. The bare soil frequency indicator is calculated as 5 the number of bare soil observations from Sentinel-2 imagery divided by the number of cloud-free Sentinel-2 6 observations for each pixel within the two most recent years of data (Demattê et al. 2020). The productivity A1 performance indicator aims to describe the vigor of existing vegetation in the city. The performance A2 productivity indicator is calculated for each unique combination of landcover and soil texture within the AOI. For pixels within each unique landcover and soil texture combination, the mean Normalized Difference Vegetation Index (NDVI) is calculated for each pixel for the most recent year of Sentinel-2 observations and is divided by the 95 th percentile of mean NDVI across all pixels in the landcover/soil texture class such that higher values represent greener vegetation. The three layers of input data for the two indicators are listed in table A.6. For each indicator of bare soil frequency and productivity performance, we apply a Jenks natural breaks clustering algorithm to create rasters categorizing pixels into five integer classes (1–5), where values equal to 1 represent the most degraded land within the AOI (high bare soil frequency or low productivity performance) and values equal to 5 represent the greenest regions within the AOI (low bare soil frequency or high productivity performance). The respective clustered rasters are added together to create a new raster with values ranging from 2 to 10, with lower values representing more degraded land and higher values representing existing green space to protect. TABLE A.6: INPUT DATA FOR NBS OPPORTUNITY CLASSIFICATION The combined bare soil frequency/productivity performance cluster raster is manually inspected to identify a threshold integer DATA LAYER DATA SET NAME AND SOURCE value where values higher than or equal to the threshold represent regions of existing green ESA World Cover (Earth Engine Data Catalog, Land cover space to protect. For all clusters with a value no date-b) less than the manually determined threshold, ISDA Soil Properties (Africa) (Hengl et al. 2021) the cluster spatial extent is intersected with the suitability extent of each NBS family to Soil Open Land Soil Texture (Hengl 2018) identify degraded land suitable for the creation properties of NBS. The resulting product is a categorical SoilGrids (Poggio et al. 2021) raster with three categories representing Surface Sentinel-2 (Earth Engine Data Catalog, (1) the protection of existing green space, reflectance no date-c) (2) NBS creation opportunities, and (3) regions without existing vegetation that are not suitable SOURCE: Original table for this publication. for NBS creation. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX A 65 A.2.3.3 ES NBS SUITABILITY MAPPING 1 2 3 The NBSOS outputs include categorized maps of the AOI highlighting regions representing the protection 4 of existing green space and all opportunities for NBS creation (see map 2 in the main text). Additionally, the 5 NBSOS provides individual maps demarcating the creation opportunities for individual NBS families, as regions 6 suitable for creation of some NBS families may not be suitable for others (see map 3 in the main text). A1 A2 Opportunities for NBS protection and creation are presented using a set of maps created using the R software environment. First, NBS opportunities for the entire AOI are provided highlighting regions for NBS protection, creation, and the extent of existing human settlement. As the AOI often extends beyond the limits of densely populated city centers, a second map of NBS opportunities is frequently provided for the core city area, where NBS creation opportunities are often concentrated. Lastly, several maps are provided highlighting regions within the core city area representing NBS creation opportunities for each NBS family of interest. A.2.4 MAPPING BENEFITS The modeling of benefits follows two different FIGURE A.4: NBSOS BENEFITS AND IMPACTS methodologies (figure A.4). Some benefits are modeled using spatial analysis, while others are estimated using a more qualitative analysis. In the case of pluvial flood risk reduction, the NBSOS quantifies the impact by calculating the effect of NBS on runoff reduction through spatial analysis, estimating flood risk reduction in different areas. Regarding heat stress reduction, the NBSOS also uses spatial analysis to reduce the exposure to hazards in places where NBS are implemented. For health, a proxy for health and recreation benefits is used, and accessibility to green spaces is considered to estimate the benefit. A similar approach is followed for social interaction. For other benefits considered in the NBSOS, the impact of each NBS on each benefit is defined as none, low, medium, or high. This is based on the Catalogue of Nature-Based Solutions for Urban Resilience (World Bank 2021). In this case, the benefits are not spatially modeled; rather it is assumed that they are delivered where the NBS is located. SOURCE: Original figure for this publication. 66 CHAPTER: APPENDIX A THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES A.2.4.1 PLUVIAL FLOOD BENEFITS 1 2 3 4 The NBSOS utilizes a spatial flood mitigation TABLE A.7: INPUT DATA FOR FLOOD MITIGATION BENEFIT MODELING 5 model to estimate flood exposure reduction 6 potential from NBS by calculating the reduced flood risk from NBS implementation as the DATA LAYER DATA SET NAME AND SOURCE A1 A2 product of annual flood probability, exposed ESA World Cover (Earth Engine Data Catalog, population, and NBS-specific runoff reduction. Land cover no date-b) For each neighborhood, flood risk is calculated as the product of annual flood probability and ISDA Soil Properties (Africa) (Hengl et al. 2021) exposed population as explained in section Soil Open Land Soil Texture (Hengl 2018) A.2.2.1. Runoff reduction potential is calculated properties as the difference between runoff in the current SoilGrids (Poggio et al. 2021) scenario and runoff in an NBS scenario where each pixel that is identified as suitable for NBS Surface Sentinel-2 (Earth Engine Data Catalog, creation is converted to the respective NBS reflectance no date-c) family. The runoff reduction potential of every pixel is linked to neighborhoods through a basic Elevation FABDEM (Hawker et al. 2022) spatial hydrological model explained in section A.2.2.1.2. Data sources for this module are listed SOURCE: Original table for this publication. in table A.7. All modeling is carried out in Python. A.2.4.1.1 RUNOFF REDUCTION THROUGH SURFACE COVER CHANGES The NBSOS estimates the runoff reduction associated with surface cover changes due to the creation of an NBS family though the Curve Number (CN) method following the InVEST Urban Flood Risk Mitigation Model (UFRMM) (Hamel et al. 2021). A pixel-based runoff is calculated for a given storm depth based on a CN associated with each combination of soil type and land use class. Following UFRMM, runoff Q is estimated in millimeters as: where P is the storm depth in millimeters, Smax is the potential retention at pixel i, and λSmax is the rainfall depth needed to start runoff at pixel i. If P < λ × Smax, Q is 0. λ is a parameter that is set by default at 0.2. Smax is a function of CN and is calculated as THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX A 67 ES For each NBS family, runoff reduction per pixel of the current situation is compared with a curve number 1 calculated with all pixels converted to the respective NBS family. Curve numbers were derived following 2 Jaafar, Ahmad, and El Beyrouthy (2019), using SoilGrids (Poggio et al. 2021) and ISDA soil properties data (Hengl 3 et al. 2021) to derive hydrological soil groups (Ross et al. 2018). Hydrological soil groups, in combination with 4 the WorldCover 2020 land cover data (Earth Engine Data Catalog, no date-b) are used to derive a land cover/ 5 hydrological soil group–specific curve number. Additionally, bare soil frequency (Demattê et al. 2020) is used 6 to differentiate three hydrologic conditions (poor, fair, and good) for each hydrological soil group/land cover A1 combination. Changes in land cover and bare soil percentage are shown in table A.8. A2 TABLE A.8: LAND COVER CONVERSION AND BARE SOIL PERCENTAGE CHANGE ASSOCIATED WITH EACH NBS FAMILY NBS FAMILY LAND COVER TYPE CONVERSION BARE SOIL (% DECREASE) Open green space Grassland 20 Urban forest Trees 40 Green corridors Trees 30 Urban farming Agriculture 10 Inland wetland Wetland 50 River floodplain Wetland 40 River renaturation Grassland 20 Terrace slope Agriculture 10 Bioretention area Wetland 40 Green roof Grassland 20 SOURCE: Original table for this publication. NOTE: Bare soil percentage decrease is a proxy for the hydrologic condition of the given pixel. Indeed, the establishment of a new NBS assumes an improvement in the hydrologic condition of the given pixel (that is, a reduction in bare soil percentage). The CN-based runoff reduction of each pixel is multiplied with the flood hazard mitigation potential calculated in paragraph A.2.2.1 to yield a final flood mitigation potential value for each pixel that accounts for downstream flood exposure reduction. 68 CHAPTER: APPENDIX A THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES A.2.4.1.2 RUNOFF REDUCTION THROUGH STORAGE 1 2 Bioretention areas and open green spaces can also mitigate flood exposure through storage, which is based on 3 4 the equations below (Reig et al. 2019): 5 6 A1 A2 where SV is the supply runoff volume entering the NBS (in cubic meters), R is the rainfall (in meters), A is the drainage area (in square meters), C is the runoff coefficient, VC is the runoff volume captured by the NBS (in cubic meters), and RRF is the runoff reduction factor, specific for each NBS. To simulate this process, compatible with the runoff reduction method described in the previous section, the NBSOS incorporates a storage model based on curve numbers by calculating a runoff coefficient (McCuen and Bondelid 1981). For each contiguous area identified as suitable for creation of bioretention areas or open green spaces, the total area is represented by S. The area (A) sent to this potential NBS patch j is calculated as where F is an NBS-specific factor (0.05 for bioretention areas and 0.03 for open greens pace (ARC 2016; McCuen and Bondelid 1981). Moreover, A is capped at 1 hectare for bioretention areas and 4 hectares for open green spaces (ARC 2016; Woods Ballard et al. 2015). Using the average CN value of area A, a runoff coefficient C is calculated as Subsequently, a new C value (Ca) is calculated: where Cb is the original C value and RRF is an NBS-specific coefficient with 0.4 for bioretention and 0.2 for open green spaces (Battiata et al. 2010; Hirschman et al. 2018). To estimate the storage effect, new average CN value for area A (McCuen and Bondelid 1981) is simulated as In a last step, the storage effect is simulated by calculating the runoff reduction using the new CN values for each modeled bioretention or open green space patch. The difference between runoff in the baseline scenario and the new CN scenario is assigned as the storage effect expressed in the same unit as the runoff reduction value calculated for the other NBS families. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX A 69 A.2.4.2 ES HEAT STRESS BENEFITS 1 2 3 Heat stress describes the suite of conditions that can occur in the human body when environmental conditions 4 preclude the shedding of excess heat. Projected increases in global temperatures are expected to intensify 5 heat stress throughout the twenty-first century, with extreme heat conditions endangering human health, 6 impairing economic growth, reducing agricultural yields, and compromising ecosystems. Cities, characterized A1 by warmer temperatures, higher population densities, and increased economic activity, are particularly A2 susceptible to the consequences of extreme heat. NBS, however, have the potential to mitigate local temperatures by providing shade and consuming solar radiation for evapotranspiration. The NBSOS includes a module to estimate spatial variability in NBS heat stress benefits based on the priority areas described in section A.2.2.2 and statistical relationships between landcover composition and air temperature. A.2.4.2.1 SPATIAL REGRESSION MODEL OF LANDCOVER IMPACTS ON AIR TEMPERATURE The NBSOS heat stress module estimates spatial variability in potential heat stress mitigation across the AOI through relationships between landcover composition and average annual air temperature (air temperature data processing is described in section A.2.2.2). For each neighborhood subregion within the AOI, the landcover composition is determined by estimating the percent landcover of eight different landcover types mapped as part of the European Space Agency (ESA) WorldCover data set, which includes built-up land, trees, grass, shrubs, wetlands, croplands, bare land, and water. The landcover composition of subregions is compared with modeled average annual air temperature in a spatial regression analysis to estimate the local sensitivity of air temperature to landcover types that can be manipulated via NBS implementation. Typically, the model estimates only the impacts of built-up land and tree cover; however, any of the eight landcover types described above could be included in the analysis. The spatial regression modeling framework includes spatial lag effects on the independent variables and model error to estimate both local and “spillover” effects of NBS implementation across the AOI. The heat benefits module utilizes a spatial autoregressive framework because of spatial autocorrelation in the temperature data, which might bias coefficient estimates from a traditional ordinary least squares regression modeling framework. The model is estimated as: where y is the dependent variable vector (average annual air temperature (°C)), X is the independent variable matrix (for example, tree cover (%), built-up cover (%)), β is the regression parameter vector, W is a spatial weighting matrix, θ is the spatial lag parameter vector, u is the spatial error, λ is the spatial coefficient of the error, and ε is the error vector of the model. Using the estimated coefficients describing the sensitivity of air temperature to land cover compositions (that is, β and θ), average air temperature is predicted under scenarios for each NBS family that assumes NBS creation in all pixels deemed suitable for its creation, resulting in a spatially variable data set of potential ambient air temperature reductions from NBS creation. To account for the importance of temperature reduction in priority areas characterized by a higher population density and average annual air temperature, 70 CHAPTER: APPENDIX A THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES spatial data on potential temperature reductions are combined with heat stress priority areas to estimate heat 1 stress benefits as: 2 3 4 5 6 where the heat stress benefits for pixel i and NBS family j are equal to the product of heat stress priority within A1 the subregion k, the absolute value of the estimated potential temperature reduction from NBS implementation A2 for pixel i and NBS family j, and a binary suitability indicator for pixel i and NBS family j. A.2.4.3 HEALTH, RECREATION, AND SOCIAL COHESION BENEFITS NBS can promote and/or improve health, recreation, and social cohesion by providing opportunities for social activities, such as walking and cycling, especially in proximity to residential areas. Urbanization results in a higher demand for recreation opportunities and increased visitor numbers in NBS such as urban forests, open green spaces, or green corridors. Recreation/social benefit potential is calculated by assuming that benefits of urban green spaces are provided in areas where residents live within 300 meters of an urban green space (Konijnendijk 2023). To estimate the potential health/recreation benefits of areas identified as suitable for the creation of open green spaces or urban forests, calculations are made of the number of people within a radius of 300 meters of contiguous potential patches larger than 1 hectare who would gain access to this green space and who did not already have access to such an urban green space within 300 meters. Effectively, a location suitable for urban forest or open green space creation larger than 1 hectare in an area with few nearby urban green spaces and high population counts would yield a higher potential benefit than a suitable location with lower population counts or a greater abundance of existing urban green space. To estimate social cohesion benefits, the NBSOS utilizes a methodology similar to that for the health/recreation benefits but reduces the minimum green space threshold to 0.5 hectares to account for green spaces that may not be large enough for recreation but are large enough to support social interaction. A.2.3.4 JOBS AND RESOURCES PRODUCTION BENEFITS NBS investment in cities represents an opportunity for the creation of new jobs through the implementation of NBS and the potential for resources production once established. For example, a community-based reforestation project in Freetown, Sierra Leone, has created more than 550 jobs to support local economies (World Bank 2021). Bioretention areas can improve the image and market value of real estate to promote economic development, generate green jobs, and increase productivity for workers with access to green areas (World Bank 2021). Urban farming increases food supply and production, reduces the distance that food must travel from the producer to the consumer, and creates jobs for agricultural entrepreneurs and workers (World Bank 2021). The NBSOS estimates potential jobs and resources production benefits by combing a spatially weighted population distribution model to capture variability in population density with proximity to NBS creation with a weighting matrix estimating variability in the provision of jobs and resources production benefits across different types of NBS. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX A 71 ES A.2.3.4.1 SPATIALLY WEIGHTED POPULATION MODEL 1 2 Population density is estimated using the High Resolution Settlement Layer (FCL and CIESIN 2016), following 3 the same methodology as the population density in the heat stress module. 4 5 For each NBS family, jobs and resources production benefits are estimated as: 6 A1 A2 where the benefits for pixel i and NBS family j are a function of the weighted population value of pixel i, the suitability of pixel i to support NBS family j, and the weighting value for NBS family j that defines the ability of the NBS family to provide benefits relative to others. Example jobs and resources production weights for a subset of NBS families are provided in table A.9. TABLE A.9: EXAMPLE JOBS AND RESOURCES PRODUCTION WEIGHTS FOR A SUBSET OF NBS FAMILIES NBS FAMILY JOBS AND RESOURCES PRODUCTION WEIGHT Bioretention area 1 Open green space 2 Green corridor 1 Urban farming 3 Urban forest 2 SOURCE: Original table for this publication. A.2.4.5 SOIL EROSION BENEFITS Urban NBS can be implemented to stabilize slopes and soils, particularly in mountainous or hilly cities characterized by loose soils or cities with seasonal streambeds susceptible to erosion from water. The NBSOS estimates the potential for NBS creation to mitigate soil erosion by combining annual soil erosion rates with NBS-specific weights characterizing the relative potential for each NBS family to reduce soil erosion. Annual soil erosion rates are estimated with the Revised Universal Soil Loss Equation (RUSLE). The RUSLE estimates long-term annual soil erosion rates (tonnes ha-1 yr-1) from water as: 72 CHAPTER: APPENDIX A THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES where A is the soil erosion rate, R is the rainfall erosivity factor, K is the soil erodibility factor, L is the slope- 1 length factor, S is the slope-steepness factor, and V is the vegetation factor. Soil erosion rates are then 2 combined with the NBS-specific weights as: 3 4 5 6 A1 A2 where the soil erosion benefits for pixel i and NBS family j are the product of the soil erosion rate in pixel i and the weighting values for NBS family j. A.2.4.6 OTHER BENEFITS Benefits from the implementation of NBS in cities that do not have dedicated modules within the NBSOS, such as water quality improvement and subsidence regulation, are estimated using matrix weighting models. In these models, relative benefits across NBS families are estimated as a function of NBS suitability and the potential for benefit provision across NBS families, where weights of benefit provision are informed by the World Bank’s Catalogue of Nature-Based Solutions for Urban Resilience (World Bank 2021). The modular structure of the NBSOS, however, allows for the constant development of additional benefits modules, with new benefits modules regularly developed and implemented to improve benefit estimation. A.2.4.7 VISUALIZATION AND MAPPING For each analyzed benefit and NBS family of interest, the benefit values are normalized for pixel i and NBS family j by the equation: such that the normalized benefit index describes values from 0 to 1, where higher values correspond to pixels with a greater potential benefit. The benefits of interest are presented through (1) a series of maps created with the R software environment describing spatial variability in the provision of each benefit across the AOI and (2) a bar graph characterizing the area of feasible NBS creation with the potential to provide low, medium, and high benefit levels (both are available in map 4 in the main report). Low, medium, and high thresholds are determined for each benefit of interest by first computing the maximum potential benefit level per pixel across all NBS families. The distribution of maximum potential benefits across pixels is then further divided into the 0 to 25th percentile (low benefits), the 25th to 75th percentile (medium benefits), and the 75th to 100th percentile (high benefits). THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX A 73 ES A.2.5 1 2 MAPPING OPTIMAL ALLOCATION OF NBS 3 4 5 In the fourth analysis step, the NBSOS implements a multicriteria analysis to identify spatial variability in 6 optimal NBS families, which can be used to inform an NBS investment plan (see box A.1 for an example of A1 decision support). In practice, creation opportunities for the implementation of NBS interventions such as A2 urban forests, open green spaces, and urban agriculture overlap. However, these NBS interventions will provide different benefits. The multicriteria optimization analysis assigns weights to the different benefits of interest (for example, pluvial flood reduction is more important than heat stress mitigation). These weights are used to estimate variability in combined benefits provision across NBS families with overlapping suitability. The result is a map depicting the optimal allocation of NBS families that maximize combined benefits. The total area of each NBS family in the map of optimal solutions can subsequently be combined with cost estimates of NBS family-specific implementation to estimate a total cost of investment (section A.2.5.2.2). A.2.5.1 BENEFIT WEIGHTING AND COMBINATION Benefit weights are determined considering the local context and as a part of task team consultation to choose which benefits are most important in the AOI. Sometimes more than one scenario is proposed (for example, equal benefit weights and heat mitigation with highest weight). Each benefit of interest is assigned a weight (as a percentage) corresponding to its project importance such that all benefits’ weight sum to 100 percent. For example, in a project primarily targeting pluvial flood reduction, while also interested in co-benefits of heat stress reduction and improved health/recreation, a flood-heavy weighting scheme may be applied such that flood benefits are assigned a weight of 60 percent and heat stress/health benefits are each assigned weights of 20 percent. To compute the combined benefits across the AOI for each NBS family, normalized benefit values for each benefit are multiplied by the corresponding weighting value and summed together. For example, in the flood- heavy weighting scheme described above, the normalized combined benefits for NBS family j and pixel i are calculated as: resulting in a combined benefits map for each NBS family of interest. 74 CHAPTER: APPENDIX A THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES A.2.5.1.1 MAPPING MAXIMUM POTENTIAL BENEFITS PER PIXEL 1 2 The combined benefits maps for the NBS families of interest are merged into a multiband raster for the AOI. 3 4 To quantify spatial variability in the maximum potential benefits per pixel, the maximum value is calculated 5 across all bands for each pixel in the multiband combined benefits raster. 6 A1 A.2.5.1.2 SELECTION OF NBS FAMILIES PROVIDING MAXIMUM POTENTIAL BENEFITS A2 A categorical raster depicting the optimal NBS family per pixel is produced by combining the maximum potential benefits raster with each NBS family-specific combined benefits raster. The optimal NBS family per pixel is defined as the NBS typology providing combined benefits equal to the maximum potential benefits (see map 5 in the main text for an example). A.2.5.2 ESTIMATE PROJECT INVESTMENT A.2.5.2.1 REDUCE OPTIMAL NBS MAP TO PERCENTILE DEFINING GREATEST BENEFITS In most cases, it is not feasible to implement all optimal NBS interventions identified by the NBSOS. The optimal NBS map can be restricted to depict only regions providing the highest level of benefits, defined in consultation with task teams according to budget constraints. Thresholds are assigned based on the distribution of maximum combined benefits, where optimal NBS families and locations providing benefits that fall below the threshold are removed from the map. For example, in the case of identifying the most beneficial 10 percent NBS opportunities, pixels falling below the 90th percentile of values in the maximum potential benefits raster are removed from the analysis prior to selecting NBS families providing maximum potential benefits (section A.2.5.1.2). A.2.5.2.2 COST ESTIMATION To estimate the investment cost, unit cost values from previous projects in the region are used for each NBS and converted to US dollars. These values are updated to current values using the Consumer Price Index (CPI). This means that price values for earlier years are corrected for inflation using the cost to the average consumer of acquiring a basket of goods and services that may change yearly. The correction is performed using the following equation: where project year is the year of the reference project that provides the reference unit cost values. The NBS cost estimated considers only capital expenditure (CAPEX) values for each NBS per unit of area or length. The final value is obtained multiplying unit implementation costs and the coverage values corresponding to the percentage of most beneficial NBS opportunities. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX A 75 ES BOX A.1 1 DECISION SUPPORT FOR NBS INVESTMENTS IN URBAN AREAS 2 Some NBS are suitable for the same space. Multicriteria The scenario shown below corresponds to the optimal NBS analysis is applied to select the most effective combination delivering the 20 percent highest level of combined benefits 3 of NBS to maximize the targeted benefits. T ​ he distribution in or, put another way, NBS that provide benefits corresponding 4 this example corresponds to the case of 60 percent weight for to the highest 80th percentile. The needed investment was 5 pluvial flood benefits and equal weight of 20 percent for heat estimated using unit costs from other projects in the region.​ stress reduction and improved access to green spaces.​ 6 A1 A2 TABLE A.1.1: BREAKDOWN OF NBS INVESTMENT IN GREATER DAKAR GREATER DAKAR (EAST) NBS WITH 20% HIGHEST BENEFITS OPPORTUNITY CAPEX (US$, MILLIONS) Urban forest​ 45 hectares​ $2.8​ Green corridors​ 50 kilometers​ $6.3​ Open green spaces​ 61 hectares​ $28.3​ Bioretention areas​ 13 hectares​ $31.4 ​ TOTAL INVESTMENT​ $68.8​ SOURCE: Original map for this publication. SOURCE: Original table for this publication. NOTE: The optimal solution comprise the NBS providing the NOTE: CAPEX = capital expenditure. 20 percent highest level of combined benefits. Investment costs were estimated using unit costs from other projects in the same region. Some recommendations • The design of multifunctional open green • The design of multifunctional green to maximize benefits in spaces designed for both stormwater corridors, offering opportunities for both this case include:​ detention and recreation.​ social interaction and heat mitigation. 76 CHAPTER: APPENDIX A THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES 1 2 3 A.3 4 5 6 A1 A2 APPLICATION OF THE NBSOS AS A SERVICE TO WORLD BANK TASK TEAMS ©anaya-katlego_unsplash A.3.1 TASK TEAM CONSULTATION Consultations with Task Teams are held at the beginning of each NBSOS application. During these meetings, different types of relevant information are collected. This helps to provide a customized service, considering local characteristics, necessities, and data. A.3.1.1 DETERMINATION OF THE AREA OF INTEREST The AOI usually covers the built-up area, or core city area, along with an extra area defined by the Task Team. If the focus is on flood mitigation, for instance, the AOI might be extended to cover the upper part of the catchment to identify opportunities for NBS to reduce runoff going into the city. When there is an interest in exploring opportunities around the city, a buffer around the urbanized area is defined with the local Task Team to determine the AOI. A.3.1.2 IDENTIFICATION OF RELEVANT ENVIRONMENTAL HAZARDS AND ASSOCIATED WEIGHTS The main urban challenges are identified by the local Task Team. Usually, between three and five challenges are selected; the most common are flood reduction, heat stress mitigation, health/recreation improvements, social cohesion enhancement, and the creation of local jobs and resources production. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX A 77 A.3.1.3 ES SELECTION OF RELEVANT NBS FAMILIES 1 2 3 Relevant NBS families for each case are also discussed with local Task Teams. Frequently, the definition of 4 local challenges to be addressed determines the main NBS to be studied. For instance, if one main challenge 5 is pluvial flood risk, bioretention areas and open green spaces will be considered among the NBS studied. 6 When heat stress reduction is targeted, urban forests and green corridors are included in the analysis. In some A1 cases, a further discussion about local preferences or characteristics may lead to the identification of specific A2 solutions not identified a priori. For example, the existence of urban streams as part of the urban fabric would lead to a study of opportunities for stream renaturation. A.3.1.4 AVAILABILITY OF SUPPLEMENTAL LOCAL DATA SETS The availability of local data to enhance and customize the NBSOS analysis is also discussed with the local Task Team. Even though the analysis can be performed using available global data, the addition of local data can improve some results or their visualization. For instance, population vulnerability data, which usually are not available as open data, can improve flood and heat risk analyses. Administrative divisions data can help to present results per neighborhood or district, helping to better communicate the NBSOS’s outputs. A.3.2 INTERPRETATION, PRESENTATION, AND RECOMMENDATIONS A.3.2.1 DELIVERABLES The NBSOS results are delivered as a PowerPoint (PPT) presentation deck, including a summary of NBSOS methodology, context analysis with indices regarding local hazards, and NBSOS results. These results include the identification of priority areas from hazard-exposure analysis, opportunity maps of different NBS, and potential benefits levels obtained from their implementation. Finally, the PPT deck also includes an interpretation of results and recommendations, along with design and implementation considerations for the NBS studied. In addition to the deck, geospatial raster data of suitability of NBS families (such as GeoTIFF), shapefiles describing climate resilience and environmental challenges (flooding, heat, lack of public green space), and indexes that describe the impact of NBS-types mitigating these challenges are also delivered. A.3.2.2 PRESENTATION TO TASK TEAMS AND CLIENTS The PPT deck is presented to the Task Team during an online meeting. During this meeting, feedback about the results obtained, their interpretation, and recommendations are collected. These suggestions are integrated into the final files for submission. 78 CHAPTER: APPENDIX A THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES A.3.2.3 PARTICIPATORY MODELS FOR CLIENT AND STAKEHOLDER ENGAGEMENT 1 2 3 4 In some cases, further engagements with local Task Teams include activities to help the use of NBSOS results 5 with local stakeholders. These activities include the presentation of results, interactive activities to facilitate 6 discussions among local actors, support during pre-feasibility studies, and so on. A1 A2 A.3.2.4 COSTED INVESTMENT PLANS TO INFORM PROJECT APPRAISAL Optimal NBS maps for highest benefits and estimations of costs can be used to inform project appraisal phases. The main limitation is usually the nonavailability of local data on NBS costs. In many cases, unit cost data from other countries in the region are used, but even this type of data sometimes presents a challenge. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX A 79 ES 1 2 3 A.4 4 5 6 A1 NBSOS VALIDATION: A2 NBS FOR FLOOD REDUCTION ©dimitry_unsplash A.4.1 INTRODUCTION The NBSOS uses input flood hazard from Fathom to assess where urban spaces get flooded (Fathom, no date) and, based on this, to determine where to implement NBS to reduce pluvial flooding. Therefore, the input data on flood hazard have a high impact on the definition of priority areas for NBS implementation for flood reduction and on the assessment of the level of benefits obtained from the proposed solutions. One way to validate the exercise’s objectives is to evaluate the accuracy of the flood hazard data used. Using data on flood hazards, the NBSOS estimates catchment areas contributing to flooded areas to establish where the implementation of NBS for runoff reduction could have a high impact on reducing this problem (priority area). Another objective of this validation exercise is to assess whether the areas identified as priority areas by the NBSOS are also seen as priority areas for NBS implementation by a local in-depth study. Finally, the NBSOS estimated degradation levels of urban green spaces to classify areas as protection areas (low degradation) and creation areas (high degradation). There is a third category implicitly included in the NBSOS results—sometimes green spaces classified as protection areas have room for enhancement. This means that green spaces with low degradation level can still be enhanced by designs to, for instance, increase storage capacity and recreation opportunities. The last objective of the validation is to compare NBSOS results on protection and creation with solutions proposed by the local study. 80 CHAPTER: APPENDIX A THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES 1 2 A.4.2 3 4 OBJECTIVES 5 6 This validation comprises three parts: A1 A2 1. One of the main factors driving the output of the NBSOS is the input flood maps employed in the analysis. For this reason, this part of the validation focuses on assessing how the input flood hazard areas, which are Fathom flood maps version 2 (Fathom, no date), compare to the flood maps obtained from a locally calibrated hydrodynamic model with higher resolution and accuracy. 2. The second part of the validation compares areas identified by the NBSOS as being of high priority to allocate NBS for flood reduction with areas where the local study recommends the implementation of sustainable urban drainage. 3. The third part focuses on evaluating the NBS suitability obtained using the NBSOS by comparing protection and creation recommendations with recommendations from the local study regarding flood reduction solutions. The second and third parts aim to assess the accuracy of the output produced by the NBSOS. The comparison is performed against the outputs obtained from an analysis that used higher resolution data, exploited in-situ data, and elicited local stakeholders’ knowledge to calibrate and validate the results. A.4.3 DESCRIPTION OF LOCAL STUDY USED FOR COMPARISON The project Flood and Coastal Risk Assessment and Priority Investment Planning for Greater Banjul was part of a technical assistance (TA) to the government of The Gambia led by the World Bank and funded by the Africa Caribbean Pacific (ACP) – EU Natural Disaster Risk Reduction (NDRR) Program through the World Bank Global Facility for Disaster Reduction and Recovery (GFDRR). The overall objective of the TA was to deliver an assessment of the flood and coastal risks in the Greater Banjul Area (GBA) and the Kombo North/Saint Mary district and to identify and prioritize, in a participatory way, measures and infrastructure investments, including NBS. To accomplish the goals of the project, the Dutch firm Royal Haskoning DHV was hired by the World Bank’s Task Team to conduct the necessary assessments. Flood and coastal erosion hazard were assessed through advanced inundation modeling and analysis of satellite images and validated with the stakeholders. Future natural hazards were estimated by extrapolation of trends in combination with the recent climate change projections from the Intergovernmental Panel on Climate Change (IPCC) 5th assessment report for time horizons 2040 and 2070. The resulting hazard maps were combined with the land use maps and damage functions to estimate the vulnerability and risk (economic and social) related to these hazards. The mapped risks were used as input to the development of investment options: packages of strategic measures that were THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX A 81 ES prioritized through a multicriteria analysis and a cost-benefit analysis. Finally, recommendations for selected 1 investments and their priorities were prepared. 2 3 Flood model simulations were run for different recurrence periods ranging from 0.4-year to 200-year and for 4 different time horizons (2020, 2040, 2070) to get a full picture of inundation extents and depths. The results of 5 hazard and risk assessment were used to identify hotspot areas (where the risk level is the highest) for which 6 an integrated strategy with a corresponding package of risk mitigating measures was developed (including A1 structural, nonstructural, and nature-based options). A2 Despite certain limitations—in particular, the fact that the area has a flat topography constitutes a challenge for the flood model—the outputs of this assessment can be considered to be of high accuracy given the accuracy and high resolution of the data employed and the validation steps employed in the overall process. Moreover, the goal of the study overlaps with the scope of the NBSOS. Therefore, we used these outputs as a benchmark to assess the quality of the NBSOS. A.4.4 RESULTS A.4.4.1 FLOOD HAZARD VALIDATION In this assessment, different return period (RP) flood maps for the Greater Banjul Area, The Gambia, obtained from Fathom v2, Fathom v3, and a calibrated hydrodynamic model (see section A.4.3 are compared (map A.1). Results for both 20- and 100-year return periods show that Fathom data accurately represents flood areas along the main drainage system, not fully representing smaller surfaces of decentralized flooding. This is an expected result due to resolution differences in the models, while Fathom has a resolution of 90 meters and 30 meters, respectively, for versions 2 and 3, the case-specific model uses a 2.5-meter resolution. For this case, better accuracy is shown by Fathom v3 in the case of RP100. However, Fathom v2 shows better results for the case of the 20-year return period. A.4.4.2 PRIORITY AREAS FOR NBS CREATION Another result obtained from the NBSOS is the identification of priority areas where it would be most effective to allocate NBS for flood reduction. These areas are shown by the hexagons in shades of green in map A.2, where the darker the green the higher the priority. The map also shows areas recommended by the local study where to implement sustainable urban drainage systems (SUDS), which is another name given to NBS specifically applied for urban drainage. Comparing both recommendations, the NBSOS recommends implementing NBS mainly in areas also identified by the local study. It is important to note that the local study looks at the availability of spaces, resulting in more precise recommendations. 82 CHAPTER: APPENDIX A THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES MAP A.1: FLOODED AREAS IN BANJUL, THE GAMBIA, BY FATHOM V2 (UPPER MAPS), FATHOM V3 (MIDDLE MAPS), AND LOCAL HYDRODYNAMIC 1 MODEL (LOWER MAPS), FOR THE CASES OF 20 YEARS (LEFT) AND 100 YEARS (RIGHT) RETURN PERIODS 2 3 4 5 6 A1 A2 SOURCE: Original maps for this publication based on Fathom v2 and v3 data and local model data. NOTE: SUDS = sustainable urban drainage systems. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX A 83 ES MAP A.2: COMPARISON BETWEEN AREAS IN BANJUL, THE GAMBIA, IDENTIFIED BY THE NBSOS AS HIGH PRIORITY FOR NBS IMPLEMENTATION 1 (GREEN) AND AREAS FOR SUDS IMPLEMENTATION RECOMMENDED BY THE LOCAL STUDY 2 3 4 5 6 A1 A2 SOURCE: Original map for this publication based on NBSOS data and local study data. NOTE: NBS = nature-based solutions; NBSOS = Nature-Based Solutions Opportunity Scan; SUDS = sustainable urban drainage systems. A.4.4.3 SUITABILITY FOR NBS PROTECTION AND CREATION Finally, an output from the NBSOS suitability analysis is the identification of areas for NBS protection, which have existing green spaces in good condition. The NBSOS also identifies areas for NBS creation, which are bare areas or areas with green spaces in bad condition. Map A.3 shows green space areas (depicted in purple) identified as opportunities for NBS protection. The map also shows (in red) the area identified by the local in- depth study as an opportunity for creating a green park along the river. Even though the NBSOS identifies this area as suitable for protection and not creation, the results are comparable. The area is a green space in good condition, and creating a park would protect it from being urbanized or degraded and enhance it to provide more benefits than it currently does. 84 CHAPTER: APPENDIX A THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES MAP A.3: NBS PROTECTION AREAS IN BANJUL, THE GAMBIA, DEFINED BY THE NBSOS (PURPLE), AND IN YELLOW, THE GREEN PARK 1 SUGGESTED BY THE LOCAL STUDY 2 3 4 5 6 A1 A2 SOURCE: Original map for this publication based on NBSOS data and local study data. A.4.5 CONCLUSIONS The flood model validation exercise provides evidence supporting the use of global flood maps in the NBSOS to identify high-priority areas to target flood mitigation through NBS. • Flood maps: Despite the difference in accuracy between the Fathom global model and the local one, the main hazard characteristics are captured and factored in the NBSOS. Fathom v3 shows a great improvement in resolution and accuracy, which will be reflected in the accuracy of future NBSOS. • NBSOS high-priority areas: Despite the huge difference in terms of resolution and accuracy between the inputs used by the NBSOS (global and low-resolution data) and the ones used by the local study (in-situ and high-resolution data), the NBSOS can correctly identify the main hotspots for risk reduction, and it is able to suggest similar recommendations regarding NBS creation, protection, or enhancement. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX A 85 REFERENCES ES 1 2 3 4 5 ARC (Atlanta Regional Commission). 2016. Georgia Stormwater Hamel, P., A. D. Guerry, S. Polasky, et al. 2021 “Mapping the 6 Management Manual. https://atlantaregional.org/what-we-do/ Benefits of Nature in Cities with the InVEST Software.” npj Urban A1 natural-resources/georgia-stormwater-management-manual/. Sustainability 1: 25. https://doi.org/10.1038/s42949-021-00027-9. A2 Bartos, M. D. 2020. pysheds. Database, accessed February 14, Hawker, L., P. Uhe, L. Paulo, J. Sosa, J. Savage, C. Sampson, 2024. https://github.com/mdbartos/pysheds. and J. Neal. 2022. “A 30 m Global Map of Elevation with Forests and Buildings Removed.” Environmental Research Letters 17 (2): Battiata, J., K. Collins, D. Hirschman, and G. Hoffmann. 2010. 024016. https://doi.org/10.1088/1748-9326/ac4d4f. “The Runoff Reduction Method.” Journal of Contemporary Water Research & Education 146 (1): 11–21, https://doi.org/10.1111/j.1936- Hengl, T. 2018. “Soil Texture Classes (USDA System) for 6 Soil 704X.2010.00388.x. Depths (0, 10, 30, 60, 100 and 200 cm) at 250 m.” Zenodo, December 24, 2018. https://zenodo.org/records/2525817. Brandt, J., J. Ertel, J. Spore, and F. Stolle. 2023. “Wall-to- Wall Mapping of Tree Extent in the Tropics with Sentinel-1 and Hengl, T., M. A. E. Miller, J. Križan, et al. 2021. “African Soil Sentinel-2,” Remote Sensing of the Environment 292: 113574. Properties and Nutrients Mapped at 30 m Spatial Resolution Using https://doi.org/10.1016/j.rse.2023.113574. Two-Scale Ensemble Machine Learning.” Scientific Reports 11 (1): 6130. https://doi.org/10.1038/s41598-021-85639-y. Demattê, J. A. M., J. L. Safanelli, R. R. Poppiel, et al. 2020. “Bare Earth’s Surface Spectra as a Proxy for Soil Resource Monitoring.” Hirschman, D. J., M. Aguilar, J. Hathaway, K. Lindow, and Scientific Reports 10 (1): 4461. https://doi.org/10.1038/s41598- T. Schueler. 2018. “Updating the Runoff Reduction Method.” 020-61408-1. Prepared for the Metro Government of Nashville & Davidson County, Tennessee Metro Water Services, Stormwater Division. Earth Engine Catalog. No date-a. CHIRPS Pentad: Climate https://hirschmanwater.com/wp-content/uploads/2018/07/ Hazards Group InfraRed Precipitation With Station Data (Version RRM-Nashville_Report_FINAL_060718.pdf. 2.0 Final) (data set). Provider USCB/CHG. https://developers. google.com/earth-engine/datasets/catalog/UCSB-CHG_ Hooker, J., G. Duveiller, and A. Cescatti. 2018. “A Global Dataset CHIRPS_PENTAD. of Air Temperature Derived from Satellite Remote Sensing and Weather Stations.” Scientific Data 5 (1): 180246. https://doi. Earth Engine Data Catalog. No date-b. ESA WorldCover 10m v100 org/10.1038/sdata.2018.246. (data set). Provider ESA/VITO/Brockmann Consult/CS/GAMMA Remote Sensing/IIASA/WUR. https://developers.google.com/ Jaafar, H. H., F. A. Ahmad, and N. El Beyrouthy. 2019. “GCN250, earth-engine/datasets/catalog/ESA_WorldCover_v100. New Global Gridded Curve Numbers for Hydrologic Modeling and Design.” Scientific Data 6 (1): 145. https://doi.org/10.1038/s41597- Earth Engine Data Catalog. No date-c. Harmonized Sentinel-2 019-0155-x. MSI: MultiSpectral Instrument, Level-2A (data set). Provider European Union/ESA/Copernicus. https://developers.google. Konijnendijk, C. C. 2023. “Evidence-Based Guidelines for Greener, com/earth-engine/datasets/catalog/COPERNICUS_S2_SR_ Healthier, More Resilient Neighbourhoods: Introducing the 3–30– HARMONIZED. 300 Rule.” Journal of Forestry Research 34 (3): 821–30. https://doi. org/10.1007/s11676-022-01523-z. Fathom. No date. “Global Flood Map,” Fathom. Accessed February 14, 2024. [Online]. https://www.fathom.global/product/global- Marconcini, M., A. Metz-Marconcini, S. Üreyen, et al. 2020. flood-map/. “Outlining Where Humans Live, the World Settlement Footprint 2015.” Scientific Data 7 (1): 242. https://doi.org/10.1038/s41597- FCL and CIESIN (Facebook Connectivity Lab and Center for 020-00580-5. International Earth Science Information Network), Columbia University. 2016. High Resolution Settlement Layer (HRSL). McCuen, R. H. and T. R. Bondelid. 1981. “Relation between Curve Source imagery for HRSL, copyright DigitalGlobe. https:// Number and Runoff Coefficient.” Journal of the Irrigation and dataforgood.facebook.com/dfg/tools/high-resolution- Drainage Division 107 (4): 395–400. https://doi.org/10.1061/ settlement-layer. JRCEA4.0001425. 86 CHAPTER: APPENDIX A THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES OpenStreetMap Contributors. 2023. Planet Dump. https://planet. Sahr, K. 2020. DGGRID version 7.05. Data set. https://h3geo.org. osm.org. 1 Sirko, W., S. Kashubin, M. Ritter, A. Annkah, Y. S. E. Bouchareb, 2 Pekel, J. -F., A. Cottam, N. Gorelick, and A. S. Belward. 2016. Y. Dauphin, D. Keysers, M. Neumann, M. Cisse, and J. Quinn. 3 “High-Resolution Mapping of Global Surface Water and Its Long- 2021. “Continental-Scale Building Detection from High 4 Term Changes.” Nature 540: 418–22. https://doi.org/10.1038/ Resolution Satellite Imagery.” arXiv. https://doi.org/10.48550/ 5 nature20584. arXiv.2107.12283. 6 Poggio, L., L. M. de Sousa, N. H. Batjes, G. B. M. Heuvelink, Wan, Z., S. Hook, and G. Hulley. 2021. “MODIS/Terra Land Surface A1 B. Kempen, E. Ribeiro, and D. Rossiter. 2021. “SoilGrids 2.0: Temperature/Emissivity Daily L3 Global 1km SIN Grid V061.” A2 Producing Soil Information for the Globe with Quantified Spatial NASA EOSDIS Land Processes Distributed Active Archive Center. Uncertainty.” SOIL 7 (1): 217–40. https://doi.org/10.5194/soil-7-217- https://doi.org/10.5067/MODIS/MOD11A1.061. 2021. WHO (World Health Organization). 2016. Urban Green Spaces Quandt, A. J. Herrick, G. Peacock, S. Salley, A. Buni, C. C. Mkalawa, and Health: A Review of Evidence. Copenhagen: WHO and J. Neff. 2020. “A Standardized Land Capability Classification Regional Office for Europe. https://iris.who.int/bitstream/ System for Land Evaluation Using Mobile Phone Technology.” handle/10665/345751/WHO-EURO-2016-3352-43111-60341-eng. Journal of Soil and Water Conservation 79 (5): 579–89. https://doi. pdf?sequence=3&isAllowed=y. org/10.2489/jswc.2020.00023 . Woods Ballard, B., S. Wilson, H. Udle-Clarke, S. Illman, T. Reig, P., W. Larson, S. Vionnet, and J.-B. Bayart. 2019. “Volumetric Scott, R. Ashley, and R. Kellagher. 2015. The SuDS Manual. Water Benefit Accounting (VWBA): A Method for Implementing CIRIA. https://www.ciria.org/CIRIA/CIRIA/Item_Detail. and Valuing Water Stewardship Activities.” Working Paper. World aspx?iProductCode=C753F&Category=FREEPUBS. Resource Institution. https://www.wri.org/research/volumetric- water-benefit-accounting-vwba-method-implementing-and- World Bank. 2021. A Catalogue of Nature-Based Solutions for Urban valuing-water-stewardship. Resilience. Washington, DC: World Bank. Ross, C. W., L. Prihodko, J. Anchang, S. Kumar, W. Ji, and N. P. Hanan. 2018. “HYSOGs250m, Global Gridded Hydrologic Soil Groups for Curve-Number-Based Runoff Modeling.” Scientific Data 5 (1): 180091. https://doi.org/10.1038/sdata.2018.91. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN APPENDIX B: DETAILED METHODOLOGY OF THE COASTAL NBSOS REPORT 88 THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN TABLE OF CONTENTS ABBREVIATIONS AND ACRONYMS 90 B.1 OVERVIEW OF PRODUCT, APPLICATION, AND METHODOLOGY 91 B.1.1 Overview 91 B.1.2 NBS scenarios 92 B.2 METHODOLOGY 93 B.2.1 Summary of the methodology 93 B.2.2 Overview of software/coding scheme 94 B.2.3 Mapping priority areas 95 B.2.4 Mapping NBS suitability 100 B.2.5 Mapping benefits 101 B.2.6 Selection of optimal solutions 108 B.3 APPLICATION OF THE NBSOS AS A SERVICE TO WORLD BANK TASK TEAMS 112 B.3.1 Project Task Team consultation 112 B.3.2 Interpretation, presentation, and recommendations 113 REFERENCES 114 THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN 89 TABLE OF CONTENTS MAPS MAP B.1: Summary of the methodological steps to assess flooding in the coastal NBSOS tool in Saint Lucia................................................................................................................................................ 96 MAP B.2: Example of a land use map for the baseline situation in 2020 and Manning coefficient for the different land uses in the baseline situation in Saint Lucia..................................................................... 97 MAP B.3: Steps for flood risk calculations in Saint Lucia.................................................................................... 99 MAP B.4: Flood maps for a coastal storm in Mombasa, Kenya, with a return period of 100 years in 2050.........107 MAP B.5: NBS locations in Mombasa, Kenya......................................................................................................107 MAP B.6: Areas north of Mombasa, Kenya, flooded by a coastal storm with a return period of 100 years in 2050............................................................................................................................................... 108 MAP B.7: Benefit-to-cost ratios in Mombasa, Kenya, for coral reef NBS per coastal section of 2 kilometers.... 111 FIGURES FIGURE B.1: Scenarios for NBS, consisting of baseline scenario in 2020 and future scenarios in 2050........... 103 TABLES TABLE B.1: Data layers of the coastal NBSOS from Google Earth Engine...........................................................94 TABLE B.2: Example of coral ecosystem service value function...................................................................... 105 TABLE B.3: Example of mangrove provisioning service value function............................................................ 106 TABLE B.4: Mangrove restoration cost function............................................................................................... 109 90 THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ABBREVIATIONS AND ACRONYMS AOI area of interest MPA marine protection areas API Application Programming Interface NBS nature-based solutions BCR benefit-to-cost ratio NBSOS Nature-Based Solutions Opportunity CBA cost-benefit analysis Scan CMIP6 Coupled Model Intercomparison NDVI Normalized Difference Vegetation Project Phase 6 Index COG Cloud-Optimized GeoTIFF PPP purchasing power parity EAAD expected average annual damage RCP Representative Concentration EAAP expected annual affected population Pathways EO Earth observation SCC social cost of carbon ESVD Ecosystem Services Valuation SFINCS Super-Fast INundation of CoastS  Database SSP2 Shared Socioeconomic Pathways 2 GCS Google Cloud Storage tCO2/ha/year tonnes of carbon dioxide per hectare GDP gross domestic product per year GEE Google Earth Engine WTP willingness to pay HyCreWW Hybrid Coral Reef Wave and Water level All dollar amounts are US dollars unless otherwise specified. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX B 91 ES 1 2 3 B.1 4 5 6 A1 OVERVIEW OF PRODUCT, A2 APPLICATION, AND METHODOLOGY ©A. Shuau (Obofili)_Unsplash B.1.1 OVERVIEW The coastal Nature-Based Solutions Opportunity Scan (NBSOS) identifies investment opportunities for protecting and enhancing mangroves, coral reefs, and beaches. This scanning tool highlights locations with high potential for coastal nature-based solutions (NBS) and quantifies the benefits of NBS in terms of climate change–exacerbated flood-risk reduction, tourism, blue carbon, and fisheries, among other ecosystem services. By estimating benefit-to-cost ratios (BCRs), the tool can support World Bank Task Teams, governments, and other investors to understand which NBS families have the most potential, identify potential project sites, and determine which ones provide the highest benefits. The NBSOS outputs and deliverables— including results, interpretation, recommendations, and a geospatial data package—can be prepared in approximately six weeks. The coastal NBSOS utilizes an array of openly available medium resolution (10- to 30-meter) Earth observation data and other geospatial data sets as inputs into an analytical workflow consisting of four methodological steps (see figure 2 in the main text for a graphic representation of the steps). The first step is understanding the problem: what is the spatial distribution and magnitude of coastal flood risks due to storms, surges, and climate change? The second step maps suitable areas for NBS (mangroves, coral reefs, and beaches) and the third step models and estimates benefits and costs of the possible NBS intervention to quantify BCRs. Finally, in the fourth step, it provides decision support through multicriteria and cost-benefit analyses. This appendix describes the geospatial data sets and models that constitute the coastal NBSOS (section 2) and outlines the deliverables of the coastal NBSOS as a service to World Bank Task Teams (section 3). The coastal NBSOS runs primarily in Python, with data and end products stored in Google Cloud Storage (GCS). 92 CHAPTER: APPENDIX B THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES 1 2 B.1.2 3 4 NBS SCENARIOS 5 6 The benefits and costs of coastal NBSOS are valued by comparing the present-day situation to future scenarios A1 that account for sea-level rise, economic and demographic growth, expected ecosystem degradation, and A2 potential effect of ecosystem protection and ecosystem enhancement by NBS. The NBS scenarios include changes in extent and condition of ecosystems, from degradation by anthropogenic action or, alternatively, from enhancement through NBS interventions. Those future scenarios consider a timeframe up to the year 2050. Given the uncertainties about how ecosystems could evolve in the future as a result of environmental and anthropogenic factors, and to facilitate comparability between different NBS, the approach compares the effect of losing and gaining 20 percent of the performance of each ecosystem through degradation and NBS enhancement, respectively. Furthermore, each typology of NBS is modeled independently, allowing a spatially explicit valuation of the services of each NBS and ecosystem at each coastal segment (see the example in map B.5 for coral reefs in Mombasa, Kenya). THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX B 93 ES 1 2 3 B.2 4 5 6 A1 METHODOLOGY A2 Nacala ©Stig Nygaard_Flickr B.2.1 SUMMARY OF THE METHODOLOGY To understand the problem, the increase in coastal flood risks due to climate change and growing socioeconomic exposure is calculated for coastal areas by modeling the flood extent during coastal storms with the SFINCS (Super-Fast INundation of CoastS) hydraulic model) (Leijnse et al. 2021). The resulting flood maps are overlaid with data sets of population and building footprint to quantify population and residential buildings affected by flooding during storm events. To map NBS suitability, potential NBS sites are identified via geospatial modeling of publicly available Earth observation (EO) and geospatial data sets. These data sets provide information about the current ecosystem cover, land cover, topographic, and bathymetric data. The NBS suitability assessment generates maps showing potential sites for NBS protection and enhancement. To model NBS benefits, the flood benefits of potential NBS sites are modeled in SFINCS, while the other co-benefits (tourism, blue carbon, and sustainable resource extraction, among others) are calculated using regression functions fitted to NBS benefit estimations from other projects. BCRs are estimated to identify locations where NBS investment would have the highest economic returns; the full methodology of the scanning tool is illustrated in figure 6 in the main text. In that figure, Step 1 is to understand and quantify the extent of coastal flood hazards in the baseline situation; the hydraulic model SFINCS is used to predict flood extent reduction using climate forcing (waves, extreme water levels, and scenarios of sea-level rise). Step 2 maps NBS suitability for mangroves, coral reefs, and beaches; data sets of the spatial domain (showing current topography, bathymetry, land uses, ecosystem presence) are used as input of the NBS suitability. Step 3 models NBS benefits, and SFINCS is applied to quantify coastal flood reduction by NBS interventions. 94 CHAPTER: APPENDIX B THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES Additional co-benefits are calculated using regression functions fitted to NBS estimations from other projects. 1 Step 4 supports decisions, and BCRs are calculated and mapped for scenarios with NBS. 2 3 4 To support decisions, maps showing the potential location of different NBS projects and illustrating their BCRs 5 are provided. 6 B.2.2 A1 A2 OVERVIEW OF SOFTWARE/CODING SCHEME The coastal NBSOS steps have been implemented in a pseudo-automatic framework in Python that streamlines the sourcing of input data, flood model setup and simulations, assessment of scenarios, and the spatial distribution of co-benefits and BCRs of NBS opportunities. The steps are described below. B.2.2.1 TABLE B.1: DATA LAYERS OF THE COASTAL NBSOS FROM GOOGLE EARTH ENGINE DOWNLOADING INPUT DATA DATA LAYER DATA SET NAME AND SOURCE The coastal NBSOS first identifies suitable land to support NBS protection Coral reef Allen Coral Atlas (Allen Coral Atlas 2022) extent and enhancement to assess potential NBS and their benefits by synthesizing Mangrove publicly available EO data. To ensure Global Mangrove Watch (Bunting et al. 2022) extent access to both recent and historical EO data, EO data are obtained through Land cover ESA World Cover (Earth Engine Data Catalog, no date-a) the Google Earth Engine (GEE), JRC Global Surface Water (Pekel et al. 2016) which provides access to a multi- Surface water petabyte catalogue of remote sensing OpenStreetMap (OpenStreetMap Contributors 2023) data that ranges from raw satellite observation bands to analysis-ready Google Buildings (Sirko et al. 2021) land cover products. The GEE’s Python Built-up land Application Programming Interface World Settlement Footprint (Marconcini et al. 2020) (API) is used for a seamless integration ISDA Soil Properties (Africa) (Hengl et al. 2021) of data acquisition in the coastal NBSOS Python framework. The tool Soil Open Land Soil Texture (Hengl 2018) uses a custom area of interest (AOI) characteristics that extends at least 4 kilometers into SoilGrids (Poggio et al. 2021) the ocean and an area inland defined by the user (for example, administrative Elevation FABDEM (Hawker et al. 2022) boundaries). All data layers (table B.1) are downloaded from GEE except for Roads OpenStreetMap (OpenStreetMap Contributors 2023) three OpenStreetMap data layers, Population High Resolution Settlement Layer (FCL and CIESIN 2016) which are downloaded using the OSMnx Python library (Boeing 2024). SOURCE: Original table for this publication. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX B 95 B.2.2.2 ES EO INDICATORS 1 2 3 The EO data derived from GEE are used in the development of spatial raster layers describing environmental 4 indicators that form the basis of the NBSOS’ subsequent modules. Examples of environmental indicators are 5 the Land Capability Classification Index (Quandt et al. 2020), tree canopy cover, and bare soil frequency (see 6 section 2.3). Analysis-ready indicators are centrally stored as Cloud-Optimized GeoTIFF (COG) spatial raster A1 format files in GCS, where data can be accessed via Python by each module in the coastal NBSOS. A2 B.2.3 MAPPING PRIORITY AREAS The mapping of priority areas consists of identifying high-impact areas for NBS investment within the AOI, which is guided by two main questions: (1) Where is hazard risk highest? and (2) Where could NBS be implemented to reduce hazard risks most effectively? In other words, this step focuses on characterizing the flood and erosion hazards and building and infrastructure exposed. B.2.3.1 COASTAL FLOODING Coastal flooding is characterized by the combination of storm surges, wave action, and sea-level rise. Flooding in the coastal NBSOS is simulated using the hydraulic model SFINCS, a reduced-physics solver that can simulate 2D compound flooding in coastal regions accurately but with low computational time (Leijnse et al. 2021). SFINCS has been applied and validated across various geographies, research studies, and projects (Deltares 2020; Röbke et al. 2021; Sebastian et al. 2021). In the NBSOS, the model is run for different climate forcing scenarios (for example, climate change) and geospatial scenarios (for example, NBS options). The resulting flood maps for each scenario are overlaid with population and building exposure, which allows a spatial identification of exposed population and buildings. B.2.3.2 INPUT DATA DESCRIPTION AND PROCESSING The flood model domain is defined using FABDEM (Hawker et al. 2022) topographic data and bathymetric data calculated from Sentinel-2 imagery. Extreme waves are obtained from the ERA-5 climate reanalysis (Hersbach et al. 2018) and water levels from the COAST-RP data set (Dullaart et al. 2022) that provides extreme sea levels associated with return periods between 1 and 100 years. Future scenarios in the tool include water levels that contain local sea-level rise estimates and changes in extreme water levels associated with the Representative Concentration Pathway (RCP) 8.5 scenario (Solomon et al. 2007). The flood model factors in different ground roughness values associated with existing land uses. The effects of these land uses, including ecosystems, modify the flood propagation inland by changing the bottom (ground) friction, hydraulic flows, and inundation patterns. Furthermore, coastal ecosystems can also reduce wave- 96 CHAPTER: APPENDIX B THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES driven water levels through wave breaking and frictional effects. These effects are included in the model by 1 decreasing the input water levels applied in SFINCS based on specific parameterizations described below. 2 In the baseline scenario, the effect of existing coastal ecosystems on coastal flooding is included through 3 4 additional ground roughness (that is, flood reduction) and by decreasing the wave-driven water levels according 5 to estimates based in literature (Medeiros 2023; Zhang et al. 2012) and detailed hydrodynamic studies. 6 An example of the input data sets for modeling in SFINCS is shown in map B.1. A1 A2 MAP B.1: SUMMARY OF THE METHODOLOGICAL STEPS TO ASSESS FLOODING IN THE COASTAL NBSOS TOOL IN SAINT LUCIA SOURCE: Original map for this publication, based on NBSOS data. NOTE: Inputs to the flood model are a digital elevation model, spatial distribution of land uses, information on waves and water levels, and projections of sea-level rise and storm surge for future climate scenarios. The model output consists of a flood map with water depths at a horizontal resolution of 30 × 30 meters. B.2.3.3 EFFECT OF COASTAL ECOSYSTEMS ON BOTTOM FRICTION When modeling flood inland, water flows will experience more resistance at locations with many obstacles or relatively more bottom roughness (for example, a dense mangrove forest) but will face less resistance in areas with bare ground (lower bottom roughness) and channels, where the flow is unobstructed and the water deeper. For this reason, the effect of coastal ecosystems on bottom friction is implemented through different values of the land use roughness using Manning friction coefficients, based on the literature (Zhang et al. 2012) (see the table in map B.2). In the scenarios of the baseline situation, existing ecosystems are given Manning friction coefficients based on present-day land use maps (year 2020), which correspond to values of 0.3 for healthy mangroves (Medeiros 2023); 0.1 for degraded mangroves; 0.2 for coral reefs; and 0.02 for sandy beaches (see map B.2 for an example). THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX B 97 ES MAP B.2: EXAMPLE OF A LAND USE MAP FOR THE BASELINE SITUATION IN 2020 AND MANNING COEFFICIENT FOR THE DIFFERENT LAND USES 1 IN THE BASELINE SITUATION IN SAINT LUCIA 2 3 4 5 6 A1 A2 SOURCE: Original map for this publication, based on NBSOS data. B.2.3.4 EFFECT OF COASTAL ECOSYSTEMS ON WAVE RUN-UP Although SFINCS is a reduced-complexity model capable of estimating flooding inland, the model is based on the shallow-water equations (similar to Delft3D and other flood physics-based models) and it is unable to simulate wave hydrodynamics in the surf zone and swash processes that are critically influenced by ecosystems such as reefs, mangroves, and beach and dune systems. To address this, the NBSOS represents wave effects on water levels using formulations that estimate wave run-up (defined as an increase in the mean water level from wave processes, which depends on the magnitude of offshore wave parameters, the topography of the coast, and wave propagation in the surf zone). Wave effects on the water levels by coral reefs and beaches are calculated using two separate methods. For coral reefs, the NBSOS uses the Hybrid Coral Reef Wave and Water level (HyCreWW) metamodel (Rueda et al. 2019), which was built based on a large set of run-up estimations for different reef morphologies, beach slopes, and wave and water-level conditions (Pearson et al. 2017). The metamodel calculates the top 2 percent of wave run-up (R2%), including the effects of wave setup, very low frequency waves, and infragravity 98 CHAPTER: APPENDIX B THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES waves. Run-up estimates are calculated as a function of several input parameters (offshore water level, 1 offshore wave height, offshore wavelength, fore reef slope, reef flat width, beach slope and coefficient of 2 friction). To assess the impact of corals on predicted run-up, two computations are performed: one with 3 4 the current width of the reef intact and another where the reef disappears, reducing its width to zero due to 5 degradation. This comparison enables the calculation of run-up reduction attributed to the reef. Subsequently, 6 this reduction is adjusted based on the NBS scenario and the assumed condition of coral health. A1 A2 Differently, for beaches, the Stockdon formula is applied (Stockdon et al. 2006. This formula was derived through field measurements and empirical data and is widely applied in coastal engineering. It computes the elevation exceeded by only 2 percent of all run-up or swash events (R2%) as a function of offshore wave height, offshore wavelength, and beach slope. The effect of mangroves on wave run-up is disregarded because there are no comparable run-up data sets for mangroves in the literature and their effects are included in the flood model through bottom friction. In the baseline scenario, wave run-up is estimated using the ecosystem extent of 2020, water- level data from the COAST-RP data set, wave data from ERA-5, reef widths and mean water depths obtained from the Allen Coral Atlas (2022), and reef and beach characteristics (beach slope, fore reef slope) from local fieldwork studies. B.2.3.5 ESTIMATION OF FLOOD RISK AND PRIORITY AREAS The flood maps are combined with spatial distribution of population and buildings (map B.3). The population within flood-prone areas is obtained from High Resolution Settlement Layer (FCL and CIESIN 2016). Exposed residential buildings are identified using the building area from the World Settlement Footprint (Marconcini et al. 2020), which provides the building coverage in a 90-meter grid. This information is combined with a vulnerability (or damage) curve that estimates the economic value. The method to assess assets at risk is explained in more detail below. The methods to identify population and buildings at risk of coastal flooding are explained below. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX B 99 ES MAP B.3: STEPS FOR FLOOD RISK CALCULATIONS IN SAINT LUCIA 1 2 3 4 5 6 A1 A2 SOURCE: Original map for this publication, based on NBSOS data. NOTE: Flood maps of SFINCS are overlaid with population data sets from High Resolution Settlement Layer and World Settlement Footprint to obtain the inundated population and residential buildings for each extreme event. The value and degree of damage of buildings is estimated using the methodology of Huizinga, De Moel, and Szewczyk 2017. Lastly, the expected annual affected population (EAAP) and expected average annual damage (EAAD) are calculated by trapezoidal integration of the different return periods for each year. For each flood scenario, the affected population and expected damages are calculated. The expected annual affected population (EAAP) impacted by flooding is estimated by overlaying the flood maps obtained from SFINCS with population data from the High Resolution Settlement Layer (FCL and CIESIN 2016) and integrating the population (Pi) within flooded areas for different return periods (Ti, where i corresponds with return periods of 1–100 years): The expected average annual damages (EAAD) due to flooding are computed in several steps: • Calculation of maximum damage due to flooding. The maximum damage is calculated by multiplying the number of buildings within the flood hazard zones (from the World Settlement Footprint, Marconcini et al. 2020) by the building value of each country and by a depreciation factor of 0.6 (Huizinga, De Moel, and Szewczyk 2017). Here all buildings are considered as residential, assuming that this is the most common building use. • Calculation of damage using flood-damage curves. The maximum damage values are multiplied by an average damage curve, which provides the degree of damage as a function of the local water depth. We assume a linear increase in damage from 0 to 100 percent between a water depth of 0 and 4 meters, and 100 percent damage for water depths greater than 4 meters. 100 CHAPTER: APPENDIX B THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES • Calculation of expected annual values of damages. The expected average annual damages (EAAD) are 1 obtained by the integration of the damage values (Di) from different return periods (Ti, varying between 2 1 and 100 years), according to: 3 4 5 6 A1 A2 The process outlined above can be applied to present-day distribution of people and buildings to characterize assets currently at risk. However, coastal development in flood prone areas can be a major factor of risks, especially given the economic and demographic concentration in coastal areas, as historically demonstrated by growth patterns in coastal cities (World Bank 2022). For this reason, the coastal NBSOS also considers future changes in exposure by assuming that the locations of population and assets stay the same as in the year 2020, but the building value is adjusted at the same rate as gross domestic product (GDP) projections (Shared Socioeconomic Pathways 2, or SSP2), since GDP has been historically correlated with built-up stock values. Comparison of the expected annual affected population and expected average annual damage between 2020 and 2050 enables the identification of locations with increasing flood risk due to climate change. B.2.4 MAPPING NBS SUITABILITY B.2.4.1 SUITABILITY CRITERIA FOR NBS PROJECTS For mangroves, the NBS scanning tool identifies two types of interventions: (1) conservation of mangroves by protecting areas where they exist today and remain healthy; and (2) restoration (enhancement) of mangroves at locations where they exist at present but are sparse, and also at locations where they may have been in the past but need an addition of environmental enablers to assist their restoration (sites within 100 meters of existing mangroves and at most 100 meters inland from the coastline, as a proxy for intertidal areas). In the NBS scenarios, it is assumed that healthy mangrove areas are preserved and that potential enhancement sites are vegetated and, therefore, are able to perform as healthy mangroves. For consistency across areas, the coastal NBSOS uses the Global Mangrove Watch (Bunting et al. 2018) as the most recent mangrove distribution data to identify the present-day spatial mangrove extent (the reference year is 2020). Subsequently, historical and current characteristics of soil and land cover are used to identify healthy mangroves (that is, suitable for conservation) and less healthy mangroves that could be restored (that is, suitable for enhancement). Categorization of mangrove regions into areas either suitable for preservation or enhancement is based on two EO indicators: (1) the bare soil frequency and (2) the productivity performance indicator. The bare soil frequency indicator is calculated as the number of bare soil observations from Sentinel-2 satellite data divided by the number of cloud-free Sentinel-2 observations for each pixel within the two most recent years of data (Earth Engine Data Catalog, no date-b). The productivity performance indicator aims to describe the vigor of existing THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX B 101 ES vegetation in the city. The performance productivity indicator is calculated for each unique combination of 1 landcover and soil texture within the AOI. For pixels within each unique landcover and soil texture combination, 2 the mean Normalized Difference Vegetation Index (NDVI) is calculated for each pixel for the most recent year 3 of Sentinel-2 observations and is divided by the 95th percentile of mean NDVI across all pixels in the landcover/ 4 soil texture class such that higher values represent greener vegetation. 5 6 For both indicators, a Jenks natural breaks clustering algorithm creates rasters categorizing pixels into integer A1 classes (from 1 to 5), where values equal to 1 represent the most degraded land within the AOI (high bare soil A2 frequency or low productivity performance) and values equal to 5 represent the greenest regions within the AOI (low bare soil frequency or high productivity performance). The clusters are added together to create a new raster with values ranging from 2 to 10, with lower values representing more degraded land and higher values representing existing green space to protect. The combined bare soil clusters of frequency/productivity performance are manually inspected to identify a threshold (integer value) that separates current mangrove areas suitable for protection or enhancement. Areas outside of current mangrove extent that could be suitable for enhancement are identified by filtering out all contiguous mangrove extent areas of at least 1 hectare and creating a 100-meter-wide buffer zone around them. Within these zones, pixels with a slope under 2 degrees, lower than 10 meters above mean sea level, without obstacles (no buildings or roads), and within 100 meters of a land pixel are identified as potentially suitable locations for mangrove enhancement and restoration. For coral reefs, potential enhancement and protection sites are identified based on the presence of coral reefs in shallow-water areas close to shore (between the shoreline and the 3-meter isobath) since these are most effective for flood reduction purposes and are locations where reef restoration, including artificial reefs, could be implemented (Reguero et al. 2018; Roelvink et al. 2021). Coral reef presence is identified using global coral reef extent data from the Allen Coral Atlas. Water depth is derived from Sentinel-2 imagery–derived bathymetry using the bathymetry based on Li et al. 2019 and Rueda et al. 2019. For sandy beaches, the protection of existing beaches is considered, and their conservation via nourishments so that they can keep their relative height with respect to rising sea levels. The location of existing sandy beaches is obtained by manually digitizing Google Earth images from 2023. B.2.5 MAPPING BENEFITS B.2.5.1 FLOOD RISK REDUCTION BENEFITS The flood risk reduction benefits and costs of coastal NBS projects identified by the suitability mapping are valued by comparing the present-day situation as a baseline scenario (with reference year 2020) with future scenarios (2050), which represent changes in the extent and condition of NBS as well as in socioeconomic exposure and in coastal hazards from climate change. The main scenarios are summarized below: 102 CHAPTER: APPENDIX B THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES • Baseline situation (2020). Expected flood damages and benefits/ecosystem services in the year 2020. 1 Friction coefficients and wave run-up are calculated using the baseline values of sections B.2.3.1 and 2 B.2.3.2. 3 4 5 • Year 2050, increased socioeconomic exposure. Expected flood damages and ecosystem services in 6 2050, assuming that (1) NBS performance remains as it is at present (through conservation strategies) A1 but that (2) socioeconomic exposure increases (in line with expected GDP growth as defined in the SSP2). A2 This increase in socioeconomic exposure is also included in all the scenarios listed below. Friction coefficients and wave run-up are calculated using the baseline values of sections B.2.3.1 and B.2.3.2. • Year 2050, increased socioeconomic exposure and coastal climate change. Expected flood damages and ecosystem services as of 2050, adding to the previous scenario climate change effects through sea-level rise and changes in storm surges, according to Coupled Model Intercomparison Project Phase 6 (CMIP6) climate projections (Muis et al. 2023). • Year 2050, NBS degradation. Expected flood damages and ecosystem services in 2050 considering that coastal ecosystems will degrade. For mangroves, the degradation scenario assumes that sparse patches of mangroves disappear while dense mangrove areas become unhealthy and offer 20 percent less roughness. For corals, the scenario assumes that roughness decreases by 20 percent and that the run- up reduction by coral reef is 20 percent lower, leading to an increase in run-up. For beaches, it assumes 10 percent more run-up and an annual coastal erosion rate of 2.5 percent to assign changes in co-benefits through beach surface area. • Year 2050, NBS protection. Expected flood damages and ecosystem services in 2050 considering that coastal ecosystems are protected. For mangroves and coral reefs, the protection scenario assumes that the ecosystems maintain their performance as of 2020. For beaches, it assumes they can keep up with sea-level rise, maintaining the run-up reduction of the baseline case. Therefore, sea-level rise is discarded from the total water level of coastal boundary points that are in proximity to beaches. • Year 2050, ecosystem enhancement. Expected flood damages and ecosystem services in 2050, considering that NBS projects are implemented, assuming that mangroves in sparse areas, which are deemed unhealthy in 2020, are restored to become denser and healthy, offering 20 percent more roughness. Furthermore, potential new mangrove areas are created outside the current mangrove extent; and coral reef restoration can offer 20 percent more roughness and the run-up reduction by reef is 20 percent higher, leading to a reduction in run-up; beaches behave identically to the protection scenario. To value each coastal NBS individually, the scenarios are simulated individually for each ecosystem (figure B.1). This is done for both enhancement and protection actions. In the case of enhancement, only one type of ecosystem is enhanced in each simulation while the others maintain their condition as of 2020. In the case of protection, only one type of ecosystem is maintained at its 2020 condition while the others degrade. Beaches represent an exception, as there is only a protection scenario where they keep up with sea-level rise maintaining the run-up reduction of the baseline case: • Mangrove enhancement only. This assumes that only mangrove enhancement NBS are implemented, whereas corals and beaches remain in their condition as of 2020. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX B 103 ES • Coral enhancement only. This assumes that only coral enhancement NBS are implemented, whereas 1 mangroves and beaches remain in their condition as of 2020. 2 3 • Mangrove protection only. This assumes that only mangrove protection NBS are implemented, whereas 4 corals and beaches degrade. 5 6 • Coral protection only. This assumes that only coral protection NBS are implemented, whereas mangroves A1 and beaches degrade. A2 • Beach protection only. This assumes that only beach protection NBS are implemented, whereas corals and mangroves degrade. FIGURE B.1: SCENARIOS FOR NBS, CONSISTING OF BASELINE SCENARIO IN 2020 AND FUTURE SCENARIOS IN 2050 SOURCE: Original figure for this publication. NOTE: This is a slightly simplified version of figure 8 in the main text. B.2.5.2 CARBON SEQUESTRATION BENEFITS The value of additional carbon sequestration by mangroves is estimated using methods and parameters from the literature (Murray et al. 2011; Pendleton et al. 2012): • Computation of additional carbon sequestration from NBS multiplies the cumulative additional mangrove area by a representative carbon sequestration rate per unit area: 6.3 tonnes of carbon dioxide per hectare per year (tCO2 /ha/year) (Pendleton et al. 2012). 104 CHAPTER: APPENDIX B THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES • Carbon stored per year is valued economically using the social cost of carbon (SCC), which is the monetary 1 value of damages caused by emitting 1 more tonne of CO2 in a given year (Pearce 2003). The SCC, therefore, 2 represents the value of damages avoided for a small reduction in emissions—in other words, the benefit 3 4 of a reduction in atmospheric CO2 in a given year. The SCC increases over time as a result of the increasing 5 marginal damage caused by additional tonnes of CO2 in the atmosphere. The NBSOS uses the US Interagency 6 Working Group series of SCC estimates for the period 2010–2050 (Interagency Working Group 2016), which A1 are in the range of World Bank carbon pricing scenarios consistent with a 2°C global warming scenario. A2 B.2.5.3 OTHER BENEFITS The values of other ecosystem services are estimated using data from the literature and value functions. Data on corals and mangroves from the Ecosystem Services Valuation Database (ESVD) (Brander et al. 2024) are used in regression analyses to estimate functions that relate the value of coral and mangrove services to the characteristics and context of each ecosystem.11 These regressive functions are subsequently applied to predict location-specific benefits accounting for variation in relevant explanatory factors (for example, the size of the ecosystem, population density, income of beneficiaries). An example of an estimated value function for coral ecosystem services is shown in table B.2, where the dependent variable is defined as US dollars per hectare per year and the explanatory variables include the area of the ecosystem patch in hectares and binary variables that indicate the ecosystem services, the Global Human Modification map, the Biodiversity Intactness Index of the ecosystem, the population density within a 30-kilometer radius of the ecosystem, and the GDP per capita (also for the 30-kilometer radius from the valued ecosystem). The explanatory variables have expected signs in terms of how they influence variation in ecosystem services values but they are not all statistically significant. For the example, in table B.2, the overall explanatory power (R2) of the model is relatively low. Ecosystem service values per unit area decline slightly with the size of the ecosystem patch (that is, total values increase less than proportionately with the size of the ecosystem) and the extent of human disturbance (anthropogenic modification) but increase with biodiversity intactness. Population density and income both have positive correlations with ecosystem services values, representing demand-side factors. 11 Details about the ESVD can be found at https://www.esvd.net/. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX B 105 ES TABLE B.2: EXAMPLE OF CORAL ECOSYSTEM SERVICE VALUE FUNCTION 1 2 EXPLANATORY VARIABLE COEFFICIENT P-VALUE 3 4 Constant −1.758 0.599 5 6 Area (hectares; ln) −0.074 0.288 A1 A2 Raw materials 3.579 0.219 Erosion regulation 1.928 0.149 Existence bequest 1.759 0.109 Extreme event regulation 2.457 0.066 Recreation tourism 1.500 0.154 Waste treatment 3.744 0.010 Aesthetic enjoyment 3.725 0.006 Food 1.285 0.248 Cognitive development −2.187 0.145 Inspiration 1.067 0.714 Human Modification index −2.890 0.010 Biodiversity Intactness Index 3.362 0.157 Population density (ln) 0.634 0.001 GDP per capita (US dollars; ln) 0.102 0.556 Adjusted R2 0.143 N 255 SOURCE: Brander et al. 2024. NOTE: The dependent variable is US dollars/hectare/year (ln). GDP = gross domestic product; ln = natural logarithm transformation; N = number of coral reefs. An example of an estimated value function for mangrove provisioning services is presented in table B.3. The NBSOS tool focuses on provisioning services to address commodity consistency in the values, while regulating service values such as carbon sequestration and flood mitigation are valued separately. The dependent variable is defined as US dollars per hectare per year. The explanatory variables include the area of the ecosystem patch in hectares, the population density within a 30-kilometer radius of the ecosystem, the GDP per capita also for the 30-kilometer radius of the valued ecosystem, the human modification index of the ecosystem, and the percentage of the area within 30-kilometer radius of the ecosystem that is designated as protected area. The explanatory variables have expected signs in terms of how they influence variation in 106 CHAPTER: APPENDIX B THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES ecosystem services values and are all statistically significant. The overall explanatory power is comparable 1 to similar analyses in the literature. Ecosystem service values per unit area decline slightly with the size of 2 the ecosystem patch (that is, total values increase less than proportionately with the size of the ecosystem), 3 4 with the extent of human disturbance (human modification), and the extent of protected area designation, 5 which is likely to reduce direct use of provisioning services. Population density and income both have positive 6 correlations with ecosystem service values, representing demand-side factors. A1 A2 TABLE B.3: EXAMPLE OF MANGROVE PROVISIONING SERVICE VALUE FUNCTION EXPLANATORY VARIABLE COEFFICIENT P-VALUE Constant −0.144 0.935 Area (hectares; ln) −0.188 0.001 GDP per capita (US dollars; ln) 0.617 0.001 Population density (ln) 0.520 0.001 Human Modification index −2.520 0.001 Protected area (% of area) −0.288 0.011 Adjusted R2 0.179 N 371 SOURCE: Brander et al. 2024. NOTE: The dependent variable is US dollars/hectare/year (ln). GDP = gross domestic product; ln = natural logarithm transformation; N = number of mangrove sites. The estimation of co-benefits from beach nourishment is focused on the value of maintaining the width of beaches for tourism. Based on results of a discrete choice experiment conducted on Anguilla that estimated the willingness to pay (WTP) of international tourists to maintain beach width (Tieskens et al. 2014), beach loss is valued at $9.91 per unit meter of shoreline retreat. Annual tourist arrivals are obtained from national tourism statistics of the case study country governments and spatially distributed to beaches using information on hotel locations. Hotel locations were derived from listings on Booking.com, extracted using Python scraping. For each coastal section of 2 kilometers, a weighted sum was calculated of the number of hotels within a radius of 1, 2, and 5 kilometers with respective weights of 0.5, 0.375, and 0.125. Tourists were then allocated to each coastal section based on the relative value of this weighted sum. In addition, the added value of accommodation expenditure was estimated using estimates based on available local data on tourism arrivals and expenditures (that is, number of nights per tourist and average hotel accommodation spending). Seventy percent of added value from accommodation was estimated to be attributed to the presence of sandy beaches, in line with similar studies conducted in the Caribbean. For each beach, the tourism value of beach nourishment is computed as product of (1) the sum of the estimated WTP and added value of hotel expenditures attributed to sandy beaches, (2) the annual number of tourist visitors, and (3) the annual loss of beach width (assumed at 2.5 percent). THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX B 107 B.2.5.4 ES VISUALIZATION AND MAPPING 1 2 3 The outputs of the scanning tool are mapped below for an area around Mombasa, Kenya, as an example, and 4 consist of: 5 6 • Coastal flood hazard maps at present and by mid-century as well as built-up areas and population at risk. A1 These maps can be generated for different return periods. Map B.4 shows an example of a storm in 2050 A2 with a return period of 100 years. • NBS opportunities: potential sites for nature-based adaptation using mangroves, beaches, and coral reefs (map B.5). • Reduction of flood risks by NBS opportunities by mid-century. Results for a storm with a return period of 100 years are shown in map B.6. MAP B.4: FLOOD MAPS FOR A COASTAL STORM IN MOMBASA, KENYA, MAP B.5: NBS LOCATIONS IN MOMBASA, KENYA WITH A RETURN PERIOD OF 100 YEARS IN 2050 SOURCE: Original map for this publication. Basemap by Esri. SOURCE: Original map for this publication. Basemap by Esri. NOTE: Blue areas show flooded regions, with different shades of blue NOTE: Mangroves are shown in green, coral reefs in blue, and sandy depending on the height of the water level with respect to mean sea beaches in orange. level. Red dots show areas with buildings exposed to flooding. 108 CHAPTER: APPENDIX B THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES MAP B.6: AREAS NORTH OF MOMBASA, KENYA, FLOODED BY A COASTAL STORM WITH A RETURN PERIOD OF 100 YEARS IN 2050 1 2 3 4 5 6 A1 A2 SOURCE: Original map for this publication. Basemap by Esri. NOTE: Blue areas show locations where flood can be prevented by enhancement measures, and red areas show sites that would be flooded if degradation of ecosystems takes place. B.2.6 SELECTION OF OPTIMAL SOLUTIONS To calculate the costs and benefits for different NBS in different areas within the project AOI, NBSOS divides the total coastline into sections. All identified 10-meter pixels of mangroves, coral reefs, and beaches are then assigned to the closest 2-kilometer section to create separate patches of each NBS per coastal section. Both avoided flood damages and co-benefits, as well as protection and restoration costs, are calculated for each 2 kilometer coastal section for each NBS to provide a spatially explicit cost-benefit analysis (CBA). B.2.6.1 COSTS OF NBS The costs of implementing NBS using coral reefs, mangroves, and beaches are estimated using data on ecosystem restoration costs from the literature. In the case of coral reefs and mangrove enhancement, available data are used in meta-regression analyses to estimate functions that relate the costs of restoration to the characteristics and context of the restoration activity. These functions are subsequently applied to predict location-specific costs accounting for variation in relevant explanatory factors (for example, the size of THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX B 109 ES the restoration site, type of intervention, purchasing power parity or PPP). In the case of beaches, unit costs of 1 beach nourishment are obtained from the literature and adjusted at the country level to account for differences 2 in price levels. All costs are adjusted to 2020 price levels using GDP deflator factors from the World Bank World 3 Development Indicators.12 To identify the cost of reef restoration, the cost per meter of constructing artificial 4 reefs is estimated. The cost for such a hybrid solution is based on a median value identified by Ferrario et al. 5 (2014) of $1,290 per linear meter. 6 A1 Data on the costs of mangrove restoration are obtained from Bayraktarov et al. (2016) and Su, Friess, and A2 Gasparatos (2021). Country-level information on PPP and GDP per capita are added to the data set; these are also from the World Development Indicators. An ordinary least squares regression model is estimated with the dependent variable defined as restoration cost in US dollars per hectare. The explanatory variables include the area of the restoration site in hectares, a binary variable indicating whether the intervention includes both hydrological restoration and mangrove planting, the PPP factor for the country, the GDP per capita, and the number of years over which the restoration activities are implemented. An example of estimated mangrove cost function is presented in table B.4. The explanatory variables have expected signs in terms of how they influence variation in costs and are mostly statistically significant. The overall explanatory power is relatively high. Mangrove restoration that includes both hydrological works and planting has substantially higher costs. Restoration costs also increase with higher price levels, income per capita, and the number of years over which the intervention is implemented. Costs per unit area decline slightly with the size of the restoration site—that is, total costs increase less than proportionately with the size of the restoration site. TABLE B.4: MANGROVE RESTORATION COST FUNCTION EXPLANATORY VARIABLE COEFFICIENT P-VALUE Constant −2.14 0.210 PPP factor 2.245 0.079 GDP per capita (ln) 0.938 0.001 Site area (hectares; ln) −0.005 0.929 Cost years (ln) 0.748 0.003 Restoration hydrological and planting 1.274 0.002 Adjusted R2 0.645 N 132 SOURCE: Su, Friess, and Gasparatos 2021. Note: The dependent variable is US dollars/hectare (ln); PPP = purchasing power parity; ln = natural logarithm transformation; N = number of mangrove sites. 12 See the World Bank DataBank at https://databank.worldbank.org/source/world-development-indicators. 110 CHAPTER: APPENDIX B THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES The cost of beach nourishment is obtained from Spencer, Strobl, and Campbell (2022), which reports a cost 1 of $2,083.11/meter in 2019 prices. This unit cost is adjusted for each country using PPP factors from the World 2 Development Indicators to reflect differences in general price levels. 3 4 5 Mangrove and coral reef protection costs were calculated using the establishment and maintenance cost of 6 marine protection areas (MPA). Estimations on the per hectare cost of MPA establishment and maintenance A1 were obtained as described by McCrea-Strub et al. (2011), using an MPA size of 2 square kilometers for coral reef A2 protection and 50 square kilometers for mangrove protection to account for the difference in magnitude. B.2.6.2 COST-BENEFIT ANALYSIS CBA is an economic methodology used to compare the costs and benefits of a proposed investment over a period of time. Applications of CBA of public investments generally take a broad perspective and aim to incorporate all relevant societal (welfare economic) benefits and costs. In a CBA, future costs and benefits are converted to and aggregated as “present values” using a discount rate. The discount rate represents the annual rate at which costs and benefits depreciate because people place a higher value on the present than on the future. This depreciation reflects general uncertainties about the future and the opportunity cost of investing capital in any given project when money invested elsewhere could have yielded equal or greater returns. In the present CBA, costs and benefits are estimated over a 28-year time horizon (2023–50) using a discount rate of 6 percent. For beaches, NBSOS presents a CBA for protecting the current extent and functions of beaches, comparing the avoided flood damages and co-benefits of the 2050 degradation scenario to the 2050 beaches protection scenario. For mangroves and coral reefs, NBSOS presents a CBA for protection, comparing the 2050 degradation scenario to the 2050 mangrove/coral reef protection scenario; and a CBA for enhancement, comparing the 2050 avoided flood damages and co-benefits of the baseline scenario to the mangrove/coral reef enhancement scenario. NBS investment options are evaluated in this report using the BCR (present value benefits minus present value costs). A BCR greater than 1 indicates a positive economic return on investment. The analysis is conducted at the level of individual patches of each ecosystem type to which restoration interventions could be applied. This enables the assessment of the relative economic viability of potential interventions across locations and the prioritization of those that yield the greatest benefits and the lowest cost. Priority areas are identified based on their BCRs, as shown in map B.7, which highlights locations with high investment potential. By selecting areas with relatively high BCR, the NBS opportunities can be organized into investment scenarios. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX B 111 ES MAP B.7: BENEFIT-TO-COST RATIOS IN MOMBASA, KENYA, FOR CORAL REEF NBS PER COASTAL SECTION OF 2 KILOMETERS 1 2 3 4 5 6 A1 A2 BEACH PROTECTION - BCR <1 1-3 3-5 >5 SOURCE: Original map for this publication. Basemap by Esri. NOTE: BCR = benefit-to-cost ratio. 112 CHAPTER: APPENDIX B THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES 1 2 3 B.3 4 5 6 A1 A2 APPLICATION OF THE NBSOS AS A SERVICE TO WORLD BANK TASK TEAMS ©Benjamin Jones_Unsplash B.3.1 PROJECT TASK TEAM CONSULTATION Consultations with local Task Teams are held at the beginning of each coastal NBSOS application. During these meetings, different types of relevant information are collected. This helps to provide a customized service that considers local characteristics, necessities, and data. 1. Determination of the AOI. The AOI usually covers the full coastal area of a country. Results can be provided for particular AOIs requested by local Task Teams. 2. Identification of relevant environmental hazards and associated weights. The main challenge of the coastal NBSOS is coastal flood risk due to coastal storms and surges. 3. Selection of relevant NBS families. Relevant NBS families (mangroves, corals, and/or sandy beaches) for each case are also discussed with local Task Teams. 4. Availability of supplemental local data sets. The availability of local data to enhance and customize the NBSOS analysis is also discussed with the local Task Team. Even though the analysis can be performed using available global data, the addition of local data can improve some results or their visualization. For instance, population vulnerability data, which usually are not available as open data, can improve coastal flood risk analyses. Administrative divisions’ data can help to present results per neighborhood or district, helping to better communicate the outputs of the NBSOS. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX B 113 ES B.3.2 1 2 INTERPRETATION, PRESENTATION, AND RECOMMENDATIONS 3 4 5 B.3.2.1 6 A1 DELIVERABLES A2 The coastal NBSOS results are delivered as a slide deck that includes a summary of the methodology, contextual analysis, background (for example, local hazards), and results of the NBSOS. These outputs include the identification of priority areas from hazard-exposure analysis, NBS opportunity maps, and the CBA results. The slide deck includes an interpretation of results and recommendations from the analysis, as well as general design and implementation considerations. The outputs are also provided in a geospatial format, including raster data of the suitability of NBS families (GeoTIFF), shapefiles describing climate resilience and coastal flood challenges, and BCRs of the NBS families. B.3.2.2 PRESENTATION TO TASK TEAMS AND CLIENTS The slide deck is presented to the Task Team in a meeting where feedback about the preliminary results, their interpretation, and recommendations is also sought. These suggestions are integrated into a final version of the NBSOS results. B.3.2.2 PARTICIPATORY MODELS FOR CLIENT AND STAKEHOLDER ENGAGEMENT In some cases, further engagements with local Task Teams may be possible and may include activities to improve the use of the outputs by local stakeholders, such as presentations of results, interactive activities to facilitate discussions among local actors, and support during pre-feasibility studies. 114 CHAPTER: APPENDIX B THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN REFERENCES ES Earth Engine Data Catalog. No date-b. Harmonized Sentinel-2 MSI: MultiSpectral Instrument, Level-2A (data set). Provider 1 European Union/ESA/Copernicus. https://developers.google. 2 com/earth-engine/datasets/catalog/COPERNICUS_S2_SR_ 3 HARMONIZED. 4 5 Ferrario, F., M. W. Beck, C. Storlazzi, F. Micheli, C. C. Shepard, and Allen Coral Atlas. 2022. Imagery, maps and monitoring of L. Airoldi. 2014. “The Effectiveness of Coral Reefs for Coastal 6 the world’s tropical coral reefs. https://doi.org/10.5281/ Hazard Risk Reduction and Adaptation.” Nature Communications A1 zenodo.3833242. 5: 3794. https://doi.org/10.1038/ncomms4794. A2 Bayraktarov, E. M. I. Saunders, S. Abdullah, M. Mills, J. Beher, H. P. FCL and CIESIN (Facebook Connectivity Lab and Center for Possingham, P. J. Mumby, and C. E. Lovelock. 2016. “The Cost and International Earth Science Information Network), Columbia Feasibility of Marine Coastal Restoration.” Ecological Applications University. 2016. High Resolution Settlement Layer (HRSL). 26 (4): 1055–74. https://doi.org/10.1890/15-1077. Source imagery for HRSL, copyright DigitalGlobe. https:// dataforgood.facebook.com/dfg/tools/high-resolution- Boeing, G. 2024. “Modeling and Analyzing Urban Networks and settlement-layer. Amenities with OSMnx.” Working paper. https://geoffboeing.com/ publications/osmnx-paper/. Hawker, L., P. Uhe, L. Paulo, J. Sosa, J. Savage, C. Sampson, and J. Neal. 2022. “A 30 m Global Map of Elevation with Forests Brander, L. M., R. de Groot, V. Guisado Goñi, V. van ‘t Hoff, P. and Buildings Removed.” Environmental Research Letters 17 (2): Schägner, S. Solomonides, A. McVittie, F. Eppink, M. Sposato, 024016. https://doi.org/10.1088/1748-9326/ac4d4f. L. Do, A. Ghermandi, and M. Sinclair. M. 2024. Ecosystem Services Valuation Database (ESVD). Foundation for Sustainable Hengl, T. 2018. “Soil Texture Classes (USDA System) for 6 Soil Development and Brander Environmental Economics. https:// Depths (0, 10, 30, 60, 100 and 200 cm) at 250 m.” Zenodo, December www.esvd.net/. 24, 2018. https://zenodo.org/records/2525817. Bunting, P., A. Rosenqvist, L. Hilarides, R. Lucas, T. Thomas, Hengl, T., M. A. E. Miller, J. Križan, et al. 2021. “African Soil T. Tadono, T. Worthington, M. Spalding, N. Murray and L.-M. Properties and Nutrients Mapped at 30 m Spatial Resolution Using Rebelo. 2022. “Global Mangrove Extent Change 1996–2020: Two-Scale Ensemble Machine Learning.” Scientific Reports 11 (1): Global Mangrove Watch Version 3.0.” Remote Sensing 14 (15): 3657. 6130. https://doi.org/10.1038/s41598-021-85639-y. https://doi.org/10.3390/rs14153657. Hersbach, H., B. Bell, P. Berrisford, G. Biavati, A. Horányi, J. Muñoz Bunting P., A. Rosenqvist, R. Lucas, L.-M. Rebelo, L. Hilarides, Sabater, J. Nicolas, C. Peubey, R. Radu, I. Rozum, D. Schepers, A. N. Thomas, A. Hardy, T. Itoh, M. Shimada, and C. M. Finlayson. Simmons, C. Soci, D. Dee, and J.-N. Thépaut. 2018. ERA5 hourly 2018. “The Global Mangrove Watch – a New 2010 Global Baseline data on single levels from 1940 to present. Copernicus Climate of Mangrove Extent.” Remote Sensing 10(10): 1669. https://doi. Change Service (C3S) Climate Data Store (CDS). https://doi. org/10.3390/rs10101669. org/10.24381/cds.adbb2d47. Deltares. 2020. “Assessing the Impact of Sea Level Rise and Huizinga, J., H. de Moel, and W. Szewczyk. 2017. “Global Flood Resilience Potential in the Caribbean. Revisiting Resilience in Depth-Damage Functions: Methodology and the Database with the Caribbean: A 360-Degree Approach.” Background Paper, Guidelines.” JRC Research Reports. JRC105688, Joint Research World Bank. https://documents1.worldbank.org/curated/ Centre. https://publications.jrc.ec.europa.eu/repository/handle/ en/599791635297691305/pdf/360-Resilience-A-Guide-to- JRC105688. Prepare-the-Caribbean-for-a-New-Generation-of-Shocks- Assessing-the-Impact-of-Sea-Level-Rise-and-Resilience- Interagency Working Group on Social Cost of Greenhouse Gases, Potential-in-the-Caribbean.pdf. United States Government). 2016. Technical Support Document: Technical Update of the Social Cost of Carbon for Regulatory Dullaart, J. C., M. Muis, N. Bloemendaal, M. V. Chertova, A. Impact Analysis Under Executive Order 12866. https://www. Couasnon, and J. C. J. H. Aerts. 2022. COAST-RP: A global epa.gov/sites/default/files/2016-12/documents/sc_co2_tsd_ COastal dAtaset of Storm Tide Return Periods. Version 2. 4TU. august_2016.pdf. ResearchData. Data Set. https://doi.org/10.4121/13392314.v2. Leijnse, T., M. van Ormondt, K. Nederhoff, and A. van Dongeren. Earth Engine Data Catalog. No date-a. ESA WorldCover 10m v100 2021. “Modeling Compound Flooding in Coastal Systems Using (data set). Provider ESA/VITO/Brockmann Consult/CS/GAMMA a Computationally Efficient Reduced-Physics Solver: Including Remote Sensing/IIASA/WUR. https://developers.google.com/ Fluvial, Pluvial, Tidal, Wind- and Wave-Driven Processes,” earth-engine/datasets/catalog/ESA_WorldCover_v100. Coastal Engineering 163: 103796. https://doi.org/10.1016/j. coastaleng.2020.103796. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN CHAPTER: APPENDIX B 115 Li, J., D. E. Knapp, S. R. Schill, C. Roelfsema, S. Phinn, M. Silman, J. Poggio, L., L. M. de Sousa, N. H. Batjes, G. B. M. Heuvelink, ES Mascaro, and G. P. Asner. 2019. “Adaptive Bathymetry Estimation B. Kempen, E. Ribeiro, and D. Rossiter. 2021. “SoilGrids 2.0: 1 for Shallow Coastal Waters Using Planet Dove Satellites.” Remote Producing Soil Information for the Globe with Quantified Spatial 2 Sensing of Environment 232: 111302. https://doi.org/10.1016/j. Uncertainty.” SOIL 7 (1): 217–40. https://doi.org/10.5194/soil-7-217- rse.2019.111302. 2021. 3 4 Marconcini, M., A. Metz-Marconcini, S. Üreyen, et al. 2020. Quandt, A., J. Herrick, G. Peacock, S. Salley, A. Buni, C. C. 5 “Outlining Where Humans Live, the World Settlement Footprint Mkalawa, and J. Neff. 2020. “A Standardized Land Capability 6 2015.” Scientific Data 7 (1): 242. https://doi.org/10.1038/s41597- Classification System for Land Evaluation Using Mobile Phone A1 020-00580-5. Technology.” Journal of Soil and Water Conservation 75 (5): 579–89. A2 https://doi.org/10.2489/jswc.2020.00023. McCrea-Strub, A., D. Zeller, U. R. Sumaila, J. Nelson, A. Balmford, and D. Pauly. 2011. “Understanding the Cost of Establishing Reguero, B. G., M. W. Beck, V. N. Agostini, P. Kramer, and B. Marine Protected Areas.” Marine Policy 35 (1): 1–9. https://doi. Hancock. 2018. “Coral Reefs for Coastal Protection: A New org/10.1016/j.marpol.2010.07.001. Methodological Approach and Engineering Case Study in Grenada.” Journal of Environmental Management 210: 146–61. Medeiros, S. 2023. “Hydraulic Bottom Friction and Aerodynamic https://doi.org/10.1016/j.jenvman.2018.01.024. Roughness Coefficients for Mangroves in Southwest Florida, USA.” Journal of Marine Science and Engineering 11 (11): 2053. Röbke, B., T. Leijnse, G. Winter, M. van Ormondt, J. van Nieuwkoop, https://doi.org/10.3390/jmse11112053. and R. de Graaff. 2021. “Rapid Assessment of Tsunami Offshore Propagation and Inundation with D-FLOW Flexible Mesh and Muis, S., J. C. J. H. Aerts, J. A. A. Antolinez, et al. 2023. “Global SFINCS for the 2011 Tōhoku Tsunami in Japan.” Journal of Marine Projections of Storm Surges Using High-Resolution CMIP6 Science and Engineering 9 (5): 453. https://doi.org/10.3390/ Climate Models.” Earth’s Future 11 (9): e2023EF003479. https://doi. jmse9050453. org/10.1029/2023EF003479. Roelvink, F. E., C. D. Storlazzi, A. R. van Dongeren, and S. G. Murray, B. C., L. Pendleton, W. A. Jenkins, and S. Sifleet. 2011. Pearson. 2021. “Coral Reef Restorations Can Be Optimized to Green Payments for Blue Carbon: Economic Incentives for Reduce Coastal Flooding Hazards.” Frontiers in Marine Science 8. Protecting Threatened Coastal Habitats. Nicholas Institute for https://doi.org/10.3389/fmars.2021.653945. Environmental Policy Solutions, Report NI R 11-04. https:// nicholasinstitute.duke.edu/environment/publications/ Rueda, A., L. Cagigal, S. G. Pearson, J. A. A. Antolínez, C. Storlazzi, naturalresources/blue-carbon-report. A. van Dongeren, P. Camus, and F. J. Mendez. 2019. “HyCReWW: A Hybrid Coral Reef Wave and Water Level Metamodel.” OpenStreetMap Contributors. 2023. Planet Dump. https://planet. Computers & Geosciences 127: 85–90. https://doi.org/10.1016/j. osm.org. cageo.2019.03.004. Pearce, D. 2003. “The Social Cost of Carbon and Its Policy Sebastian, A., D. J. Bader, C. M. Nederhoff, T. W. B. Leijnse, J. D. Implications.” Oxford Review of Economic Policy 19 (3)? 362–84. Bricker, and S. G. J. Aarninkhof. 2021. “Hindcast of Pluvial, Fluvial, https://doi.org/10.1093/oxrep/19.3.362. and Coastal Flood Damage in Houston, Texas during Hurricane Harvey (2017) Using SFINCS.” Natural Hazards 109: 2343–62. Pearson, S. G., C. D. Storlazzi, A. R. van Dongeren, M. F. S. Tissier, https://doi.org/10.1007/s11069-021-04922-3. and A. J. H. M. Reniers. 2017. “A Bayesian-Based System to Assess Wave-Driven Flooding Hazards on Coral Reef-Lined Coasts.” JGR Sirko, W., S. Kashubin, M. Ritter, A. Annkah, Y. S. E. Bouchareb, Oceans 122 (12): 10099–117. https://doi.org/10.1002/2017JC013204. Y. Dauphin, D. Keysers, M. Neumann, M. Cisse, and J. Quinn. 2021. “Continental-Scale Building Detection from High Pekel, J. -F., A. Cottam, N. Gorelick, and A. S. Belward. 2016. Resolution Satellite Imagery.” arXiv. https://doi.org/10.48550/ “High-Resolution Mapping of Global Surface Water and Its Long- arXiv.2107.12283. Term Changes.” Nature 540: 418–22. https://doi.org/10.1038/ nature20584. Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Averyt, M. Tignor, and H.L. Miller, eds. 2007. Climate Change 2007: The Pendleton, L., D. C. Donato, B. C. Murray, et al. 2012, “Estimating Physical Science Basis. Contribution of Working Group I to the Global ’Blue Carbon’ Emissions from Conversion and Degradation Fourth Assessment Report of the Intergovernmental Panel on of Vegetated Coastal Ecosystems.” PLoS ONE 7. https://doi. Climate Change. Cambridge, United Kingdom and New York, NY, org/10.1371/journal.pone.0043542. USA: Cambridge University Press. 116 CHAPTER: APPENDIX B THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN ES Spencer, N., E. Strobl, and A. Campbell. 2022. “Sea Level Rise Tieskens, K. F., S. W. Schep, and P. van Beukering, I. J. M van Beek, under Climate Change: Implications for Beach Tourism in the and E.M. Wolfs. 2014. “Mapping the Economic Value of Ecosystems 1 Caribbean.” Ocean & Coastal Management 225: 106297. https://doi. on St Eustatius.” IVM Institute for Environmental Studies. https:// 2 org/10.1016/j.ocecoaman.2022.106207. www.dcbd.nl/sites/default/files/documents/R14-13-Value- 3 mapping-of-nature-on-St-Eustatius.pdf. 4 Stockdon, H. F., R. A. Holman, P. A. Howd, and A. H. Sallenger. 5 2006. “Empirical Parameterization of Setup, Swash, and Runup.” World Bank. 2022. Living on the Water’s Edge: Flood Coastal Engineering 53 (7): 573–88. https://doi.org/10.1016/j. Risk and Resilience of Coastal Cities in Sub-Saharan 6 coastaleng.2005.12.005. Africa. World Bank Group. https://documents1. A1 worldbank.org/curated/en/099042623122580007/pdf/ A2 Su, J., D. A. Friess, and A. Gasparatos. 2021. “A Meta-Analysis of P17703604875c205e0a95b0d3f8997b32c8.pdf. the Ecological and Economic Outcomes of Mangrove Restoration.” Nature Communications 12 (1): 5050. https://doi.org/10.1038/ Zhang, K., H. Liu, Y. Li, H. Xu, J. Shen, J. Rhome, and T. J. Smith. s41467-021-25349-1. 2012. “The Role of Mangroves in Attenuating Storm Surges.” Estuarine, Coastal and Shelf Science 102–03: 11–23. https://doi. org/10.1016/j.ecss.2012.02.021. THE NATURE-BASED SOLUTIONS OPPORTUNITY SCAN LEVERAGING EARTH OBSERVATION DATA TO IDENTIFY INVESTMENT OPPORTUNITIES IN NBS FOR CLIMATE RESILIENCE IN CITIES AND COASTS ACROSS THE WORLD