Le Corridor de Développement du Nord, Haïti Étude sur le Développement Urbain et le Changement Climatique

[page 1] ER ASS ra iii À LL à LA 2 : Æ” : + + ES ua 77) à LD —_ El . - PT C4 | AIDES. FPE 0) MALE d | * | F” æ À Na! LUE mi Le > En # Ÿ és The Northern Development € ” À; € , - FA À Corridor, Haïti CASA Ta de 7 ani Urban Devel dci PT <<. D * Nil: he s À rban Development and Climate RS ce 20 Tan : . d( © > #5 | Change Study (Final Draft) DS PP 14) n AC k ft Le : PP) = ERPD IA ere) 1 Fe LM 25h: February 2015 8 RP. 4 vd y TA ef | Environmental Resources Management Le d | Mr Set 2 G oj 0 # fPéfcheë > Lo WWW.erm.com ss. EMERGING «: 4 KE Initiative ERM [page 2] TABLE OF CONTENTS 4.7 Future Studies 34 8.1 Study Focus 93 5. Future Growth Projections 37 8.2 Smart Development Scenario 93 List of Appendices i : 5.1 Future Development Projects 37 8.3 Challenges to be Addressed 95 List of Figures ï . : 5.2 Population and Demographics 37 BIBLIOGRAPHY 98 List of Tables iv 5.3 Urban Area Needs 41 Acranyms “6. Geospatial Model 42 . " Sospatia Moce LIST OF APPENDICES Acknowledgements vi . 6.1 Introduction 42 1. Introduction 1 = Indivi j uçtl 6.2 The Geospatial Modelling Process 42 APPENDIX 1: Individual GIS Maps for the Ecological 1.1 Background and ESCI 1 : System 6.3 Modelling for the NDC 43 12 ESCiin Haïti 1 6.4 Restriction Factors Sub-Model 44 APPENDIX 2: Individual GIS Maps for Urban and 1.3 Urban Development and Climate Cha Study 3 VE op NAS Imate 18e 6.5 Attractions Factors Sub-Model 47 Infrastructure Development 2. Methodology and Approach a 6.6 Future Development Projects Sub-Model 50 APPENDIX 3: Climate Studies by the University of itabili j West Indies 2.1 Methodology for the Risk and Urban Studies 4 6.7 Suitability Analysis 50 2.2 Study Area n 7. ga etopment of a Sustainable Growth Scenario APPENDIX 4: Hazard Profiles 2.3 Building upon Key Planning Efforts 5 Len . 7.1 Land Suitability 53 APPENDIX 5: Characteristics of Assets Exposed 3. Baseline Conditions 9 um 7.2 Densification 55 3.1 Current Study Area 9 APPENDIX 6: Impacts and Losses 7.3 Capacity of Existing Townships 57 3.2 Current and Historical Land Cover 13 7.4 Capacity in Trou-du-Nord 59 APPENDIX 7: Restrictions Maps 3.3 Physical, Biological and Hydrological Baselines 75 Capacity in Limonade 64 16 ‘ Pacity In mon APPENDIX 8: Attractions Maps 3.4 Cultural Heritage 19 7.6 Capacity in Terrier Rouge 68 Hi ï APPENDIX 9: Future Development Maps 3.5 Urban, Commercial and Infrastructure 20 7.7 Capacity in Bord de Mer de Limonade 72 4. Hazard and Risk Assessment Studies 21 7.8 Capacity in Caracol 74 APPENDIX 10: Cost-Benefit Analyses 41 Prioritized Hazards 21 7.9 Capacity in Jacquezy 7 APPENDIX 11: IDB Water Study Simulation Model 7.10 The Neighborhood of the Caracol Industrial ° ater Study - Simulation Mode 4.2 Methodology 22 Park 79 Development Results 4.3 Climate Change Projections 22 . 7.11 The Three Bays Marine Park 83 4.4 Hazard Profiles 24 . . . 7.12 Risk Reduction Recommendations 86 4.5 Vulnerability Assessment 32 . . 8. Conclusions and Recommendations: À Smart 4.6 Loss Estimation 33 Growth Scenario 93 > EMERGING Lo «CES GIDB NORTHERN DEVELOPMENT CORRIDOR, HAITI i Initiative ERM [page 3] LIST OF FIGURES Figure 18 - PGA probabilistic seismic hazard map for Figure 35 - Map of the attractions sub-model: 10% probability in 50 years, i.e. 475-year return period composite of maximum attractiveness factors 49 Figure 1 - Area of study — Northern Development 24 Corridor 1 Figure 36 - Attraction factors: Development Projects 50 Figure 19 — Coastal flooding with climate change y pile Figure 2 — Area of Study 4 projections for a 50-year return period 26 Figure 37 Land Suitab ility Model Based on : Attractions and Restrictions 52 Figure 3 - Composite of some ofthe mapping Figure 20 - Key watersheds in the study area 27 . imized land developed by the AIA Study (Illustrative Only) 6 Figure 38 — Optimized land use map 54 Figure 21 - Inland flood hazard map with climate y 1 Iti-dwelli d raised Figure 4 - Extract of the CIAT Strategic Plan showing change 50 year return period 28 Figure 39 - Examples of multi-dwelling and raise four poles of economic attraction that would result | | dwellings presented at the Zorange Expo 56 from the implementation of two new urban centres Figure 22 na nle areas in Haiti as 30 Figure 40 - Detail of elements analyzed for each one of {Champin and Carrefour Chevry) 8 presented by ’ the townships in the study area. 58 Figure 5 — Human settlements in the NDC 9 Figure 23— Monthly variation in water demand'and Figure 41 - Main land uses identified in the township availability (current conditions) 31 d n Figure 7 - Typical township in the NDC area (Terrier of Trou-du-Nord. 59 Fi 24 - Monthl jation i ter d d'and Rouge) 10 Igure 29 AONENIy VOTIQUON I Water cemandan Figure 42 - Trou du Nord - Areas selected for availability (projection for 2040 including climate Iculatina the building densi Figure 6 - Typical hamlet in the NDC area. (Paulette) 10 change) 31 calculating the building density. 61 Figure 8 - Typical road-side settlements 11 Figure 25 - Distribution of block boundaries in the Figure 43 : Trou-du-Nord . Current land uses, areas for | d 32 densification within the urban setting and proposed Figure 9 - EKAM neighborhood development study area expansion areas. 63 developed by USAID 11 Fi 26 - Distributi dE Vali Igure : IE “ on and Exposure Values of Figure 44 - Main land uses identified in the township Figure 10 - Footprint growth of urban settlements in Residential Buildings in the study area 33 of Limonade 64 the NDC 12 Figure 28 - Risk Map: À Annualized L he ke s se dont ns ized Loss for Figure 45 - Limonade - Areas selected for calculating Figure 11 - Evolution of a housing development in the Earthquake Hazard, Residentia 34 the building density 65 PIC area 13 ï - ili 7 Figure 12 and 13 - Land Use for 1986 and 2010 Y densification within the urban setting and proposed respectively for the NDC 14 Figure 29 - Development projects 37 expansion areas. 67 Figure 3.2-2 Land Use in 1986 14 Figure 30 - Average annual growth of total population Figure 47 - Main land uses identified in the township Figure 14 - Urban footprint growth 1986-2010-2013 39 of Terrier Rouge 68 for key urban areas 15 Figure 31 - Urbanization rate in the municipalities of Figure 48 - Terrier Rouge - Areas selected for Figure 15 - Urban intensities in 2010 16 the study area 39 calculating the building density 69 Figure 16 - Main ecological system on the Northern Figure 32 - Topics and elements considered to be Figure 49 - Terrier Rouge - Current land uses, areas for Development Corridor 18 restrictions for development 44 densification within the urban setting and proposed ; ue expansion areas 71 Figure 17 - Cultural heritage 19 Figure 33 - Map of the restrictions sub-model: composite of maximum restrictions 46 Figure 50 - Main land uses identified in the township Bord de Mer de Li di 72 Figure 34 - Topics and elements considered to be of Bord de Mer de Limonade attractions for development 47 > EMERGING Lo NORTHERN DEVELOPMENT CORRIDOR, HAITI il initiative ERM [page 4] Figure 51 - A pilotis - supported house developed for the Zorange Housing Expo 73 Figure 52 - Main land uses identified in the township of Caracol 74 Figure 53 - Main land uses identified in the township of Jacquezy 77 Figure 54 - The ‘neighborhood” of the Caracol Industrial Park 80 Figure 55 - Areas that should be considered for future development 81 Figure 56 - Preferred locations for consolidating new urban settlements in the PIC area 82 Figure 57 - Creating a planned, integrated community with the PIC as pivot. 83 Figure 58 - Preemptive zoning classes proposed for the Three Bays Marine Park 85 Figure 59 - Framework for Relative Risk Evaluation 86 Figure 60 - Standardizing loss damage recurrence comparison for the study area 87 Figure 61 - Smart Development Scenario for Haïiti's Northern Development Corridor 94 == EMERGING Lo NORTHERN DEVELOPMENT CORRIDOR, HAITI ji initiative ERM [page 5] LIST OF TABLES Table 17 - Total area and current land use distribution Table 31 - Jacquezy - Urban land uses inside and in the township of Limonade 64 outside the high risk flood areas, and ‘true’ available Table 1 - Land use change observed between 1986 land. 78 and 2010 in the study area 14 Table 18 - Limonade - Urban land uses inside and : outside the high risk flood areas, and ‘true’ available Table 32 - Jacquezy - Distribution of urban land uses Table 2 — Haitian stakeholders 22 land. 65 under a ‘good practice’ scenario 78 Table 3 — Summary of Climate Change Projections for Table 19 - Limonade - Capacity for residential Table 33 - Total areas of expansion that would be the 20405 for Northern Haiti 23 developments inside the urban setting and areas required to accommodate the housing demand Table 4 - Coastal flooding projections (including for required for expansion in the 2040 fast growth expected by 2040 in the ‘fast’ population growth climate change to 2040) 26 scenario 66 scenario 79 Table 5 — Inland river flooding projections (including Table 20 - Total area and current land use distribution Table 34 - Comparison of Hazards for the study area87 for climate change to 2040) 28 in the township of Terrier Rouge 68 Table 35 - Potential Losses, 106USD 91 Table 6 Summary of water balance for the study Table 21° Terrier Rouge - Urban land uses inside and Table 36 - Summary of risk mitigation measures 92 area 31 outside the high risk flood areas, and ‘true’ available land. 69 Table 7 - Summary of Impacts and Loss Estimates by Hazard 36 Table 22 - Terrier Rouge - Capacity for residential developments inside the urban setting and areas Table 8 — Northern Region Population and Growth required for expansion in the 2040 fast growth Projections (source AIlA Study) 38 scenario 70 Table 9 - Place of residence of PIC workers 39 Table 23 - Total area and current land use distribution Table 10 - Projections of the population base - in the township of Bord de Mer de Limonade 72 scenarios of slow growth 40 Table 24 - Bord de Mer de Limonade - Urban land uses Table 11 - Projections of the population base - inside and outside the high risk flood areas, and ‘true’ scenarios of High growth 41 available land. 73 Table 12 - Summary of the main restriction factors 45 Table 25 - Bord de Mer de Limonade - Capacity for residential developments inside the urban setting 74 Table 13 - Summary of the main attraction factors 48 Table 26 - Bord de Mer de Limonade - Distribution of Table 14 - Total area and current land use distribution urban land uses under a ‘good practice’ scenario. 75 in the township of Trou-du-Nord 60 Table 27 - Total area and current land use distribution Table 15 - Trou-du-Nord - Urban land uses inside and in the township of Caracol 75 outside the high risk flood areas, and ‘true’ available land. 60 Table 28 - Caracol - Urban land uses inside and outside the high risk flood areas, and ‘true’ available land. 75 Table 16 - Trou du Nord - Capacity for residential . : . developments inside the urban setting and areas Table 29 - Caracol - Capacity for residential required for expansion in the 2040 fast growth developments inside the urban setting 76 scenario 61 Table 30 - Total area and current land use distribution in the township of Jacquezy 77 == EMERGING Lo NORTHERN DEVELOPMENT CORRIDOR, HAITI iv initiative ERM [page 6] ACRONYMS NATHAT National Hydrologic Assessment Tool AAL Average Anualized Loss NDC Northern Development Corridor AFD Agence Française de Développement NDVI Normalied Difference Vegetation X AIA American Institute of Architects _. : OAS Organization of American Sates CARE Humanitarian Organization . LL. OCHA UN Office for the Coordination of Hu- CDMP Caribbean Disaster Mitigation Project manitarian Affairs CELADE- Population Division of the Economic oSM Open Street Map ECLAC Commission for Latin America and the OXFAM H itarian O LL Caribbean umanitarian Organization CIAT Comité Interministériel d'Aménage- PDNA Post Disaster Needs Assessment ment du Territoire PGA Peak Ground Acceleration CNGIS Centre National de l'Information Géo- PIC Parc Industriel du Caracol graphique et Spatiale PML Probable Maximum Loss DTM Digital Terrain Model PRECIS Regional Climate Model ENSO EI Niño Southern Oscillation (Episode) RN# Route Nationale # ERM Environmental Resources Manage- SEMANAH Service Maritime et de Navigation ment Inc. d'Haïti ESCI Emerging Sustainable Cities Initiative SRES IPCC Special Report on Emission Sce- FAO Food and Agriculture Organization narios FED Fonds Européen de Développement UCE Unité de Coordination et d'Exécution FEWS Famine Early Warning System Network UHN-RHC National University of Haïti Roi Henri GoH Gouvernement du Haïti Christophe Campus in Limonade IADB Inter American Development Bank UN United Nations IHSI institut Haïtien de Statistique et USAID United States Agency for International : : Development d'Informatique IPCC Inter Governmental Panel on Climate USAID- USAID Office of US Foreign Disaster OFDA Assistance Change LAC Latin America and the Caribbean UTE Unité Technique d'Exécution LANDSAT Satellite Imagery Acquisition Program WB The World/Bank LEC Loss Exeedance Curve MEF Ministère de l'Economie et des Fi- nances MINUSTAH Forces de l'ONU en Haïti == EMERGING Lo «CHE GIDB NORTHERN DEVELOPMENT CORRIDOR, HAITI v initiative ERM [page 7] e Ana Maria Säiz - Fiscal and Municipal Manage- e Jean Frantzdy, Assistant Technique du Cap- ACKNOWLEDGEMENTS ment Specialist Haïitien This report has been prepared by Environmental e Thierry Delaunay - Water Specialist ° Ravaz Josselin, Assistant Technique du Cap- Resources Management (ERM) on behalf of the + Sarah Romain — Water Specialist Haitien Inter-American Development Bank’s (the Bank) e Carlos Faleiro — Consultant, Water and Sanita- Emerging and Sustainable Cities Initiative (ESCI). tion Other Organizations and Representatives: The contributions and support of the following or- + Agustin Filippo - Operations Senior Specialist ° _ Agro en Action - Frantz Varella, Director ganizations and representatives are acknowledged: + Jose Luis lrigoyen — Operations Specialist ° American Institute of Architects - Erica Rioux e Peter Sollis - Social and Economic Specialist Gees, Director Emerging Sustainable Cities Initiative | * American Red Cross - Anna Konotchick ° Ellis J. Juan - General Coordinator Other Bank representative: + CNIGS - Boby Piard + Horacio Terraza - Sector Coordinator, Infra- e Guirlaine Denis, Ermithes Lauture, Cedrick Jo- + COSMHANNE - Communauté OpenStreetMap structure and Environment Sector seph, Stephanie Brackmann, Melissa Barandar- Haïti Nord et Nord-Est + Patricio Zambrano-Barragän — Urban Specialist, ian, Crystal Fenwick, Andy Drumm, Marie Bon- e DINEPA - Lesly Dumont Infrastructure and Environment Sector nard, Bruno Jacquet, Michel Vallée, Frederica e FAES- Julio Martinez and Patrick Anglade ° David Maleki - Climate Change Analyst Braun, Priscilla Rouyer °__Fmg Municipal Nord-Est - Marjorie Victor Dan- + Maricarmen Esquivel — Specialist, Climate iel Change Division Comité interministériel d'Aménagement du Terri- ° Jude Marie St. Martin, LOKAL+ ° Fernando Miralles-Wilhelm, Specialist, Water toire ° MARDNR/MICTD - Helliot Amilcar and Sanitation Division ° Michèle Oriol - Secrétaire Exécutif + MPCE/DATDLR - Alex Julien e Carlos Mojica, Specialist, Transportation Divi- ° Rose-May Guignard - Urbaniste Senior + MTPTC- Yolene Surena sion + Other CIAT representatives: Christelle Baptiste, + OXFAM - Agathe Nougaret and Laurence Des- e _ Ginés Suérez - Consultant, Environment, Rural Eleonore Labattut, and Erdem Ergin vignes Development and Natural Disasters Division + SONAPI - Georgemay Figaro + Raül Muñoz - Consultant, Water and Sanitation Unité Technique D’Exécution - Ministère de + UNFPA - Gabriel Bidegain Division l'Économie et des Finances e l’Université Roi Henri Christophe - Jean Marie . menin Kerres — Consultant, Climate Change ° Michael DeLandsheer - Executive Director Theoder President of the Board of Manage- e Gisela Campillo - Consultant, Climate Change ° Reynold Pauvo : Technical Director e USAID - Christopher Frey and Chris Ward Division e Alix Clement - Division Chief . . Local City Officials ERM Partners Inter-American Development Bank Country Office ‘ : In the delivery of this work, ERM also wishes to + Agustin Aguerre - Country Representative : Ge Use Maire du Linenade thank its key partners: e Gilles Damais - Chief of Operations : ’ nm : e Arcindo Santos - Fiscal and Municipal Man- ° Vercius St-Preux Mairie du Terrier Rouge ° CEEPCO Engineering Haïti agement Specialist . Samuel Romain lunior, Mairie du Trou-du-Nord ° New Haiti Institute e Pierre-Louis Annot, Directeur Planification du Cap-Haitien > EMERGING Lo NORTHERN DEVELOPMENT CORRIDOR, HAITI vi initiative ERM [page 8] 1. INTRODUCTION Initiative (ESCI). The purpose of this Initiative is to designed to facilitate sustainable city planning. It contribute to the improvement of the quality of life leverages its capacities as the leading source of 1.1 Background and ESCI in LAC's cities in terms of environmental, urban, and development financing for the region and applies its fiscal sustainability. To accomplish this, the ESCI long experience in supporting the countries of LAC. Cities and urban areas play a key role in the provides decision makers with tools, data and initial economy of Latin America and the Caribbean (LAC) frameworks for managing urban growth and 1.2 ESClin Haiti through generating opportunities, such as diffusion itori: i of expertize and innovation, concentration of spe- ferritorial expansion Haiti's Northern Development Corridor (NDC), cialized labor, and provision of educational, cultural, Through the ESCI, the Bank combines the expertise shown in Figure 1, presents a special case that re- and recreational services. With these opportunities of its different sector departments in the quires flexibility in the implementation of the ESCI come challenges such as poverty created by in- formulation of comprehensive action plans methodology. In mid-2013, ESCI launched the im- migration and an increasing and often unsatisfied demand for urban and social services, decent housing conditions, and opportunities to generate Atlantie Ocean es, Done Dételopment income. Overcoming these challenges require a : ; Corridor - Area of Study comprehensive approach that promotes both # 76 A sustainable growth and the improvement ofs N enchitle : X : citizens’ quality of life. si "LA 2e ° # F1 OX È Formal and informal growth often leads to negative ee 3 = PA) N environmental, social, and economic impacts. à. 17%, "Mens Municipal policy makers usually lack adequate data £" à HIER and analysis to inform the design of policies that À \ >. SN 4 help promote growth in a sustainable way. In many LEA vi à a cases, the implications for the municipal budget in Golfe'de laG 2 terms of financing infrastructure development and have , + « operation costs have not been clarified in newly > 73} — urbanized areas. Additionally, the environmental * «à impacts of city growth are often not typically fully PES F. considered. Areas for conservation and aquifer <t recharge need to be established or protected, and EL 2 ) vulnerability to natural disaster and the effects of ’ V4 ré x Gr. » climate change reduced. Anticipatory planning can Te 7 AÉÈTE : rat Les also help reduce greenhouse gas emissions (GHG) ven \ 30 orne. 12, SA f as a major factor affecting climate change. ; } = Le dE Te, £ o . . LL Ccatibbe, , j In response to this situation and in light of RE Le de. continuing urbanization process in the LAC region, L the Inter-American Development Bank (the Bank) Figure 1 - Area of study — Northern Development Corridor launched its Emerging and Sustainable Cities > EMERGING = Lo (Je SUSTAINABLE KZ NORTHERN DEVELOPMENT CORRIDOR, HAITI 1 LX EU BIDB ERM [page 9] plementation of an adapted version of its method- nance, ESCI is working to help mitigate urban de- seismicity; hurricanes; and drought — and using ology for Haiti's NDC, currently home to approxi- velopment impacts and catalyze interdisciplinary newly developed digital terrain models, in- mately 500,000 people in the country’s Nord and planning processes in Northern Haïti. ESC/’s ulti- cludes a probabilistic modeling of their impact Nord-Est departments. The NDC includes the com- mate goal for the NDC is: to deliver site- and city- on the region’s natural and urban landscapes munes closest to the Caracol Industrial Park (PIC), a specific plans for urban and infrastructure devel- and an estimation of impacts on existing infra- flagship economic development project that may opment in the municipalities closest to the PIC, structure. bring up to 25,000 new jobs to the region in the namely Limonade, Trou-du-Nord, Terrier Rouge next few years, unlocking rapid demographic and and Caracol. To achieve this goal, ESCl's tailored 2. Urban Growth Study. This study presents multi- urban growth and putting pressure on the region's approach in Haiti involves the implementation of horizon projections of urban and demographic services and resources. four baseline studies: growth with two basic scenarios (rapid versus slow) and their respective spatial distribution In partnership with strategic actors in Haïti, such as 1. Vulnerability and Risk Assessment of Natural and impact on existing ecological and urban as- the Interministerial Committee for Territorial Plan- Hazards. The assessment focuses on four risk sets. The growth models include the potential ning (CIAT) and the Ministry of Economy and Fi- categories — flooding (inland and coastal); spatial and growth impacts of new develop- ments (e.g., port upgrades in Cap Haïtien) on the four communes’ area of influence. Note that these two studies form the basis for this report entitled Urban Development and Climate Change Study (referred to hereinafter as the ESCI Growth Study). 3. Sustainable Mobility Plan. The plan will engage in unprecedented data collection exercises in Northern Haiti, including an origin and destina- tion survey and counts. Based on this data, the « Plan will include demand projections and draw L ë À| = F recommendations for priority mobility pro- ss CR. : : Ë > jects, such as transport hub infrastructure, : = Re ie LE esta s ze en fous multimodal options, and improved services for En eZ se RE Re | PIC workers. The geographic focus is threefold: STE. sa ee = É - es d the PIC, the surrounding communes, and Route hd PR Pr " National 6. ——— 4. Living Conditions Survey. There are considera- = —— nr E———e ee ble in social and ic informati = — = gaps in social and economic information, - = especially with regard to wages and labor, Er health and education levels, access to services, = DÉRTREE S Sa disaster preparedness, etc. In order to develop nn so ee cn an Re ee em ie me planning strategies and instruments based on NORTHERN DEVELOPMENT CORRIDOR, HAITI 2 À EU IDB ERM [page 10] up-to-date and reliable information, ESCI will 1.3 Urban Development and Climate information contained in the referenced appen- implement a complete household survey in the Change Study dices and supporting deliverables such as a GIS da- urban and rural areas of Limonade, Terrier tabase. Rouge, Trou-du-Nord, and Caracol, and gather Environmental Resources Management, Inc. (ERM) basic social and demographic information as was engaged by ESCI to undertake the first two This report presents the results of the two studies well as select information on public opinion. studies for the NDC: the Vulnerability and Risk As- (the ESCI Growth Study), as well as recommenda- The ultimate goal is to develop a baseline of in- sessment of Natural Hazards (Risk Study) and Urban tions to address the complex question of where and formation about the households adjacent to Growth Study (Urban Study), collectively presented how urban development should occur in the NDC the PIC. in this report as the Urban Development and Cli- given various dynamics that shape population mate Change Study (referred to hereinafter as the growth in the area. This report is intended to pro- Building on the results, community feedback, and ESCI Growth Study).These two studies build upon vide planning tools and insights, and a building recommendations from each of these studies, ESCI the work and methodologies ERM has used and block upon which more prescriptive plans and plan- will provide and promote four site- and city-specific developed in conjunction with the ESCI team for ning policies can be developed. It will also help urban development plans for Limonade, Terrier similar studies in Cochabamba, Bolivia and Mana- guide decisions about accommodating and influenc- Rouge, Trou-du-Nord and Caracol. The plans will gua, Nicaragua. This report presents the consolidat- ing future growth. also build on previous planning exercises by local ed findings of the study, with further details and partners such as the CIAT, which have outlined a regional vision for the NDC but require local speci- ficity and consideration of future development al- ternatives. This dual approach — to develop a foun- dation for planning based on detailed studies, as well as to build on relevant, past efforts — will en- sure that the four urban plans help guide Haïitian stakeholders and their domestic and international = En — a partners in key urban development areas for the es, DEN ERP ES NDC. ESC/'s work will include proposed interven- L tions at a pre-investment level, so as to facilitate = EE” k swift action according to local priorities. 2 GC. SDS - En These proposals touch on areas such as mixed-use s - — nn TT. development strategies, design-driven conservation - æœ-_ =—— = of landscape and resources, more resilient siting = proposals for topics such as housing, and recom- mendations for transportation infrastructure at both the commune and regional levels, including mobility options to connect the PIC and its sur- rounding communes with new regional hubs such . : M Caracol Industrial Park as an upgraded port in Cap-Haïtien. = EMERGING Lo 6 SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 3 a gs Le IDB ERM [page 11] 2. METHODOLOGY AND APPROACH er feedback as part of an engagement work- proximity to the cities of Cap Haïtien to the west, shop held for the project. and Fort Liberté to the east, partially covering the 2.1 Methodology for the Risk and Urban : : North and Northeast Departments. Geographically, Studies v. Section 7: Development of a sustainable the area of study is located at the Plains du Nord, growth scenario for the study area, consider- the coastal plateau where the Massif du Nord The two studies presented in this report (Risk and ing both slow and fast growth, presented in mountainous chain meets the Atlantic Ocean. Urban Studies) both have clear and defined objec- Section 7. tives as well as significant overlaps and connections The study area comprises the key towns of Limo- to each other. The following methodology was vi. Section 8: Presentation of key conclusions nade, Trou-du-Nord, Caracol and Terrier Rouge. used to capture the key elements of the individual and recommendations. The study area boundaries have been defined studies and their respective connections: based upon the following: 2.2 Study Area i. Section 2: Development and understanding of LC: ° The main core of the study area is to be the current baseline conditions in the study area The study area shown in Figure 2 covers an area of NDC and the PIC along Route National 6 (RN6) using: 49,391 hectares along the Atlantic Ocean, in close e readily available information from key stakeholders regarding physical, biologi- cal, hydrological and urban systems; ODE e defined studies and assessments com- = prising current and historical land cover N assessments; Section 3: Results of the baseline assess- / ments. 2 ii. Section 4: The baseline is further supplement- ÿ ed by hazard and risk assessment studies. ë m ii. Section 5: Identifying and defining future - $ growth and development considerations in- ë cluding future population growth projections | and future hazard and risk considerations, while accounting for climate change projec- tions and future development projects. This is presented in Section 5. iv. Section 6: Geospatial land suitability model- ling using the baseline conditions and future projections to identify land potentially suita- Figure 2 — Area of Study ble for development. Also includes stakehold- + <s EMERGING" (NERO Lo # SUSTAINABLE KZ NORTHERN DEVELOPMENT CORRIDOR, HAITI 4 EX gs BIDB ERM [page 12] given the socio-economic development and ° The Plan d'Aménagement du Nord / Nord- community levels, with clear indications re- growth being promoted as a result of the con- Est: Couloir Cap - Ouanaminthe, CIAT (the garding the roles and responsibilities of each; struction of this corridor; CIAT Strategic Plan). This strategic plan pro- (ii) the mechanisms for implementation, in- ° The southern and eastern boundaries are de- duced by the CIAT is aimed at developing the cluding administrative measures such as zon- fined by the water catchment zoning; Haitian territory, and in this case the north, ing, and (iv) what the plan calls ‘supplemental’ e The west is defined by both the water catch- With greater regard for its natural resources, measures, such as community compacts, an- ment area of the Grande Rivière du Nord and the risks and vulnerabilities it faces, and the chor investments, and others. the municipality of Limonade; and opportunities it offers. It comprises two doc- ° The proposed Three Bays Marine Park (Parc uments, the first of which, Haïti Demain (Haïti + Based on the diagnoses and analyses carried Marin des Trois Baies) will be taken as part of Tomorrow), offers a country-wide framework; out at the regional and municipal levels, the the study area and the Three Bays Marine the second, La Boucle Artibonite (The Arti- plan reiterates or defines new, significant de- Park will define the northern boundary, albeit bonite Loop), defines the space and function- velopment projects at those scales, which the study has been limited to the coastline. ality that should be considered for develop- should catalyze development around them- ment in this area that could turn it into a ma- selves and regionally as a result of their ag- In delineating the study area, it is acknowledged jor economic pole. gregation. that the urban areas of Cap Haïtien, Fort Liberté and Ouanaminthe are important influencers There are a number of areas in which the ESCI + In terms of spatial form and land transfor- throughout the NDC, and these factors have been Growth Study contributes to the AIA Study and the mation, the plan suggests an area of urban considered during the study. CIAT Strategic Plan, and further commentary on this development the south side of RN6. The plan is provided below. seems to indicate that this area would in reali- 2.3 Building upon Key Planning Efforts ty be formed of two separate areas: one to The NDC has been the subject of a number of im ee AA Stuay the south east of Limonade, with limited or portant recent planning efforts. This ESCI Growth The AIA Study is divided into three volumes: i) re- Deer ou du Nord the Pre ares on Study does not seek to replace or supersede these gional comprehensive plan; ii) urban growth plans Terrier Rouge other studies, but rather to complement and build for the different municipalities; and ïïi) detailed ° upon them. The two key planning studies of interest analyses by sectors (or ‘focus areas” as it refers to ° The plan then shifts to the local level, by are: them), which add up toa cumulative impact as- means of offering local development plans for sessment report that is the basis of both the re- : Le : : ù u the eight municipalities that comprise their ° The Cap Haïtien-Ouanaminthe Development gional comprehensive plan, and the urban growth area of study (the whole North and Northeast Corridor Regional Comprehensive Plan, AIA area plans. The following are some key aspects of Region). These local plans focus on their ur- Legacy (referred to hereinafter as the AIA this work: ban and immediate areas of influence, provid- Study). Published in December 2012, follow- ing clear guidance on green and grey infra- ing a year of work carried out by the American * ltpresentsa framework for how future devel- structure, social services, road and transport, Institute of Architects nonprofit foundation, opment of the region should occur, including housing and other land uses. The plans in- AIA Legacy, with funds from the IDB and the ) the principles that should guide develop clude key data with regards to urban growth, United States Agency for International Devel- ment in terms of natural resources, economic based on population projections and housing opment. The study was carried out in close in- growth, infrastructure support and human and land demand teraction with the CIAT and the Technical Exe- development; (ii) the levels of plan implemen- ‘ cution Unit of Economy and Finance. tation, that is, the regional, municipal and > EMERGING Lo sl * SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 5 EX EU % IDB ERM [page 13] 5 — a ———7 ——— development processes of each municipality. The R= La ; sé = ë - E AIA Study represents the process of settlement F h ré À ea, a mp 5... along RN6 as occurring mostly along the south side dy JUN Ù \ ; us f the corridor and the local plans are focused on ) QAY LE TS (4 ” en En o p Y 4-5) CF ÿ 6 ? A Ç the urban areas only. This does not acknowledge A D v AT n 3 ( development north of RN6 or the rural and inter- PNR AL M un 4 ? à mediate areas where settlement may and is likely Kk X =, An KA 6 s LE to happen. This ESCI study will be contributing in RE = = [US == 2 | this regard by including information from recent a surveys and aerial imagery from the area, analyzing fé what would the actual and projected growth pat- VE 2 æ: } re se (Ca terns at the regional level be, and comparing them 2. TS EE Ya, with the results of the AIA regional comprehensive Ÿ » + Ÿ À plan. This will include not only the main townships 2 QE œ} 2 but also smaller villages or hamlets. x L: A en Ne ten : D (dirt AS \ Secondly, the ESCI Growth Study will enhance the — = — = “& Re SR E mu * © AIA analysis of potential settlement areas by: bi . @, terre | RP Sr à o A * | _Æ 4 >? Re é De, % F4 QE ES N K:) + _Introducing a larger number of variables DR /E > 23 tra = SUR “0 4 LC R ; to be considered and analyzed as part of AA # 1 ts ERA À à * Et :) is CE the urban planning and land use assess- À ( L es % + ÿ Æ 5 ” RS n x ° ___Introduce greater resolution to the geog- = À 1) _ _ . raphy of some these variables. EL a æ D. É D + Application of a geo-spatial model to iden- RE: C9 Matos @D cnmonet ŒD Arrmrak ŒD coûte = Oertegmmt Comte |h] © = += = S—- _ He © romridranet CE mous D cennintepair nn tify the areas suitable for human settle- Figure 3 - Composite of some of the mapping developed by the AIA Study (Illustrative Only) ment. : . . . This enhancement will enable recommendations to e The plan includes a proposal for a new devel- of aggregating them. There are four specific contri- : IT: : : ne be provided that will either confirm, or suggest re- opment for 6,000 people, Caracol Nouvelle, butions that the ESCI Growth Study work will bring LU : ù . : : orienting, of the results of the AIA regional compre- that is shown as a model of a planned, mixed in relation to the AIA Study. . : | hensive plan in terms of where should development use and environmentally sound settlement. occur The plan suggests that this could serve as First, because the regional comprehensive plan is a ur seed for additional settlement. high-level, multi-sectorial, multi-institutional and Third, the AIA Study implies that the core ofthe multi-leveled instrument, it is not intended to pro- os : Le nu: . : : : | : k . . . urbanization of the entire region is to occur within Figure 3 provides a composite that illustrates some vide detail on the spatial configuration of the region the NDC area, with measures to prevent spatial of the elements and plans considered by the AIA as a whole, other than that which would result from rowth of Quartier Morin and Lnonnde he Study/s regional comprehensive plan and the results the process of settlement along RN6, and the local a and the implication that the stronger eco- = EMERGING Lo CHE GIDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 6 À initiative ERM [page 14] nomic links between Ouanamithe and Fort Liberté The Plan d'Aménagement contains five compo- it as such, like the new planned development, in the east will reduce the pressure for urbanization nents: i) an urban program; ii) an economic pro- but also the proposed ‘urban’ corridor be- beyond Terrier Rouge. It can however be expected gram around major projects; iii) a management tween Sainte Suzanne, Trou-du-Nord, Cham- that RN6 will continue to attract settlement along strategy; iv) a transport plan and forms of habitat; pin and Caracol. This is a fundamental ele- its edges, which could turn the area into a long, and v) access to public services. The third part of ment not visible in the AIA Study, for it repre- semi-urbanized strip, and better understanding the the Strategic Plan focuses on implementation, and sents the only direct link between the moun- attraction factors for development, which this ESCI provides a discussion on the governance and opera- tain areas of the Massif du Nord, the plateau Growth Study seeks to do, will enable a better un- tional structure that ought to be set; a series of and the coast, between Cap Haïtien and derstanding of existing, trending and more ‘intelli- orientations or guidelines for elaborating urban Ouanaminthe. gent’ urbanization patterns that the region should development plans at the local level; and an in- exhibit as a whole. vestment plan. There are four specific contributions that ESCI Growth Study brings to this CIAT plan. First, an up- Lastly, each of the local development plans pro- Important considerations from the CIAT Strategic dated demographic analysis has been performed duced by the AIA Study offers a series of elements Plan for the ESCI Growth Study include: that yields more detailed and disaggregated infor- associated to the growth process. Of particular sig- mation useful to offer a better approximation of nificance are the delineation of the existing urban + The plan reads as being supported and sup- where would human settlement likely grow, and boundary, an approximation to the extent of each portive on the results and the process of the the extent of this. Secondly more refined approxi- municipality’s population growth, and the areas AIA Study, and reiterates numerous elements mations will be offered to the geography of the where this should be accommodated. With the re- of the diagnosis. elements that are fundamental to the economy: sults from the ESCI Growth Study modeling, an as- e __Itoffers a series of guidelines on the areas agricultural milieus, economic development pro- sessment of the location of these areas will be of- that ought to be developed at the local level jects, protected areas, etc. fered. and the criteria or principles that should be followed in each municipality. Because of its Third, the CIAT Strategic Plan’s vision will be em- 2.3.2 CIAT Strategic Plan scope, it does not provide detailed approxi- braced, such as a network of services that comple- The CIAT Strategic Plan comprises three elements, mations to the geography orthe land use pat- ment each other, as opposed to compete against under a 2012 — 2030 time horizon: i) a regional di- terns to which many of its propositions should each other; and acknowledging the geographical Lo translate. areas that should be protected for their potential agnostic; ii) a Plan d’Amenagement (layout plan), : and ii) a series of implementation measures. ° The plan does not appear to endorse the con- economic value. solidation of an urban continuum along RN6; . . . The regional diagnostic focuses on seven primary rather it highlights the need to urbanize the Finallv, alternative locations fornew planned set- challenges: triangle within the Champin area as the fun- tlements will be identified should the suitability damental intervention (in the NDC) that will analysis yield areas more attractive as a result of 1. Accompanying the growth in population ensure the balanced distribution of popula- the modeling. 2. Town structure tion in the region. 3. Transforming economic structure e The planis very clear in the pursuit to consol- 4. Modernizing agriculture idate four polarities in the region, one of 5. Enhancing heritage which would be the area covered by the pre- 6. Reducing vulnerabilities sent study (the Champin pole — see Figure 4). 7. Ensuring good collective management It also offers key elements that would identify > EMERGING Lo sé * SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 7 EX EU % IDB ERM [page 15] Monte Cristi 30 km Le do République Dominicaine @ Pôle de Cap-Haïtien Dajabon @ Pôle de Trou-du-Nord è Santiago @ Pôle de Fort-Liberté pris @ Pôle de Ouanaminthe Caballeros 134 km Figure 4 - Extract of the CIAT Strategic Plan showing four poles of economic attraction that would result from the implementation of two new urban centres (Champin and Carrefour Chevry) Le EMERGING Lo «4 GIDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 8 lnitiative ERM [page 16] 3. BASELINE CONDITIONS 3.1 Current Study Area F #7. Fe, &” mn The proposed study area presented in Figure 2 is # ‘ À Ne defined as follows: lt. NET Pier URL Lan, ; ° The main core of the study area comprises la CRT Eaux : \ Î ÿ the west-east corridor along RN6:; N +.) Emme 2" “| Pro | É H } 7. } + Î en Cortrée ° The Trou-du-Nord catchment boundary de- rl DEP P'é EE cp L 1 mi] . e DR : fines the southern extent of the study area, js 7 ag Ma LE pu LAC me à ê and the coastline and associated bays along : FPE I #. Tee p : a À 4 : à { —" Canbégne E” Gun A ë the northern coast naturally define the north- , se à l J Po ere + ern extent of the study area; +. 5 d k / Ps Fe à . . 1 LP Ce £ \ pars Dares # e The easterly boundary is partly defined by the PA 2x PE + patin 7 « Nr Trou-du-Nord watershed and in the north- LŸ En sp: ÿ 4 LOT Sn PT Lo pou > east by the Terrier Rouge and Fort Liberté 1 L Shen x * na + municipal boundary; and x s £ + ° _Tothe west, the study is defined by the Limo- € CA ne K : : De C'ohal T ] ‘ nr nade and Quartier Morin municipal boundary. x \ { o 25 5 LR À In defining the study area, the urban areas of Cap Haïtien, Forte Liberté and Ouanaminthe are recog- LI feu area : : : imite commune nized as important influencers and have been fully nn cie considered. Em Urbain 2013 Reocif corallien The overall NDC area can be characterized as rural Im Océan Atlantique because of the traditional land use patterns that it Figure 5 — Human settlements in the NDC exhibits, including large plantations of sisal and : . plantain, and traditional small to medium scale Map source: From geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), lADB (c.2013), NATHAT (2010), . £ inef ducti £ frui | in li k USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing arming for production of fruits, plantain, livestock, analysis from satellite imagery 1986, 2010 and 2013. cassava, vegetables and others. The agglomeration patterns that have emerged are townships, ham- 3.11 Townships ‘organic’ grid of roads that configure blocks of ir- lets, farms, ‘linear’ settlements along roads, and In the NDC. there are three main townships: Limo regular shape and dimensions. These tend to be planned settlements, which are presented in Figure d . ps: between a quarter and a third of a hectare in area. nade, Trou-du-Nord and Terrier Rouge, represent- : : : 5. . . . Inside these blocks there is a parcel structure that in ing a first tier of settlements. They are founded at : : . Le . k l general terms is half built, with the remaining area regional crossroads, and their urban structure is : : . . being used as solarium and orchard. Consequently, largely formed by a town center with the main ad- : : : ne . Le LL the idea of a town in the NDCis largely that of an ministrative and religious buildings, followed by an = EMERGING = Lo 8 GIDB L NORTHERN DEVELOPMENT CORRIDOR, HAITI 9 Initiative ERM [page 17] agglomeration of single-family dwellings with a trative posts and other service-related buildings are of the surplus; to large estates exploited for sisal, yard. located. This leads into a network of a few streets plantain, citrus, cane and other crops. The pres- or paths that reach the surrounding areas, some of ence of small farmers along the internal roads in The township is surrounded by countryside of me- which are used for agriculture. An example is given the NDC is partly a result of the fact that they used dium sized farms, whose produce is generally sold in Figure 6, which shows the hamilet of Paulette. (or continue) to be employed in the large planta- or ex-changed at the township’s market. In terms of tions, whose owners have given them small parcels land use, the townships are almost entirely config- International relief organizations have developed to settle. Other, medium parcels and farms are the ured as mixed use, with buildings serving as home housing programs in Paulette and other hamlets. In result of subdivision of the large plantations over as well as office, store, training center, phone these cases, a difference is clearly visible between the years. booth, restaurant, etc. An example is shown in Fig- the ‘organic’ grid of the ‘natural’ town and the or- ure 7, which shows Terrier Rouge. thogonal grid of the planned settlement. As a consequence of population growth and subdivision of the original farms along the road, Hamlets are the second tier settlements present in In the NDC there are numerous types of farming linear areas of agglomeration begin to form, as the NDC, and include places such as Caracol, Bor de operations, from small farms alongside roads that shown in Figure 8. In some cases because of their Mer de Limonade, Phaeton, Paulette and Jacquezy. house two to three generations of a family, and population size, they become recognized as ‘urban’ These are normally located along secondary or ter- used for family subsistence and market exchanges tiary roads, and exhibit a much simpler structure. In hamlets the main road widens and becomes the main center, in which churches, playfields, adminis- / f” 129) 2 PP ARENE + L A LE Tree) Che De Le - D. 2 Lola A7 Le Dal F \ à - l : ÉRRISE A À nr US SA 2. =: (EE NE À D E _— NORTHERN DEVELOPMENT CORRIDOR, HAITI 10 t< as ERM [page 18] by the IHSI. ‘ ni jrs Donne 4 S ÉN, SS CR # AE a ES ES The NDC has also been the subject of numerous 6 4 # ù ‘à Se # efforts by international relief and cooperation or- 4 Fr: CR il ganizations to provide Haitians with shelter and + <> A sanitation, including Food For the Poor, the Red NY 2 à 4% 2% Cross, CARE, OXFAM, CHF and others. USAID has 7 4 4 D de NS 7 been intensely involved, having pledged after the » à » eg. 4 2010 disaster to build 3,000 homes as part of the L . “à où coordinated efforts that brought about the PIC, the a 3 ” Me completion of RN6, the upgrading or reconstruction 1 Fr 24 à . of several water and sanitation facilities and other L 4 +4 ’ projects of similar nature. | k 4 “ai Er % - es + | à The EKAM project is the largest residential devel- & $ - # à opment in the study area and was developed by 5 t’ D , USAID (see Figure 9). The project is located equidis- 4 # tant to the PIC, the University of Limonade and the township of Trou-du-Nord. It includes 750 homes, a community center, shopping areas and recreational spaces, to be placed on a 47 hectare site approxi- 1] mately. 4 î ; Another project for approximately 135 homes is to LAS be built on an 8.5 hectare site that is located just # + ( east of the PIC and the crossing known as Jesüs. & ba TIME CE re This project is immediately flanked to the west by a + , Ë customs facility, and to the east by a farm of ap- ru A : proximately twice the size that is being exploited Le LÉ beri Done with sisal that is then sent to a factory located in- > Æ | are ble L side the CIP. This sisal exploitation is expected to 4 , | Ê “2 SA er 5 grow to a 5,000 hectare operation that is described Ê— PERRET Sr later in this document. “#% GIDB © NORTHERN DEVELOPMENT CORRIDOR, HAITI 11 initiative 7 ERM [page 19] 3.1.6 Growth of settlements NDC (see Section 3.2 for more details on the meth- Trou-du-Nord, Terrier Rouge, Caracol, Paulette and odolo it has been possible to assess the changes Phaeton. At the crossing of the road that connects Using remote sensing analysis of two LANDSAT sat- Ev) i POsS! oécs 8 . 8 : 7. and growth of the urban footprints in the study RN6 with Bord de Mer de Limonade was also a ellite images (1986 and 2010), and through the use : Loc: : : : k . area. Asillustrated in Figure 10, in the year 1985 small settlement. Twenty five years later, in 2010, of a 2013 high resolution image collected and used : : : : Le : the main settlements in the NDC were Limonade, the coastal towns of Bord de Mer de Limonade, by the IDB to create a digital terrain model for the 07... : Na DEN 5 he (| TmoMsde Lé4 IN Î , à, } de € ET Phaeto | d j # À HutUN À Terrier Rouge ff = 4 f” ” ë # Ç v « { x 4 à \ { o É 25 5 1. À C2 Study area !__: Limite commune Section communale M Huelia urbana 1986 M Huella urbana 2010 nu" Huella urbana 2013 Recif corallien I Océan Atlantique Figure 10 - Footprint growth of urban settlements in the NDC Map source: From geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. = EMERGING = Lo e## GIDB |: NORTHERN DEVELOPMENT CORRIDOR, HAITI 12 initiative. ERM [page 20] Caracol and Jacquezy had seen the largest growth. : Bord de Mer de Limonade passed from an almost 2007 / invisible concentrated area at the regional scale, to 1 an area with the largest footprint of all three. Cara- ] col saw its area almost tripled, and Jacquezy had (| also grown dramatically. 1 The inland townships, however, saw a more normal L D URL LA expansion, in which Terrier Rouge exhibited the TS " largest expansion with 2.5 times the area of 1985, 7 followed by Trou-du-Nord with an expansion of u double the size of 1985, and Limonade, with a di ÿ 1 2013 slightly lower expansion. By 2013, what clearly ap- : peared in the map as agglomerations were the se- A ‘3 ki: : e . ries of linear settlements along RN6 and some of = +3 < the secondary roads connecting the different town- % EN ships. The settlements along the roads connecting ; es % | RN6 to Bord de Mer de Limonade as well as the ; +6 À road connecting Limonade with the township of 4 \ té Campegne to its South appear to be the more de- } . fined ones. Figure 11 - Evolution of a housing development in the PIC area Presently, this linear pattern seems to be the one 2009, jumping to 27 homes in 2010 and 37 in 2013. known as ground-truthing. Data collected in the acquiring more speed. Figure 11 exhibits a series of The settlement is likely to continue growing, and if field was used to calibrate the training regions in images from the area along RN6, in the vicinities of a measure of the intensity with which it has grown the most recent supervised classification and in- the intersection of this thoroughfare with the road from its inception was applied, the result in 20 form the execution of all other supervised classifica- that connects to Caracol and serves as one of two years would be a settlement of anywhere between tions on the historical imagery. The results provide access points to the PIC. The series begins in 2007 70 housing units if it grew conservatively to 160 an understanding of the evolution of land cover in and ends in 2013, passing by images from 2009 and dwellings if it grew with intensity. the northern development corridor from 1986- 2010. The area bound by a continuous yellow line is 2010. where a piecemeal process of settlement has oc- 3.2 Current and Historical Land Cover curred, likely with the support of an international As Figure 12 demonstrates, the urban areas and organization. The settlement is currently comprised A remote sensing analysis of two LANDSAT satellite corridors have grown noticeably. In 1986, the urban of a total 37 homes and what appears to be a small images (1986 and 2010) covering a span of 24 years footprint of the NDC only occupied 0.4 percent of communal or commercial facility. The area was was performed to analyze eleven land cover classes the total area of study expanding to 7.7 percent of rural up until 2007 and during 2008-2009 settle- for urban, rural and natural areas over time. the total area in 2013. In addition, the non-urban ment appeared, which is also when the construc- | de . lands appear to illustrate a significant deterioration, tion of RN6 took place. According to the images, the Final land cover classification Was checked for quali- with a reduction in natural and agricultural lands. settlement began with approximately 13 homes in y assurance and quality control issues (QA/QC) through a site survey of the area of study, a process > EMERGING = Lo NORTHERN DEVELOPMENT CORRIDOR, HAITI 13 initiative ERM [page 21] In terms of green infrastructure, the NDC experi- enced its significant deforestation before the period : es, D 1986 of analysis and a slow, but sustained deterioration 7, dm NS y A ù of its hydric system. The area has lost 5.030 hec- nadg Se tares of cultivated areas despite having vast exten- REA Carscbl ke Fi : sions agrological classified soils and in 2010 only PA - E half of the area suited for agriculture (6,315 hec- È NS 0 5 ; Î tares) was utilized. DEP LS EP K 4 Ë As shown in Table 1, the most important land use - É à F + ; ë changes occurred in rangeland increasing by 6,647 ? Gé © hectares, cultivated areas decreasing by 5,030 hec- À 2 F tares and barren land decreasing by 2,158 hectares. E ”s Besides rangeland, urban areas have increased sig- À 5 sc nificantly: medium intensity 55 hectares, low inten- 4 sy L 2 . —— sity 363 hectares and open space 530 hectares as- sociated to road development. In contrast, the hy- E " 2010 dric system lost 49 hectares of water bodies and Æ MT 257 hectares of forested wetlands associated to - |. 28 (es deforestation, erosion, flooding and pollution or the CRÉT re à s sr rivers, wetlands, mangroves and riparian forests. 4 Le dé : + 2 : f Table 1 - Land use change observed between 1986 and ‘tn a LEE à Es ; 2010 in the study area p 57 *< à £ Fe. Fopenspae 1 0 | 0 | © | D a raioppé # moyenne intonsté M Pier d'eau mn Développé faible intensité Les zones humides boisées [Forest "| 0 | 0 | 0 | PET M Mare can nant ÆM Marécages a vegetation herbacee DM Océan Atlantique mm Stérie Figure 12 and 13 - Land Use for 1986 and 2010 respectively for the NDC Source: Remote sensing analysis from satellite imagery 1986 and 2010 [ee EMERGING= Lo NORTHERN DEVELOPMENT CORRIDOR, HAITI 14 1nitiative ERM [page 22] 3.2.1 Urban Areas October 2013. Between 1986 and 2010 the urban One fourth of the 2010-2013 growth is the area of footprint grew 419 Hectares at 17 hectares per the PIC (246 Ha), the other three thirds are housing 2010 LANDSAT anaace described above, and'ae year) in contrast with a 340 Ha between January projects such as EKAM, the UNH-RHC in Limonade third reference point was also obtained from a high 2010 and October 2013, at a rate of 113 Has per and organic urbanization processes associated with resolution image collected and used by the IDB to year. Figure 12 illustrates the urban growth be- these development projects, along the RN6, around create a digital terrain model for the area. Urban tween 1986 and 2013, and Figure 14 focuses in on the existing towns and in the region between Limo- land use was classified into four categories based key areas of the NDC. Figure 15 provides the urban nade and Quartier Morin, as seen in Figure 14. on imperviousness levels (the area that ceases to be intensities {in terms of density of development) as his devel h dth banizati surfaces such as buildings and roads): i) developed " ° °P high intensity, ii) developed medium intensity, iii) developed low intensity; and iv) developed open F LIMONADE ARACOL | Trou ouh | ÿ TERRIER ROUGE space. The first three are occupied by buildings and Es … Î Î the fourth one is composed by parks, roads, 2] d & - NT dr: Fils . se cemeteries and infrastructure inside the urban 4 LM È . ' 4 j footprint or connecting discontinuous urban areas, ; s| * , d * such as the RN6. é , An average urban density of 78.09 inhabitants per $ hectare was reached in 2012 and surpassed the 100 Inhabitants per hectare in Caracol and Limonade, 2 however this analysis of imperviousness levels did os vos | _od EURE | not show areas classified as developed high intensi- 1 le. Î ty. Developed areas of medium and low intensity / XY RS | increased significantly during the 24 years of the è ds £ ‘à Es analysis and the urban footprint of the NDC grew #$ # { ; more than three times its size between 1986 and ec s } 2010, while open space, basically road infrastruc- Ê« * ture, doubled during the same period. — 2, Te * ; Comparing the 1986 and 2010 urban footprints C1 Zone d'étude shows that the NDC is mainly developed with lower Limite commune levels of density, predominantly compact built envi- | mu SL LS LE AUS | Cne ronment concentrated around traditional settle- MM Empreinte urbaine 2010 ments and more recently along roads. = Smerep on mm Océan Atlantique In addition, after the 2010 earthquake and the con- Figure 14 - Urban footprint growth 1986-2010-2013 for key urban areas De sean rh à tal 304 H eee Source: Remote sensing analysis from satellite imagery 1986, 2010 and 2013 NORTHERN DEVELOPMENT CORRIDOR, HAITI 15 NN ERM [page 23] urban to 1.2 percent at the beginning of 2010 and Like most part of Haïti, the mountains of the north- areas might have been displaced by urbanization 1.9 urban in 2013. These growth trends, in combi- ern coast have been deforested due to demandés for and charcoal production, and areas around Terrier nation with the development and infrastructure agricultural land, charcoal and construction materi- Rouge have faced irrigation problems. projects expected for the area, will attract more als. There was also a sharp decrease of cultivated population and are likely to continue the urbaniza- areas, despite the rich quality of soils present in the 3.23 The Hydric System tion process. area of study, in contrast with a steep increase in Four land use categories were evaluated to analyze ? ° scrub wetland and emergent wetlands. However The findings demonstrate a lack of forested areas. halted production due to market conditions, other the forested wetlands and scrub wetlands were consolidated because they belong to the mangrove RAGOLT ] RUES 5 MARNE . ecosystem in the area of study. The area of water J d } bodies has not changed significant with a reduction p F—. + nl û of 49 hectares (0. 5 percent) between 1986 and | Bus # / 2010. pi Î 4 * j The relative isolation of the northern coastline of \ Caracol and Limonade has protected its hydric sys- \ tem and ecosystems; however, the environmental À impacts of deforestation, erosion and flooding have L slowly deteriorated the riparian forests of tributary vonphn pre Be | rivers, reducing the area of forested wetlands from ut À | ed H 4,318 to 4,061 hectares (0.6 per cent). Finally, it is . | med A À important to clarify that the areas subject to emer- # ss | HS : LL gent wetlands were difficult to identify because the : ; study was performed on satellite images without s \ #: j : cloud cover and they correspond to the dry season. { , di 3.3 Physical, Biological and Hydrological FR Nr À Baselines C1 Zone d'étude ; Regional baseline information has been obtained : Limite commune from a variety of sources. A GIS database has been es Ever hp =D ie assembled from the various information sources Développé faible intensité including: Récif corallien cd ul + Topography, which shows that more than half Figure 15 - Urban intensities in 2010 of the study area is relatively flat and its southern border is part of the Grand Massif Source: Remote sensing analysis from satellite imagery 2010 du Nord, a mountainous formation that ele- == EMERGING = Lo 6 SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 16 EX EH % IDB ERM [page 24] vates from the southern border of RN6 to- tems, the plains have smaller ecosystems VIII. It is also noted that ecologically-valuable wards the center of Hispaniola island. composed of disperse woods and riparian lands belonging to classes VI, VII and VIII are forests. located closer to the coastline or in the moun- ° The hydric system composed by superficial tain areas, and coincide with marine-coastal water, underground water, riparian forests + The normalized difference vegetation index and highlands ecosystems in the region as dis- and flood plains, as well as all bodies of water, (NDVI), an indicator of vegetation health, cussed above. wetlands, reservoirs and protection buffers used to monitor degradation processes. NDVI around them, aquifer recharge and discharge is the result of a remote sensing process in These information sources have been used to cre- areas and the coastal and inland flooding are- which vegetation health is evaluated accord- ate individual GIS layers and maps of the area spe- as defined by the Risk Study (see Section 4). ing to the reflection intensity in different color cific to the technical topic, and Appendix 1 contains bands captured by the satellite, in particular all of the individual maps. Figure 16 provides a e Strategic ecosystems and protected areas, infrared. In general, vegetation in the NDC is composite map showing the collective main ecolog- comprising the region’s two main strategic healthier than in most of Haiti, however, it is ical system (comprising physical, hydric and biologi- ecosystems: marine-coastal and highlands. important to highlight that the healthier vege- cal aspects) for the NDC. Slightly less than half of The Three Bays Marine Park was created the tation in the NDC corresponds to agricultural the surface area of the main ecological structure in December 2013. This covers an area of ap- areas. In contrast, the highlands ecosystems corresponds to three main groups: areas with po- proximately 90,000 hectares that includes the are less healthy, which explains why the re- tential for forest protection (22%), areas crucial for bays of Limonade, Caracol and Fort Liberté, as mote sensing process classified most vegetat- the protection of water (14%) and key ecosystems well as the Lagon aux Boeufs, east of Fort ed areas in the mountains as scrub. (12%), mainly mangroves, disperse forests and Liberté. The newly established Three Bays dunes and beaches. Marine Park will help protect the mangroves, + Agrological quality of soils classification, eel grass beds, reefs and habitats housing im- where the NDC area is rich in land classes | portant fisheries that are crucial for providing through IV, while limited in classes V through livelihoods to nearby communities. It will also help protect the area from storm surges and provide local communities with ecosystem tt ue, services such as carbon sequestration, tour- ism value and more. The MPA is also home to RO numerous threatened species, including sea . Sie turtles, whales, manatees and migratory = birds. The highlands ecosystem is composed by areas adequate for reforestation and ripar- ian habitats, crucial for the improvement of the hydric systems, from the head of the riv- ers and their watersheds. Like most parts of Haïti, the mountains of the northern coast have been deforested pushed by demands for F : É > s : : Agricultural land with Grand Massif du Nord:in the distance agricultural land, charcoal and construction materials. In between the two main ecosys- == EMERGING Lo 6 SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 17 EX EU % IDB ERM [page 25] [” l'a ou ; rm] F2 CA Î nf" — + LE Berre f = Rés TT j Le Fr n CS - » = Pre nn { 16 SS | 24 ve 7 5 2 ) } A y j & | d Z , : PET 2 & 5 ] < 2 = £ ë 47 F4 Fr "4 - 0 k } ? Be à” (' L ( CE à »: * AURRNRE D: EN à C1 Study area Reservoirs — Route principale ce | en «2 = Limite commune Zone humide — Route secondaire Dther 2357 Section communale DM VI Hydric system — * MM Zones urbaines En Vi Recif corallien Ein MM Espace boise VII Océan Atlantique D Lac - étang — Rivière principale nn . Mangiier Rivière secondaire IE Hs Plages et dunes MMM Lits fluviaux et alluvions récentes Fiogrs 2 arm | — Map source: From geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. E EMERGING = R 4] Lo «TE SIDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 18 itatire ww ERM [page 26] 3.4 Cultural Heritage Northern Haïti has a vast and valuable cultural her- itage, documented through field surveys and histor- ical research, asillustrated in Figure 17. The region possibly has the earliest Pre-Columbian Amerindian Ep sites in the Caribbean, and the earliest documented 4 AI pr in Haïti (early lithic, possibly 4,000 B.C.), as well as À F = Ra sites associated with the Arawak and Taino cul- EE \ A è à tures. La Navidad, the first known European settle- ÿ s [TS |] ment in the Americas, is located in the shoreline £ - < N GET A TS between Limonade and Caracol, and in front, with NÉ CS per PI, 6, > s its traces lost in the Ocean, is the potential site of Do hat Se ñ 3 the wreckage of Christopher Columbus’ Santa Maria I77 LS ss il ship, the best known of several underwater archae- D bed : 8 ology, traditional shipbuilding, fishing traditions, 3 and the historical implications of maritime trade #3, F F4 throughout the Caribbean Basin. S Spanish and French colonial heritage left forts, forti- fications and military buildings, and later, inde- d pendence, with the earliest and most significant slave re-volts, ended in the first free Black Republic TS F ré of the Americas and their heritage. The eighteen, PS os suses rar mme eu 1) nineteen and twentieth century contributed with C1 Zone d'étude + Période occupation américaine! Période espagnole sugar, indigo and sisal plantations and the remains Parc des Trois Baies + Sites sacrés du vodou Colonisation francaise f their facilities, includi : t ts fl M Zones urbaines T Pélerinage catholique vodou EM Période amérindienne oi eir facilities, Including Some Investments from * Période amérindienne + En Bas Saline Zones de naufrage potentiels United States. + Période espagnole * _ Naufrage de la Santa Maria — Route principale + Colonisation francaise + _ Colonies principaux de la côte nord —— Route secondaire : : : : + Période haitienne PNH CSSR Récif corallien Such history also left a rich architectural heritage * Période flibustiers, boucaniers EM Architecture vernaculaire EM Océan Atlantique and construction techniques: traditional wattle- . h u . Figure 17 - Cultural heritage and-daub vernacular architecture, Caribbean wood, rubble and ashlar and stone masonry colonial struc- Map source: From geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), tures and early modern reinforced concrete typolo- USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sens- : . ing analysis from satellite imagery 1986, 2010 and 2013. gies. Urban fabric and settlement patterns mostly display European colonial array in contrast with disperse Caribbean configurations. > EMERGING = Lo NORTHERN DEVELOPMENT CORRIDOR, HAITI 19 initiative ERM [page 27] 3.5 Urban, Commercial and Infrastruc- ture The urban and infrastructure development has also been mapped, and has comprised the following: e Industrial uses such as the PIC and mining concessions; + Roads acknowledging the hierarchy of roads including national routes, urban roads, sec- ondary roads and tertiary roads. + Publicutilities, acknowledging that the NDCis significantly undeserved and the 98% of the study area that is currently rural lacks basic utilities, such as water supply, wastewater collection, storm water drainage, solid waste management and electricity; and + Employment and key economic activities such as commercial (transnational, regional and lo- cal markets), industrial (the PIC), services (ho- tels, financial institutions and fuel stations), institutional (health facilities, universities, po- lice stations, wastewater and solid waste treatment areas) and mining activities. + Social services and infrastructure including health and education facilities. These information sources have been used to cre- ate individual GIS layers and maps of the area spe- cific to the technical topic, and Appendix 2. == EMERGING NORTHERN DEVELOPMENT CORRIDOR, HAITI 20 initiative ERM [page 28] 4. HAZARD AND RISK ASSESSMENT materialized as earthquakes, and can trigger northern regions, where impacts of include STUDIES tsunamis. The boundary between the two flooding, loss of life, livestock, destruction of plates is not defined by a single border line, agricultural lands, erosion, river siltation, in- This section presents the work undertaken for the but instead by a zone where several tectonic creased incidence of water-borne diseases, baseline risk assessment and vulnerability analysis fault systems which run across the island and and famine. of the NDC. It outlines the results of a probabilistic show evidence of historic and/or pre-historic risk assessment, impact analysis and mapping of activity. Within the system, there have been ° _Inland Flooding: Flooding is by far the most prioritized hydro-meteorological and geophysical historic evidence of several destructive earth- destructive hazard in Haiti. The country’s hazards. The analysis utilizes a common risk frame- quakes, the most recent being the earthquake most populated cities are all nestled in flat work, where risk is a function of hazard, exposure of January 12th, 2010. The northern coast has coastal valleys. Widespread deforestation in and vulnerability. The results from this aspect of the been struck repeatedly by earthquakes and the upper reaches of these valleys, coupled study will assist decision makers: tsunamis. with the lack of storm water drainage infra- structure in urban areas, creates an environ- ° better understanding hydro-meteorological ° _ Hurricanes: The World Banks Climate Risk ment conducive to inland flooding. and geophysical hazards; and Adaptation Country Profile states that ° identify which assets are most exposed to the over the past 30 years, Haiti has been hit by + Coastal Flooding: Settlements along the coast natural hazards; and six hurricanes, where the impacts of these and in low lying areas, along with damage to + understand the most serious potential conse- storms can be expected in all areas, including and dwindling mangrove assets and deteriora- quences of climate change such as physical damage, economic loss and loss of human life. 3 Fr. M De 4.1 Prioritized Hazards vi % br: nee j 7 fr À L ” ny g Based upon a review of available hazard records : À à: fr kr 4 mer and information, discussions with the IDB specialists # so S « - and Haïitian stakeholders (see Table 2), the follow- LE d ing five hazards were prioritized and studied: + ni e Seismicity: Haiti shares the island of Hispanio- k Bed & | ie L. . : AA 4 €. la with the Dominican Republic. This portion ES (PRES m 2 AC Lu F KI LA 8, { of the Greater Antilles is located at the north- De " ÿ ern edge of the Caribbean tectonic plate, at _æ: = x 4 BEA its boundary with the North-American plate. Pre = ” | 4 $ The limit between the two tectonic plates is ns NH |A defined by a strike-slip left-lateral motion, ne À { à since the Caribbean plate moves relatively to > L the east-northeast and the North-America = plate moves relatively to the west. This kind Elood'MEmersde of interaction induces a strong liberation of | mechanical energy, typically and frequently > EMERGING = Lo et SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 21 EX EH hd IDB ERM [page 29] tion of littoral environments, result in expo- Section 7.12 then presents risk reduction recom- tries in order that they may develop climate change sure of populations and communities along mendations, where hazard losses are compared and scenarios at national centres of excellence, simulta- the northern coast to coastal flooding. a series of sustainability recommendations are pre- neously building capacity and drawing on local cli- sented for each hazard, as well as a cost-benefit matological expertise. ° Drought: The north of Haïti has frequently analysis for five mitigation strategies. experienced repeated droughts, brought For this project, the PRECIS projections assumed an about by a combination of erratic rainfall pat- 4.3 Climate Change Projections A1B emissions scenario. Under this scenario, as terns coupled with a limited water manage- defined in the Inter-governmental Panel on Climate ment infrastructure. In Haiti, droughts have 4.3.1 Approach Change’s (IPCC) Special Report on Emission Scenari- destroyed crops, reduced agricultural produc- : : : : . os (SRES), it relates to a future world of very rapid : VE Crops, reduce agricu'tura proeuc The available information and studies on climate ( ), . ly ap! tion, and decreased food security. Missing or . . economic growth, global population that peaks in : change scenarios relevant to the study area in : : : poorly managed water infrastructure makes Le . mid-century and declines thereafter, and the rapid : : Lo northern Haiti were researched and summarized. : : the agricultural regions and hence, the liveli- Lu : . . introduction of new and more efficient technolo- : The key findings were then inputted into the risk : hoods that depend on them, particularly vul- : gies. A1B assumes a balance across all energy : : assessment work in order to enable the natural : s : nerable to a changing climate. . . . sources (where balanced is defined as not relying hazards of flooding (both inland and coastal), hurri- . L dd ht to b d when taking int too heavily on one particular energy source, on the Table 2 — Haitian stakeholders canes and crougnt to be assessed Wnen taking Into assumption that similar improvement rates apply to consideration future climate change predictions. Il | d'end technologies). N Stakeholders engaged a energy supply an enruse ee no ogies . No ° Comité Interministériel d'Aménagement du Territoire The Climate Studies Group (at Mona, Jamaica), part doses Fe able for ot s emissions sce- (CIAT) of the University of West Indies, was commissioned Pate ort NA Sas oncentration ° Direction Nationale de l'Eau Potable et de l'As- to undertake an assessment of climate change pa- athways ( ). sainissement (DINEPA) rameters and projections applicable to northern ° Ministère de l'Agriculture de tu P : L PP : 4.3.2 Temperature and Precipitation 0 Ministère des Travaux Publics, Transport et Communi- Haïti. This work is presented in Appendix 3, and . : : cation (MTPTC) comprises the following: Future change data are provided for five variables ° Ministère de l'Economie et des Finances (MEF) when considering an A1B emissions scenario. For ° Centre National de l'Information Géo-Spatiale (CNIGS) ° Projected Changes in 5 Atmospheric Variables four of the five variables the data was provided as + Direction de la protection Civile (DPC) for selected grid boxes over Haïti from the absolute change. These variables are: minimum . Institut Haïtien de Statistique et d'Informatique (IHSI) PRECIS RCM, February 2014; and temperature (ci), maximum temperature (0), *____Bureau Des Mines Et Energies + Evaluation of trends in sea levels and tropical mean temperature (°C) and 10 m wind speed (m/s). storm intensities, February 2014. Percentage change is provided for precipitation. 42 Methodol The change for each variable and for each period is : etnodology The climate change parameter projections were calculated for the 20405 consistent with the time The process for the risk and hazard assessment obtained thraugh the running of PRECIS (Providing horizons of the overall ESCI study. Table 3 summa- comprised the following steps: REgional Climates for Impacts Studies), which is a rizes the future change data ranges. Regional Climate Model (RCM) and was developed i. Climate change assessment; by the Hadley Centre (UK) in order to help generate ii. Development of hazard profiles; high-resolution climate change information for as ii. Vulnerability assessment; and many regions of the world as possible. PRECIS is iv. _Loss estimation. made freely available to groups of developing coun- > EMERGING TRS SUSTAINABLE IDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 22 initiative ERM [page 30] 43.3 Sea Level Rise and Tropical Cyclones An assessment of current literature on current and Di projected trends in sea level rise and storm intensi- ! ties with particularly emphasis (where possible) on SOÙ N ul D future values for the Caribbean region was also AM , r undertaken. >, ZON DANJE Sé gi ; 1 1 'l ; C1 IURERES Table 3 — Summary of Climate Change Projections for ré F 4 FD, | ll 1111 f Ï the 20405 for Northern Haiti Î | 2) 7 le X | (] | | 22 É \ 7, Pie = 22, L ï Fr 16 STE) Mean Min Temp Max Precip 2 Due OP É | = ;) ts Temp (°C) (©) | Temp(o) | (%) A LIN Eu Le } NDJ 137to 1.62 to 14ito |-1103to ) PESTE Se | 2 4 1.74 1.85 1.76- -3.01 IE £ mar mn CUTS NOW bé La …. FMA 1.36 to 1.63 to 1.45 to -0.83 to p F £ »' 2 = 1.60 1.94 1.78 9.98 co Di Ces “4 mr. 1.77 2.06 2.11 -443 ASO 148to 172to 1Sito |-15.72to En 1.91 2.02 2.07 -7.14 Tsunami watning in Cap Hatien Annual 1.38 to 1.67 to 1.48 to -9.50 to 1.74 1.97 1.93 -3.69 Data is averaged for over three month seasons: November- intensity and fraction increases in the number of reflect projected climate change scenarios. The January (NU), February-April (FMA), May-July (MI) and most intense storms. resultant maximum wind speed with projected cli- August-October (ASO), roughly consistent with the Caribbean : dry season and wet season mate change scenario are compared to modeled | The IPCC Fifth Assessment Report (IPCC 2013) indi- wind speeds for Haïti. For sea level rise, projected increases in global cates that the frequency of the most intense storms mean sea levels were taken from IPCC (2013), rela- Is more likely than not to increase by more than tive to 1986-2005 as a baseline; suggest a likely +10%, while the annual frequency of tropical cy- range of 0.17-0.38m increase in the 2046-2065 clones are projected to decrease or remain relative- timeframe. ly unchanged for the North Atlantic. For storm intensities, simulations are consistently This suggests no major change in the frequency of | finding that greenhouse warming causes tropical hurricanes in North Atlantic region comprising Haïti. cyclone intensity to shift towards stronger storms The SRES scenario A1B for study area of Haiti sug- by the end of the 21st century (2 to 11% increase in gests that the wind speeds are projected to de- mean maximum wind globally). When simulating crease by very small magnitude of 0.25 m/s ( 0.559 21st century warming under A1B, the present mod- mph) over the projected for the 20405 relative to els and downscaling techniques suggest increases in the 1960-1990 baseline. These projected changes have applied to model wind speed over the return period to develop wind hazard maps for Haïti that == EMERGING TRS SUSTAINABLE IDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 23 initiative ERM [page 31] 4.4 Hazard Profiles and southeastern coast and strike-slip fault zones applied to determine the seismic hazard for the that transect the northern and southern portion. study area, which has involved assembling relevant For the five prioritized hazards, profiles were devel- seismic data and information for Haiti, geo- oped by investigating the various natural hazard While there is a verifiable record of earthquake referencing the information to the study area, and occurrences within the study area. The hazard iden- occurrences dating back more than 500 years in the overlaying against local geological characteristics tification process included consultations with key Caribbean, in general, the occurrence of seismic and slope indices in order to generate earthquake Governmental, NGO and community stakeholders, events in Haiti has been poorly recorded. A review hazard maps, expressed in terms of peak ground as well as observations during field missions. Infor- of the information available has indicated that since acceleration (PGA) soil values at 10 m horizontal mation on past hazard events were also download- 1750 approximately 19 major earthquake events resolution. The PGA is a measure of how hard the ed from the Disaster Information Management Sys- have been recorded, culminating in the tragic event earth shakes in a given geographic location, in other tem (http://www.desinventar.net/). The profiling of in January 2010 which claimed the lives of over words the intensity of an event. A series of maps hazards includes determining the spatial extent of 200,000 people. has been developed for different return periods, hazards, where possible (i.e. maps), understanding and Figure 18 provides an example hazard map the frequency or probability of future events, their Appendix 4 provides full details of the methodology magnitude, and climate variability factors that may affect their severity. Each identified hazard has unique characteristics that can impact northern s : Haiti. Appendix 4 provides the detailed hazard pro- A » ” <- files, and these are summarized below. | éd : j À É > : AA.1 Seismic Lo hi ” d l b À ES An earthquake is caused by a sudden motion or | Le trembling of the earth due to an abrupt release of { stored energy in the rocks beneath the earth’s sur- À s ; face. When stresses due to underground tectonic Al { forces exceed the strength of the rocks, they will es S \ abruptly break apart or shift along existing faults. : { | ; The energy released from this process results in Le ) Ÿ à vibrations known as seismic waves that are respon- DR 4 : j sible for the trembling and shaking of the ground \ | during an earthquake. 4 \ 27. Len Ÿ d [TT study Area The seismic hazard in Haïti has its origin in the in- & ; PGA (g) 475 YrRP| teraction of the North American and Caribbean No, 4 ur plates, which have a relative eastward movement à. à Lis EE of 2 cm/year (20 mm/yr). The island of Hispaniola is Ê considered a complex area of deformation which presents both subduction zones off the northern Figure 18 - PGA probabilistic seismic hazard map for 10% probability in 50 years, i.e. 475- year return period == EMERGING Lo 6 SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 24 a gs F IDB ERM [page 32] developed for the study area. One of the most serious components of hurricanes areas, particularly when storm surge coincides with is high winds. Because of the extensive size of a normal high tide. The assessment indicates that areas at higher seis- catastrophic hurricane, a storm need not pass di- mic risk (as indicated by the darker areas in Figure rectly over Haiti to cause severe damage. A hurri- The storm surge is produced by water being pushed 18 which correspond to higher PGA values), princi- cane passing within close proximity to the island of toward the shore by the force of the winds moving pally due to the underlying soil conditions, are Hispaniola can also cause major damage to proper- cyclonically around the storm. The impact on surge those areas closer to the coastline where softer and ty and even loss of life. Essentially there are no of the low pressure associated with intense storms deeper soils exist. areas of Haiti that are free from hurricane force is minimal in comparison to the water being forced winds. The coastal and low lying areas, such as toward the shore by the wind. The intensity of the 4.4.2 Hurricanes those of the study area, experience the first effects storm surge is affected by the width and slope of Hurricanes and tropical storms are large-scale sys- of damaging winds. The rains that accompany hur- the continental shelf. A shallow slope will potential- tems of severe thunderstorms that develop over ricanes are intense and last for several days. Intense ly produce a greater storm surge than a steep shelf. tropical or subtropical waters and have a defined, and prolonged rainfall, winds and pressure can The narth af Haiti is prone to storm surge. organized circulation. Hurricanes have a maximum cause both coastal flooding (see Section 4.4.3) and . sustained (meaning 1-minute average) surface wind inland flooding (see Section 4.4.4). Towns such as Bor de Mer de Limonade, Caracol speed of at least 74 mph; tropical storms have wind : un and Phaeton are susceptible to coastal flooding speeds of 39 mph to 74 mph. The methodology developed for the identification caused by storm surge. The American Association of of wind hazards for this study was based on numer- Architects, indicate that these settlements are in a Hurricanes get their energy from warm waters and ical modeling of hurricane motion using existing precarious location to shoreline {American Institute typically lose strength as the system moves inland. models and verifying and calibrating this work of Architects, 2012). There 1, however, limited Hurricanes and tropical storms can bring severe against the local study area parameters, as well as documented history concerning storm surges in winds, inland flooding, storm surges, coastal ero- building in the projections for future climate change Haïti, let alone well documented instances of sion, extreme rainfall, thunderstorms, lightning, and implications. Appendix 4 provides full details of the coastal flooding within the NDC. tornadoes. Hurricanes and tropical storms typically methodology applied to determine the hurricane . . have enough moisture to cause extensive flooding. hazard for the study area. To assess coastal flooding, a regional model was utilized and adopted to understand wave and surge Haïti is among the most hurricane-prone locations These results model wind speed over various return heights in the study area. The information utilized inthe world. In 2004, the Food and Agriculture periods in order to develop wind hazard maps for for this study effort ss derived from the Atlas of Organization (FAO) reported that during a period Haiti that reflect projected climate change scenari- Probable Storm Effects in the Caribbean Sea, which from 1909 - 2004, forty-seven (47) tropical storms os. Hurricane risk, associated with wind speed, is a was developed under the Caribbean Disaster Miti- and hurricanes hit Haiti. From 2004 to 2012, twelve relatively homogeneous factor across the study gation Project (CDMP), a joint effort of the Organi- (12) wind storms have made landfall in Haïti, Data area, and therefore no discernible geographical zation of American States (OAS) and the US Agency from the Prevention Web , which provides infor- variation was noted for the study area. for International Development (USAID). mation on human and economic losses from disas- k : : ters, indicates that between 1980 and 2010, over sus Coastal Flooding Masai saeneignts for four return ou four million (4,171,407) persons have been affected High waves associated with tropical storms and et specific oints AO te Hal eo Far by hurricanes in Haïti, with 4,390 deaths caused and hurricanes are potentially very dangerous and dam- this study, wave height and surge heights that were scrnomie impacts for the same period of US$ 822 aging to the coastal settlements due to the storm reported for Cap Haitian were adopted for the en- surges which can cause extreme flooding in coastal tire study area. These water levels were then pro- > EMERGING Lo sl * SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 25 EX EU % IDB ERM [page 33] jected onto the coastal land using a GIS to demar- Figure 19 shows the 50-year return flood map for Table 4 - Coastal flooding projections (including for cate the horizontal extent of inundation. The effects the study area showing the potential for inunda- climate change to 2040) of projected climate change were also integrated tion, andillustrating the vulnerabilities of coastal _ _ into the assessments, including minor changes in settlements such as Caracol, Jaquezy, Borde de Mer RER Predicted Flood Predicted Area of : : : A : Period Level (m) Flooding Inundation sea level rise as per Section 4.3.3, are represented de Limonade, and Phaéton to flooding. (km£) on the extent of flooding. Appendix 4 provides full details of the methodology applied to determine the coastal flooding hazard for the study area, and Table 4 presents the projected results of coastal flooding. 4.4.4 Inland Flooding Floods can arise from a variety of causes. The most commonly understood floods occur when water levels in rivers rise and the waters overtops their à PS banks, and adjacent floodplains and lowlands are 4 s EN subject to recurring floods. This type of flooding LS _—. : —— usually occurs after intense or prolonged rainfall. A SHARE e Be } 1) EE à . SR second type of flooding can also occur due to heavy ia : . 4 | \ | 4 rains where infiltration of rainfall is impeded or p k x is - Do Phaët A 5 Ï bn, (through either impermeable soils or increase in d Ph F4 ABfnonade . A EE ? Pallete Le by : impervious surfaces due to development). É | j 2) ‘ #. grerrier Rouge f Le Î Ë Floods in Haïti, as in other Caribbean islands, follow Ë Es Frou-du-Alord (Q J — i à tropical weather patterns. Haïti has two distinct SC * \ € 5 | 6 rainy seasons, one from April to June and another FLAT 4 - L 4 2 DD) G PE ME from October to November. There have been a LE 2.4 > Ps LE 7 Re du D, number of large-scale devastating flooding events Ê H $ £ i fi E, in Haïti over time and most of the flooding events x D KA of (| have been linked to large-scale climatic events (i.e. # É Le = : STE 7. À tropical cyclones), as well as more recently smaller À PRES. ? L i —E LE low pressure systems which have impacted Haiti on CT Zone d'étude Récif corallien a yearly basis. !___! Limite commune DM Océan Atlantique M Zones urbaines Très faible Haiti’s rugged and mountainous terrain coupled = Here with environmental degradation and poor water- En Éievé shed management has created optimal conditions M Très élevé : for flooding problems. Figure 19 — Coastal flooding with climate change projections for a 50-year return period = EMERGING = Lo et # GIDB À NORTHERN DEVELOPMENT CORRIDOR, HAITI 26 À ivitiative ERM [page 34] ERM Flooding Study The Trou-du-Nord watershed measures 110 km?. shed measures 680 Km?. The Rivière Caracol and . . The Rivière Franiche, Rivière Pilette and Rivière Rivière Cartache are main tributaries to this river. As part of this ESCI study, a flooding study was per- Lo : Lo Loi : : u es x Cabaret are the main tributaries to this river and The Rivière Caracol is a permanent river with a con- formed on the two principal watersheds in the : . : Lois : . are intermittent streams and are dry part ofthe stant source of water, while the Rivière Cartache is study area, the Trou-du-Nord and the Grande Rivi- Lo pas : . . : : : : x . , year. The Petite Rivière, an intermittent river, is an intermittent stream and is dry part of the year. ère du Nord (see Figure 20). ERM's study only as- ou : se d ri lated floodi also located in this basin and drains into the Trou- sessea river-related flooaing. du-Nord plain. The Grande Rivière du Nord water- The vulnerability of these two watershedbs is signifi- cant, with widespread deforestation, clearing of q land for agriculture and increased urbanization all . Le contributing to the flooding problems in the region. É a FA The commune of Quarter Morin, which is situated … NX he in a moist, low lying alluvial plan and bordered on #1 6 the east by the Grand Rivière du Nord, is prone to kr É $ Le flooding. Several factors have worked to increase à LÉ -'R \& the susceptibility of flooding, including more in- ...: D à %, : da s tense climatic events, increased run-off, and the Un } accumulation of debris downstream. Limonade is F je l bordered by the Grand Rivière du Nord on the west. SPA T: FES ns While the Barrage de Tannerie previously helped to Mn 1 F contain flood waters and provide irrigation during $ a ÈS. . the dry season, the dam failed in the 1960s and has 07, Le not been repaired. Intense rainfall causes flooding 2 s . Ca and the accumulation of water in low-lying areas À which are slow to drain following flooding events [ due to limited or inadequate storm water drainage infrastructure. Limonade receives an average of 1200-1400 mm rainfall annually. In Trou-du-Nord, the terrain, rainfall and soil types, in both the mountains and plains, give rise to a fair- ly dense network of rivers. The Trou-du-Nord river is the most important river system. In low-lying portions of commune, there are a series of smaller tributaries and torrential gullies. Large areas of the plain are subject to frequent flooding caused by Figure 20 — Key watersheds in the study area torrential downpours. The heavy clay content of the soil causes erosion and results in frequent sediment Source: CNGIS . : . build up in streams. Urban areas are adjacent to the main river with development occurring in ripar- 3 EMERGING = Lo NORTHERN DEVELOPMENT CORRIDOR, HAITI 27 N ERM [page 35] jan zones. Historically, the city has been flooded analysis and a probabilistic simulation of rainfall of the challenges encountered. severely. which has considered climate change. Hydrologic modeling was undertaken to simulate the precipita- Table 5 presents the projected results of inland Flood risks are also present in the northeast portion tion-runoff processes, and hydraulic modeling of river flooding across the study area, and Figure 21 of the Terrier Rouge commune, sometimes impact- the main rivers identified above was also undertak- shows the 50-year return flood map for the study ing the city on its northern edge. To the south, set- en to develop probabilistic flood forecast maps for area showing the potential for flooding. The results tlements experience higher annual rainfall six return periods (2-, 5-,10-, 25-, 50-, 100-return shows that with climate change on an average flood amounts, and as a result, experience flash floods. periods) for the portions of the basin that intersect depth will increase by about 0.23 m (23 cm) across The urban development of the city is constrained by the study area, and including climate change pro- all return periods. low lying topography, which is prone to flooding. jections. Appendix 4 provides full details of the methodology applied to determine the inland river Table 5 — Inland river flooding projections (including for A detailed flood hazard assessment methodology flooding hazard for the study area, including some climate change to 2040) was undertaken which included a meteorological Return Predicted Flood Predicted Area of gen En gn aus Fun Period Level (m) Flooding Inundation \ Ô , Ç ) rom: pa — —— TT Te . FT À | IDB Water Study 1 A key information gap that has been identified as development of the PIC has progressed is a reliable quantitative assessment of water availability and ne | | ex quality for the PIC. This assessment needs to con- sider the water availability and quality of contrib- | | Legend uting surface watersheds, groundwater sources and vers their surrounding ecosystems, and to recognize the Flood 50 Year RP With Climate Change needs and demandés of all water users, as well as Dm nn 00 possible impacts of future development in the re- | gion including population growth, land use change, ES PET 5 oomas pu con TT new infrastructure, and climate change. For in- [ou Propose Stucy Are stance, the PIC site's primary source of surface wa- pen me = pe en en = ter, the Trou-du-Nord watershed, feeds into Caracol Bay, a potentially sensitive ecological resource. Figure 21 - Inland flood hazard map with climate change 50 year return period Challenges include a lack of data on the ecological == EMERGING Lo sé * SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 28 EX EU % IDB ERM [page 36] conditions and characteristics of Caracol Bay, the potential impacts associated with the PIC de- e The developed model can be used to design absence of data to assess critical environmental velopment currently and over time, particularly onsite drainage improvements in detail, as well flows to the bay, salinity intrusion inland towards related to flooding at the PIC and its surround- as infrastructure to prevent/mitigate inflow sensitive wetlands and groundwater, and extremely ings, and water quality impacts in the Trou du from the river floodplain. limited hydrometeorological and water quality data Nord river as it flows towards its discharge in e The water quality simulations suggest that the for the region, specifically for the Trou-du-Nord the Bay of Caracol; proposed wastewater treatment plant at the watershed. + Within the watershed: support the develop- PIC will not have a significant effect on the riv- ment of an integrated water resources man- er’s water quality, or on the downstream dis- Although preliminary estimates suggest there is agement (IWRM) plan for the Trou du Nord charge, based on key water quality indicators ample groundwater available within the underlying Watershed/MTA area in Northern Haiti; and that have been simulated. Massacre Transboundary Aquifer (MTA) to meet ° Within the country: serve as a pilot project for the PIC's water demands, the aquifer is believed to a future program designed to assess water The simulation model development results have be unconfined and overlain by highly porous, alluvi- availability at the watershed level throughout been disseminated through two capacity building al sands, rendering the aquifer vulnerable to con- the country, by scaling up the IWRM approach workshops (one in July 2014 at the PIC; the other in tamination. to the regional and national levels. October 2014 at CIAT, see Appendix 11 for details) with participation of several interested stakehold- Further, climate change has the potential to further By the time of this report’s publication, the project ers (CIAT, UTE, Ministry of Agriculture, IDB). These strain the availability and quality of water resources had made progress on the development, testing, workshops have been conducted as part of this in the area. Global climate models indicate increas- and implementation of a hydrologic and water qual- project, so that they contribute to solidify the inte- ing temperatures for Haïti, while a rising sea level ity simulation model for the PIC and its area of dis- gration of stakeholders and institutions that carry and an increased intensity and frequency of hurri- charge to the Trou-du-Nord river. This simulation out water resources management activities in the canes are likely in the future. It is therefore deemed model was developed using the IDB’s Hydro-BID Trou-du-Nord watershed. crucial to include existing climate projections and system, which was tailored for this project to ena- their impacts in the water management plans in ble flood and water quality calculation capabilities. order to provide a basis for successful adaptation in the area of the PIC and its surrounding Trou du The Hydro-BID 2D simulation model has been pa- Nord/MTA system. rameterized with high-resolution topography ob- tained through ESC/’s and ERM's work in the region, Based on this background, in 2014 the IDB commis- soil data analysis based on aerial photography sioned a non-reimbursable technical assistance (complemented with Hydro-BID’s data base), and project, water availability, quality and integrated rainfall data obtained through the SNRE in Haïti. water resources management in Northern Haïti, The modeling results to date indicate the following: focused on quantitatively assessing current and future water availability and quality and water de- e The PICis highly susceptible to flooding, even mand by all stakeholders as key inputs to integrated without rainfall occurring directly onsite. water resources management (IWRM) in northern Flooding occurs for a rainfall with return peri- Haiti at three connected scales: ods in the 25-50 yr. range upstream, while it floods if a rainfall event with a 1 yr. return pe- e Within the industrial park: improve the ability riod occurs onsite. to analyze existing (baseline) conditions and > EMERGING Lo sl * SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 29 EX See F IDB ERM [page 37] 4.4.5 Drought drought hazard (see Figure 22). The cumulative recurring phenomenon in the northern coastal . . . effects of longer dry periods are crop losses and plains and reported that “rainfall in wet mountain grougne mantsetin NAS that more families are becoming reliant on food areas [of the Northeast] has helped spur crop of tic Fall terne durin the two istinct assistance during the “hunger season”, which is the growth and development, while crops in Ferrié, Fort rainy seasons: À Pure and Oétober November three-month period between rainy seasons, in Liberté (except on the Maribaroux plains), Terrier Y : : AP Le which there is little harvesting and employment Rouge, Caracol, and Trou-du-Nord have failed due According to the World Bank, El Niño/ ENSO epi- mn ne : : . . opportunities. to the drought conditions in these areas, prolonging sodes have tended to delay the arrival of the rainy à D : season(s) and create drought conditions in the the lean season, which generally “ends in June... country. NATHAT (2010) “E à Rétional laval hazard The Famine Early Warning System Network (FEWS [and that]... virtually the entire northern region has assessment indicated that farmers are reportin Net) reported In August 2011 that the north and been affected by the drought which delayed the longer dry seasons and wetter and one rain 8 northeast were affected by drought and estimated start of the spring planting season, which eventually 8 “ NATHAT study has al . " that major crop yields would be diminished by 20 got underway in June with the onset of the rains.….” seasons. The study has also categorize percent. The FEWS Net furthered detailed what is a (FEWS 2012). most of the northern coast as being susceptible to i i + ï i i H i ï i î Of greater concern to stakeholders is the impact Légende : ER de. 6 that these short term fluctuations in precipitation | | Localté MNT (m.s.n.m.) + _—. © GFDRR will have on the surface and subsurface water sup- | Sèche EE 20 _ TS - an ply or the hydrological regime of watersheds that Countor du s - Pacs LT 2 ee 4 _— : . ee — LAS Re — | %BID Là &] ®] intersect the NDC. It takes longer to recognize the DM 540-1565 ET a. Sn D. | effects of hydrological drought on soil moisture . [_] 1366-2663 né Es AT … RÉPUBLIQUE D'HAÏTI levels, stream flows, as well as in groundwater and RCE en ae Démon devenus reservoir levels. The frequency of hydrological : ET ES PE alien on Ha drought is typically measured over the longer term n ps De, ml raté pere Crea and predicates a need for understanding of both De LE ion den the supply and demand for water. Hydrological me = 2. E s 452 mn drought concerned with the problems associated | - Re op / \gun | Susceptibilité à la with deficiencies in precipitation (the supply) and _ nn gen |" PR 3 5 jp me Res that of competing interests for water access and | b on DR. FC IN utilization (the demand). + VS Pate Cneg ue Dnssss RÉ ep CE AE pen ga sd e [180000 An assessment of drought has been performed fo- Chu pre Fe cure DR pains Soi or Leader dau : : Lo | ; Ê Tr ES me te, ae à es RL SR Ban ae Grn cusing on the influence of precipitation, and how MT Re RE s” p: rte Häsue, Unvarié Parde is is coupled wi e anticipated effects of de- CL + Per # RE 7 fee th led with the anticipated effects of d Frans Per IR, D. [Enr velopment and climate variability, will impact the us &x EE Fr : RS re current and future water supply. Current and future | | | — ee | | | | | | | TRE EEE water balance was estimated for the two main wa- li Ë î ÿ f i I 1 H Lie ee tersheds that intersect the study region. This study Figure 22 — Drought susceptible areas in Haïti as presented by NATHAT, 2010 does not seek to address the broader environmen- = EMERGING = Lo «E GIDB | NORTHERN DEVELOPMENT CORRIDOR, HAITI 30 À ivitiative ERM [page 38] tal, political and socio-economic factors that also and development growth presented in Section 5) reduce the available water stock and make the im- play a role in water access issues. that takes into consideration climate change. pacts of prolonged periods of water deficit more significant. Such impacts will become more pro- The hydrological drought assessment has been per- Table 6 — Summary of water balance for the study area nounced during years with below average rainfall. formed by estimating components of the classical hydrological cycle. The movement of water through Current |. Future Pro- 45.0 . . eme . jections, 2040 the hydrological cycle varies significantly in both 3; 0 time and space, The hydrologial cycle emphasizes je : n ÿ 300 8 25. ° Precipitation; Rural Population, Mm 24 | s | 3150 e Surface runoff; and Water Use and Demand, °° 3 15 5.9 00 ° Groundwater. Mm jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec For this analysis, the hydrological models that were Mm a ——_—_—_————————————————— 3 us nes for the pod parer dssessmnen 2one Figure 23 — Monthly variation in water demand and witn other conventional methods of hyarologica Water Availability Potential, Mm° availability (current conditions) assessment, have been used to assess potential - water availability for the watersheds of Grand River Run, Vs Potenitial Du Nord and Trou-du-Nord. The climate change = u | Ground Water Potential projections have also been integrated into the as- (Recharge), Mm° 103.7 152.9 45.0 sessment. n 400 . : : 5300 Appendix 4 provides full details of the methodology Surplus/Deficit, Mm À 250 applied to determine the drought hazard for the & 200 . . 3 : study area. Based on the estimates of water availa- $ 15.0 bility potential and demandés, overall summary of 8100 water balance is shown in Table 6. Monthly variation shows that during the dry season Jan Feb Mar Apr May Jun Jui Aug Sep Oct Nov Dec The current water availability potential is consider- of the year, the gap between demand and availabil- Month ably more than the demandés. Butin future projec- ity increases as compared to the wet season. There — Available Potential, Mm3 — Demand, Mm3 tions, the water availability potential is merely suf- is however a significant gap between the demand . hl d d'ond ficient to meet the projected demands. and availability potential particularly during the dry Figure 24- Monthly variation in water demand'an availability (projection for 2040 including climate season (June to October) as compared to the wet h The monthly variations in the availability and de- season. change) mand potentials are shown in Figure 23 for present day, while Figure 24 shows the monthly variation This water deficit during the dry period indicates for future growth (2040, using growth projections prolonged periods of hydrological drought, and climate change and projected growth will further == EMERGING TRS SUSTAINABLE IDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 31 initiative ERM [page 39] 4.5 Nulnerability Assessment e Limited access to basic needs such as clean low for the definition of an appropriate scale from . : water; and which to capture inventory elements. Appendix 5 The vulnerability assessment considers the study Environmental degradation such as deforesta- provides a more detailed description of the meth- area’s social vulnerability as well as the more tradi- tion. odology applied to characterize the assets exposed, tional assessment of the potential impacts to the and a brief summary is applied below. built environment. The social assessment seeks to 4.5.2 Characteristics of Assets Exposed identify a variety of indicators to inform of the un- This mapping process provided the basis for classi- derlying causes of vulnerability in the region, while The inventory of exposed assets involves under- fying buildings and for using a suitable classification the more traditional vulnerability assessment iden- Standing the distribution of people, buildings and hierarchy for the capture of a wide range of struc- tifies assets, characterizes structures and infrastruc- infrastructure that may be affected by natural phe- tures and densities. Such an approach is consistent ture so as to determine the built environment’s nomena. Remote sensing along with a rapid field with standard methodologies used to develop ex- potential performance to different levels of hazard assessment method was used to estimate the num- posure models and supports the required inputs for intensity (i.e. acceleration, flood depth, etc.). A ber and distribution of assets in the study area. This undertaking a probabilistic risk assessment by vulnerability assessment is performed to assess the included understanding the building density and providing an approximate spatial location of ex- specific damage and loss characteristics of each types for each land use category within defined posed elements for each block subdivision within asset identified. administrative boundaries. Administrative bounda- the section communal (see Figure 25). ries (i.e. section communal) were then subdivided 4.5.1 Social Vulnerability based on density of building footprints so as to al- Building occupancy mapping and distribution was This section informs the underlying causes of vul- nerability in the region and the potential impacts of ES NORTH ATLANTIC OCEAN gd: the identified hazards to demographic groupings in vo ON 5 the study area. Vulnerability considers the social smenane AD 4 — and environmental aspects that increase and accen- LES Ti è tuate impacts of hazard events. Social vulnerability Er À LS f us en... focuses on the economic, educational and financial a LR = [ea Ve: mé, CRAN > factors that impact the ability of people or commu- RÉ 2 — D Sgen ÈS: nities to adapt to hazards. The vulnerability of the STE K7 <> ? 1e « study area is exacerbated by the many factors that EN D 2 6 define the NDC including: Ve LC RS sue _ e Extreme poverty; = Fe x EN ° __ Demographics with more children present Et Q,. $77 Lu {and therefore vulnerable); nest ee Ÿ Ke > e Areliance on self-employment; St ous e Limited education opportunities; 2 ormerre pets e Gender inequalities; cuœace _ PE CA ren Study Area e Land access challenges; es Bcoruens: encens Migur"GE En pes | ° Food shortages and reliance on subsistence D farming; Figure 25 - Distribution of block boundaries in the study area == EMERGING Lo 6 SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 32 EX See % IDB ERM [page 40] then undertaken and structure classification of tioners, and were then aggregated to create eco- ards using standard risk assessment methodologies structures applied. Structural information is an nomic values for assets in the different blocks, such that take into consideration hazard parameters, in important factor in determining the vulnerability or as the example shown in Figure 26. Similar maps conjunction with damage ratios, to determine the how likely structures are to fail when they are sub- have been produced for commercial and industrial economic loss potential for each hazard (with the jected to hazards, such as wind pressure that ex- buildings. exception of drought). Appendix 6 provides a more ceeds their design. The models of the different detailed description of the methodology applied for types of infrastructure were determined based on A similar process is also followed for facilities and loss estimation, and the associated results, and a experience with the typical construction of Haiti. infrastructure, including hospitals, schools, roads, brief summary is presented below. The basic structural systems were grouped accord- and bridges, utilities such as water, electricity and ing to the following general construction: rein- wastewater. These estimates should be used to understand rela- forced concrete, masonry structure, unreinforced tive risk from hazards and potential losses and are masonry, and earthen. 4.6 Loss Estimation not intended to be predictive of precise results. de : : Uncertainties are inherent in any loss estimation LL : Probabilistic loss estimates were then determined Les : : Replacement values for buildings are then estimat- . k . methodology arising in part from incomplete scien- . . : . for seismic, hurricane, coastal and inland flood haz- Le : ed through a review with local construction practi- tific knowledge concerning natural hazards and eu un on eu their effects on the built environment. Uncertain- ties also result from approximations and simplifica- 3 — tions used in the development of hazard maps or 1944ne the inability to perform a more detailed inventory assessment. Vulnerability can be assessed by considering the ë potential and performance of the built environment _— : Do at different levels of hazard intensity (i.e. accelera- ÿ tion, flood depth, peak gust, etc.). Vulnerability functions were therefore developed for seismic hazards such as earthquake, hydro-meteorological hazards such as flood (both inland and coastal _— ge floods) and hurricanes. Legend Vulnerability functions relate the damage or loss pol ns É magnitude to a specific intensity of a hazard. Vul- | 0-2283122 nerability functions are specific to the particular M2 12637230 937N structure type and must be assigned to each asset 193ZN- EMI 6,372,370 - 13,757,970 , . os . DM 12.757.971 - 23.400,05 according to their characteristics. Figure 27 pre- ns prysiiieu sents an example vulnerability function created for es coastal flooding for a low rise masonry structure. FT Fr s re Te Appendix 6 presents the range of vulnerability func- Figure 26 - Distribution and Exposure Values of Residential Buildings in the study area tions generated as part of this study. > EMERGING = Lo os! SUSTAINABLE S NORTHERN DEVELOPMENT CORRIDOR, HAITI 33 EX EH % IDB ERM [page 41] The risk metrics described above, particularly, the choices for risk reduction measures. Appendix 6 12 AAL, can be used to provide an understanding of presents the full set of results from the loss esti- & 1 the spatial extent of losses and help to identify and mates based upon the above risk metrics, and Table É 08 | Il Il | prioritize the urban areas or localities that are un- 7 provides stakeholders with an overview of im- H os | | ( | | | der risk. A street light indicator methodology, pacts expected for each hazard and a summary of Ê where the colors on the map coincide with the level aggregate loses (maximum probable losses), which É ‘+ | of risk, has been used to map risk. include economic losses for general occupancy clas- Ë 02 / Î ses and infrastructure. 0 By mapping risk at the block level (see Figure 28 as 00 10 20 30 40 50 60 an example) stakeholders have a better under- 4.7 Future Studies Flood Depth, m standing of where potential losses will be the high- . …. : est and where monies should be allocated for risk The hazard and risk assessment studies presented RP RE reduction. All the areas and exposure categories in this section represent initial attempts to define that have a high and very high risk are automatic and understand the scale and magnitude of risks Based upon these results, estimates of the losses row ren La ren ren attributable to each hazard can then be made, and A PE A D D the findings can be used to support local and re- n_ hs gional decision makers in their understanding of the … : | | potential impacts of each hazard and allow a com- parison of hazards by quantifying potential impacts. à | | These estimates can be used to understand relative ] risk from hazards and potential losses. us 4 = ne ÉD p E The economic loss results are presented here using three risk indicators: ù j e Probable Maximum Loss (PML) - an estimate of losses that are likely to occur, considering ne In. mes existing mitigation features, due to a single Laser hazard event; Annual Average Losses (USD) e _Loss Exceedance Curve - plots consequences Éee Residential (losses) against the probability for different MN 2556-6960 events with different return periods; and = 0e are le. e Average Annualized Loss (AAL) - estimated or] JR 525 7742 Li) 25 5 10 Kilometers TT long-term value of losses to assets in any sin- gle year within the study area. Figure 27 - Risk Map: Average Annualized Loss for Earthquake Hazard, Residential == EMERGING = Lo 46! SUSTAINABLE KZ NORTHERN DEVELOPMENT CORRIDOR, HAITI 34 EX gs dd IDB ERM [page 42] facing the NDC. Further studies would enable this + Conduct bathymetric survey of northern work to be progressed and improved, and the fol- coastal area geared toward regional coastal lowing provides a summary of recommended next hazard modeling, sediment management, envi- steps for interested stakeholders to further advance ronmental assessment, emergency response. the hazard and risk work: + Study and map of severe repetitive loss and repetitive loss properties/ infrastructure, con- duct limited fieldwork, and evaluate hazard mitigation measures that would cost- effectively address clustered repetitive loss properties/infrastructure. + Conduct detailed land use planning study at the watershed level to understand the inter- play between deforestation in watersheds and the rapid urbanization so as to adequately quantify the levels of infiltration in upland are- as and overland flows in urbanized areas with the focus on improving storm water drainage infrastructure. Such a study along with a re- vised hydrological and hydraulic (H&H) model would provide the quantitative basis for as- sessing flood mitigation measures on basin and sub-basin level. + Related to the above, define and implement arrangements for the collection of data on pre- cipitation, including the entry of manual histor- ical records, so as to build an appropriate in- ventory of data for evaluating hydro- metrological hazards in the study area (flood- ing and drought). + Conduct detailed geological survey for the up- date of geology maps, with special attention for detailing of earthquake history, fault slip rates, and site soil types at an adequate scale so as to develop detailed seismic hazard risk maps for ground shaking and liquefaction haz- ards in the study area. NORTHERN DEVELOPMENT CORRIDOR, HAITI 35 EH ERM [page 43] Table 7 - Summary of Impacts and Loss Estimates by Hazard : Aggregate Economic ue us [mme . Damage to all structures (residential, commercial and industrial), reconstruction costs can be substantial (heavy walls poor reinforcement); . Loss of Life; . Loss of business and industrial production; Earthquake 2,500 . Loss of business production; : 1694.47 . Disruption of transport, failure to bridges; . Disruption to electrical network, substation damage; . Extensive damage to lifelines (i.e. water pipe networks, pumping stations, sewage treatment); . Social displacement and unrest; . Extensive residential and commercial property damage, roof failures, water damag- es . Moderate damages to industrial property, short term loss of industrial production . Loss of life, Hurricane 1,700 . Emergency evacuation needs; 815.81 . Commercial - Extensive property damage, roof failures, water damages . Property damage to critical facilities, transportation infrastructure (associated flooding biggest threat to transportation infrastructure), substantial damage to electric distribution network, utility disruption . Loss of housing, particularly on areas where there are rapid flows; . Undercutting of rural residential buildings or failure along river banks, where the water may erode soil under foundations . Damage to houses, commercial buildings and critical facilities due to inadequate Flooding 100 drainage infrastructure dueto prolonged flooding; | | 10.78 . Damages to transportation infrastructure, especially in areas of confluence of rivers (i.e. bridges, culverts), restricts transportation throughout study area. . Water and Waste Water Infrastructure - Inadequate drainage prone to blockage by silt, earthen debris, and many cases, garbage; damages to culverts; increase in con- tamination and secondary threats such as waterborne and vector-borne diseases . Extensive damage to coastal settlements, and supporting infrastructure. . Health and education facilities located within inundation areas prone to extensive damages as Flood- 100 . Damages to transportation infrastructure, restricts transportation. 93.47 . Damages to water lines below ground level and well above ground level . Increase in contamination and secondary threats such as waterborne and vector- borne diseases 5 EMERGING Lo TRS SUSTAINABLE ad IDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 36 as ERM [page 44] 5. FUTURE GROWTH PROJECTIONS tion presents this information. + The National University of Haïti Roi Henri Christophe Campus in Limonade (UHN-RHC); Sections 3 and 4 have presented baseline infor- 5.1 Future Development Projects + Housing developments (EKAM in Caracol: 750 mation that helps define the current interaction of The future development projects include on-going Housing units and 535 plots, Feed the Poor natural and urban activities and which can be used and planned development projects, and GIS layers projects along RN6 and fishing villages); and as a basis for determining potential suitability for were developed for the following, which are also + Anexisting quarry. land use. The projection of future land use needs shown in Figure 29: : based on economic development and growth, as Planned Development Projects: well as population increases are also needed to be Committed Projects: . us inputted into the modelling process, and this sec- * Sea Port expansion in Cap Heïtien; ° The Caracol industrial park (PIC); e Mining concessions and infrastructure pro- jects (thermic plants, water networks, solid waste treatment plants and wastewater treatment plants); and ‘ er e Projected housing (PIC resettlement called k Ve, EN Calles or Faias: 572 housing units and Food for 4 # NN AP the Poor in Terrier rouge: 242 housing plots). 5.2 Population and Demographics 1e w HU 0 : 5.21 Overview ? 2 ° Ë A detailed analysis of the population of the area of “4e 5 study was performed to understand the dynamics F. ë in place and better predict the impact of develop- É ment projects and investments and the future y à needs. This analysis focused on two population © 4 projections to be calculated for the year 2040: ñ tres mn. À e Slow Growth: based on the assumption of ee a mr me non-fulfillment of identified development pro- Section communale NS Schéma directeur de eau potable (DINEPA) jects as described in Section 5.1are delivered, I Zones urbaines © Centrales thermiques ñ EM Parc Industriel — Route principale and the planned projects do not progress); I UNH-RHC Route secondaire and 1 Projets de logement © Pot Concessions minières des points Récif corallien Figure 29 - Development projects ° Fast Growth: based on the assumption that all projects get implemented as described Map source: Variables built from geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), above. IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). == EMERGING = Lo 6 SUSTAINABLE K NORTHERN DEVELOPMENT CORRIDOR, HAITI 37 EX EH % ID B ERM [page 45] The population assessment has been undertaken, 2020, the region would reach a population of the existing projections remain valid, and where utilizing, and where appropriate supplementing, the 505,743, representing a growth rate of 1.3% over appropriate update them based upon best available AIA Study and the CIAT Strategic Plan. The Institut the period. Of these projections, the NDC area data including from IHSI, which is the only source of Haïtien de Statistique et d'Informatique (IHSI) was would be characterized by a higher growth rate and reasonably accurate information and projections in used as the primary data source (IHSI 2004, 2009a, represent a total of 67,381 of the total projected Haiti. 2009b, 2009c and 2012; IHSI and CELADE / ECLAC population for 2020. A growth rate of 1.2% was 2008), although it is also acknowledged that IHSI then applied to the 2020 to 2030 period resulting in 5.23 Demographic Trends has made only two censuses during the period 1980 a total population of 598,587 in 2030, of which the Existing demographic trends and baseline data ex- to 2003, and while a census is being planned for NDC would account for 79,753 of this total. Based ists for Haiti through a number of studies and as- 2014, data presented for recent years represents on historical growth rates, the AIA Study projects sessments as follows: projections made by IHSI. that the northern region would grow by 1.54 times : _. between 2009 and 2030, (i) National Demographic Trends (1950-2050). Data 522 AIA Population Projections : . : for national demographic trends comes from IHSI According to the AIA Study, and as illustrated in D eh ROM A resume chatte ro M CELADE / ECLAC 2008, n this paper, projections Table 8, a total of 387,339 people were living in the . ! . are calculated using a methodology and a model northern area of Haïti in 2009, and specifically in posed investments in the area progress. This pro developed by IHSI with the help of the CELADE and the NDC area, the population is 51,607 (shown as jection assumes a 2.15 growth rate reaching a the United Nations Fund for Population Activities the PIC node, comprising Limonade, Caracol, Terrier population of SC Mo ofthe grown (UNFPA). The methodology is based on data from Rouge and Trou-du-Nord). Was anticipate aroun the PIC and in Limons e, the General Census of Population and Housing con- Trou-du-Nord, Terrier Rouge and Fort-Liberté, and ducted in Haïti in 2003 and recent economic and The AIA Study also projects that if the same growth the NDC would contribute 136,172 of this total. demographic surveys, and provides national popu- rates of the recent past are extrapolated out to This ESCI study seeks to verify the extent to which lation projections (1950 to 2050) and average growth rates for both the urban and rural popula- Table 8 — Northern Region Population and Growth Projections (source AIA Study) tions. The key trends from this work predicts strong growth in both urban and rural areas, although fronuanon | 20 rates of urbanization are much higher than rural, [___Baselne | High Growth | Baseline | HighGrowth | and a gradual lowering of the growth rate. These [ville de Cap-aïtien | = EE D 2 ee ii = D national trends are reflected at the local level and : M ES DE 7 DE D [subrous forthe capmaenurennode | unes | mms | sens | amsn | ss | (i) Demographie trends at the municipal level (2000-2015). The data from IHSI analyses (IHSI 2009b) presents projections of the total, urban and rural population at the departmental and munici- pality level and the trends for key towns in the [ue seroreene man À on Tom À nan [us | th are shown in Figure 30 and Figure 31. This Fute ion ge Dee Tous zen rss Tue um] Hi Le the municipal level does not extend beyond 2015. Cora sense | sossres À sr2xss [0 sssser [° ssuxss | == EMERGING Lo NORTHERN DEVELOPMENT CORRIDOR, HAITI 38 initiative ERM [page 46] 1.74 90.00 1% 80.00 1.68 70.00 1.66 60.00 1.64 50.00 nue 40.00 160 30.00 1.58 ° N . N 20.00 as S où COS 10.00 PE ES LE © sul &® & NS S & « - S < Ss ® < PS Caracol Terrier Rouge Quartier Morin Limonade Trou du Nord m 2000-2005 m2005-2010 m2010-2015 m2000 m2005 m2010 m2015 Figure 30 - Average annual growth of total population Figure 31 - Urbanization rate in the municipalities of the study area Source: IHSI 2009b Source: IHSI 2009b Migration trends over time play an important role remaining 82.6% live in other towns of the Depart- (ii) Increasing urbanization - the data above can be in demographic change. The data indicates that on ments of the North and Northeast. used to calculate the rate of urbanization for key average, 10.76% of the population of the municipal- municipalities in the north, which reinforces the ities in the north are migrants, and specific to the Table 9 - Place of residence of PIC workers trend of towards concentration of the population in study area, migration rates of 6.8%, 9.25% and near me oo urban centers. 16.8% applied for Trou-du-Nord, Terrier Rouge and Caracol respectively. The results also show that the lcarecol | 47 | uns | 524 Migration Trends average percentage of migrants is higher in urban Caracol 347 11.34 Understanding migration patterns to urban centers than in rural areas, and that most migrants in the [Trou-du-Nord 77] 664 | 2170 | and rural areas allows a better prediction of future north are from local adjacent departments (6.86%) population trends and projections. Migrants are compared to departments from other regions of | Quartier Morin" | 54 | 1% | considered those persons whose place of birth is Haïti (2.61%) and abroad (0.81%). not the place of current residence, and the number . k and percentage of migrants is an important indica- Daily Commuting | tor of the attractiveness of a given locality. To better understand migration flows associated with project implementation, IDB collected data for Etrangers | se | 183 | Long-Term or Permanent Migrations the PIC workers was also analyzed (IDB 2012). This The General Census of Population and Housing data details the residency of employees, and can be 2003 (IHSI 2003) contains useful information on consider as a sample of the working population in - migration including place of birth and length of terms of daily commuting for employment. The residence, as well as an analysis of migration flows. data is summarized in Table 9 and shows that 19.4% of PIC workers live in Caracol and EKAM, and the == EMERGING = Lo NORTHERN DEVELOPMENT CORRIDOR, HAITI 39 initiative ERM [page 47] 5.2.5 Population Projections Slow Growth Scenario Table 10 - Projections of the population base - scenari- The results of the previously presented data and Given the absence of municipal-level demographics os of slow growth Roren was used as à baselie from which ro de- projections for the post 2015 period, an adjusted ESCI Projecti lop projections of the population of the NDC growth rate was calculated that takes into account AIA/ audy re through to he Year 2040. The method- those trends observed at the national level (see re 200 ere ology was as follows: Section 5.2.3 above). The adjusted rate of growth (Slow) ame | a has been applied to the 2012 population and pro- 1. Calculation of the annual growth rate from jected through to generate population estimates for 2000 to 2015 from the IHSI 2009b data intro- 2030 and 2040 in a scenario of non-Implementation Quartier de Grand duced in Section 5.2.3, noting that the urban of development projects, as shown in Table 10. population in these documents is not dis- High Growth Scenario asreaee Aude the different urban The AIA Study and the CIAT Strategic Plan estimate ° the demographic impacts of a similar scenario Nord ’ ’ d . 2. Adjustment of the annual growth rate (from 1 where development projects would be implement- above) for the period 2015-2040. An adjusted ed. These estimates assumed strong regional plan- growth rate was calculated by the following: ning controls to limit the expanding population and growth of the towns and urban centers in favor of * Calculate difference (variation) between the development of a new planned town in the the annual growth rate at the national Champin area (Nouvelle Caracol). The approach level: 2005-2015 vs. 2015-2040. adopted for this ESCI Study has assumed a more Ville de Quartier °__ Application of this variation in growth realistic approach of recognizing the limited gov- rate for 2010-2015 for the municipali- ernmental capacity to steer growth and influx to ties to generate an adjusted growth the desired areas, and assumes to some extent that Rouge É É É É rate. existing trends continue. The trends are defined by °_ Usingthe adjusted growth rate, calcu- pou rade already developed ses (er socti ici : glomeration) and commuting patterns, and have De to taken into account the current growth trends at the obtain estimates for 2030 and 2040. local level, as well as data on migration flows (long- + Populations in 2030 and 2040 are calcu- term and daily commuting) The adjusted rate of growth has been applied to the lated as the base population of the 2012 population and projected through to generate study with the assumption that no de- population estimates for 2030 and 2040 in a scenar- velopment projects are implemented io of non-implementation of development projects, (slow scenario). as shown in Table 11. 3. Disaggregation of the of the communal urban population agglomeration. For the urban mu- nicipalities, it provides the total urban popula- tion which in some cases includes several ur- ban areas in the municipality “#4 GIDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 40 as ERM [page 48] 5.3 Urban Area Needs scenario and the high-growth scenario, and is de- tailed in Section 7.3 where the capacity of the exist- The area needs for urban development can be es- ing townships to accommodate further growth and timated based on the previously presented popula- development is further explored tion projections and also the household numbers, average household size and housing density target. With respect to the population analysis and carrying . capacity it is currently assumed that household Table 11 - Projections of the population base - scenar- sizes stay constant from 2012 to 2040. It is recog- ios of High growth nized that there are different arguments related to this assumption and whether in fact average ESCI 2040 . | Current AIA/ CIAT Pattes households in the future may lower due to increas- 2012 Ten (High ing GDP, lower fertility rates, higher education 8 Growth) rates, increasing life expectancy, etc. Conversely, it pairs po de could be separately argued that “non-traditional” er de imonace jobs will be created which will take parents outside = = have the immediate family (mother, father and Quartier de Petite Anse 93,586 181,779 188,199 children) but also other family members (such as Ville de Trou-du- grand-parents, aunts etc. to help take care ofthe Nord children while the parents are at work. There are Ville de Caracol therefore many potential household size scenarios and given this situation, the projections and as- sumptions used in this ESCI Growth Study are con- Ville de Fort Liberte sidered appropriate and based upon the based Ville de Limonade available data from IHSI. minthe Furthermore, the criteria used to calculate the ca- Ville de Quartier : ss : Morin 4,125 7,136 23,743 pacity of existing townships to accommodate future " = housing demand was very conservative so that Ville de Terrier Rouge 13,876 32,260 46,285 there was flexibility and capacity in the planning Ville du Cap Haïtien exercise to accommodate potential variances in zanne cussed in Section 8.3. The number of households was estimated by divid- ing the projected populations by the average household size, which was taken from IHSI data (IHSI 2012). This was performed for the baseline E EMERGING NORTHERN DEVELOPMENT CORRIDOR, HAITI 41 Initiative ERM [page 49] 6. GEOSPATIAL MODEL development. The aim is to carry out this process by exercised a greater influence over the others. In a means of ‘comprehensively’ and ‘equitably’ consid- situation in which the scope of the work had been 6.1 Introduction ering all the elements that represent both pursuits, larger, the logical approach would have been to such that no one element ends up receiving, sup- survey the area of study and engage the different As stated in the Section 1, the goal of this study is to porting or ‘suffering’ the negative effects of anoth- communities so as to identify as to the specific determine the areas in the NDC that should be er. This is the basis of modern planning and is what forces that ‘pull harder’ than others and provide preferentially considered for future urban growth regional planners have called multi criteria decision this as the balancing criteria. and settlement, in such a way that this contributes analysis. The ‘art’ of planning has always resided on to a sustainable setting. The fundamental principle how that balance is demonstrated to have been As Joerin and Thériault (2001) demonstrate there is is that the areas in which future urban settlement reached in a particular situation. a vast array of ‘models’ or ‘approaches’ to the ques- occurs should be those that result from protecting tion of which elements and criteria to use in the or setting aside key areas for cultural, ecological In the past this type of analysis was carried out with running of the multicriteria tool. However, the aim and environmental reasons, including those that tools and references that had less precision that of ERM in this particular project was not to adhere are exposed to natural phenomena whose occur- today, but in both cases we are faced with the same to any model in particular, but simply produce the rence could not be mitigated, thus threating life. elements that represent the forces of economic land suitability map that would result from consid- development and nature conservation. Roads, ering all elements in equality of circumstance. By This section provides an explanation of the method schools, agglomerations, continue to be the ele- applying this approach we are ensuring the arrival and criteria used in this ESCI Study to define suita- ments that attract a settler, particularly for the way to at least the closest representation of the bal- ble areas for future growth. The approach builds in which they positively impact the family or firm's anced result that would be drawn with or without upon the AIA and the CIAT studies by: economic bottom line. And wetlands, woods, areas the modelling tool — hence representing what of bio-diversity are all elements that could also at- would likely be the picture if it had been drawn + Anumber of the elements used to judge the tract the settler but should be preserved or pro- following traditional mapping and planning tech- suitability of land for development were de- tected because of their natural value to society. niques. fined with greater precision, through the work on risk and vulnerability as well as the update Land suitability analysis through GIS modeling is a In determining which areas would be recommend- of a number of ‘layers’ provided by ESCI and well-known computational tool that helps provide able for human settlement in a given city and re- CIAT. planners with a more precise delimitation of differ- gion, the approach carried out uses a geospatial e The ‘combination’ of all the pieces of infor- ent elements as well as a clear delimitation of areas modeling process that simulates what is commonly mation (or ‘factors’) which is the basis of a that would result through a simulated combination known as attraction and restriction factors. The planner’s decision about where to allocate de- of all those elements under pre-determined combi- attraction factors represent those elements that velopment and where to protect, was not em- nation criteria. In other words, it is a computational will encourage and attract development by virtue of pirical but undertaken through a geo-spatial tool that rigorously applies multi criteria decision the services or value they provide. For example, modelling process. analysis. roads are a strong attracting factor for develop- ment given the access they provide, and similarly 6.2 The Geospatial Modelling Process Because of the scope of the study and the limited the presence of utilities due to the service they In determining which areas ought to be trans- amount of information in Haïti, in the Northern provide. Employment areas and social infrastruc- formed from their natural state to a developed Development Corridor study ERM made a funda- ture facilities can also be strong attractors. Con- state, communities, planners and decision makers mental, very conservative assumption: consider of versely, restriction factors are those elements that are faced with the challenge of reaching a balance equal value each element that was identified as an elther needs protection due to the Inherent value between the conservation of nature and economic attraction or a restriction, such that no one element they provide e.g. areas of high biodiversity, cultural = EMERGING = Lo s8 * SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 42 EX See F IDB ERM [page 50] sites or aquifer recharge areas, or are areas where ment to appropriate areas. In this context, growth ‘maximized’ so that their influence in the selection development should be actively discourage e.g. and new settlements tend to be steered towards of suitable land is very clear. The computational areas prone to natural hazards such as flooding. The already developed areas. process selects and creates a hierarchy of all the baseline information presented in Sections 3 and 4, areas with the highest degrees of attraction (from as well as the future growth and development con- This approach is widely and almost universally rec- the baseline components that have been classified siderations presented in Section 5, is used and cate- ognized as the way to build sustainable cities and as such e.g. roads, infrastructure, schools), but dis- gorized in terms of the attraction and/or restriction regions, for it fosters the following key results: cards, also hierarchically, all those areas affected or forces that they present. controlled by factors that should impede or prevent + It promotes densification (better known as settlement (e.g. locations prone to flood or earth- By combining these attraction and restriction fac- compact cities), which, in turn, makes the de- quake, or the presence rich agricultural soils or bio- tors through geospatial modelling, an understand- livery of water, sewer and other infrastructure, diversity). ing can be gained in terms of preferable and less social services, transport much less costly and preferable locations for future urban development. much more effective on a per capita basis. The modeling outputs are computer generated Combining the attraction and restriction factors is + _Ittends to increases the mixture of land uses maps of potential suitability based upon the applied undertaken through the lens of the different devel- per unit of area, which, in turn, increases the attraction and restriction factors. These outputs opment or growth scenarios being envisioned. likelihood of pedestrian or bicycle home-to- are only a guide, and must not be used as a defini- work travel, reduces travel distances that re- tive output for suitability. The outputs must then For example, in a location with limited controls and duce emissions and enhances health, and oth- undergo detailed analyses to then determine po- enforcement of urban growth, the influence of the ers. tentially suitable development areas based on a restrictions can be significantly diminished and range of factors such as the capacity of the areas to therefore urban growth, or ‘sprawl’, can occur in an 6.3 Modelling for the NDC absorb new development, current land uses, resi- ad-hoc and uncontrolled way which is responding : . dential density and the general characteristics of only to the attractions. This can produce a setting The modelling approach used builds off successful the locations. that tends to be marked by mid-density settlements similar studies undertaken by ERM and ESCI for the inside the urban areas, surrounded by expanding, metropalitan reglons of Cochabamba and Managua. The analysis also includes due consideration of the low to very-low density settlements. ‘Informal’ or However, recognizing the unique characteristics of expected population growth of the region and a ‘extra-legal’ settlements also occur on empty, un- the NDC area and Haiti more broadly, a particular calculation is made to derive the time horizon in protected public lands or in areas with little value focus has been placed on the speed and/or pace in which the suitable areas would be reaching capaci- because of their exposure to natural hazards or lack which development could occur in the area, recog- ty. This effort yields, among others, public policy of public utilities and /or social services. nizing both the recent growth due to investments recommendations with regards to the areas of the like the PIC, and the many proposed additional de- territory that should be considered urban, for- Conversely, in a scenario or strong planning con- velopment projects. expansion, and rural or peri-urban, and when, if at trols and enforcement, the restriction factors can . k all, should their boundaries and general terms of play a strong and influential role, limiting where in addition, the modelling approach has sought to development be modified. This work is further de- future urban growth occurs. Growth will still follow integrate from the beginning sustainable planning scribed and presented in Section 6.7. and respond to the attractions; however it will also considerations (also referred to as ‘smart or intel- respect and avoid designated restricted areas. This ligent” growth). This means that the geospatial The modeling process used involved the following latter scenario can be viewed as embracing sustain- modelling has sought at all times to recognize the stages and components: able planning principles in that it supports conser- importance of restrictions, and therefore respect vation and protection efforts and focuses develop- and avoid the defined restrictions. In modeling terms, this means that the restrictions have been “#5 GIDB L9 NORTHERN DEVELOPMENT CORRIDOR, HAITI 43 as ERM [page 51] 1. Using the baseline GIS information introduced 6.4 Restriction Factors Sub-Model in Sections 3, 4 and 5, attractions and re- strictions are determined. 6.4.1 Individual Restriction Factors 2. Assigning attraction/restriction scales to each The topics and elements considered restrictions for one of the variables based on analyses of each development in the NDC Study Area are illustrated layer according to the degree to which they in Figure 32, where there are twenty-two elements influence (or should influence) development. assessed. Having defined the elements and parame- For example, based on professional judgment, ters that will be considered as restriction factors, a road such as RNG is given 1 km of influence then they are analyzed and processed using a geo- on both sides, considering that there are no spatial model. The analysis of each element enables additional roads of similar status. Whereas an understanding of the degree with which it secondary roads are given not more than 500 should be considered as a restriction in the model, m of influence, considering as is the case with as well as ideas about the more suitable areas for the roads that serve Trou-du-Nord that after human settlement. This is explained further for that distance another road will probably ap- each key restriction factor in Table 12, and the indi- pear. vidual restriction maps are presented in detail in Appendix 7. 3. Developing individual sub-models, one each for attraction factors and restriction factors. These sub-models combine all of the relevant traction maps. This allows the “isolation” of the intensity of attraction from that of re- striction, making it possible to understand the ( srimeéismeninan ESS push and pull separately. 4. The restriction and attraction maps are then Come Re one ga EE combined to generate an overall model map, which shows the land suitability based on the mal restrictions exist ES Ce This model map is then translated into a general ete LS LE ae land use scenario, and used to define the general ES EA aa ES ES regional planning policies and actions on land use, and road and transport infrastructure. This work is presented in Section 7. Lo ae EE Figure 32 - Topics and elements considered to be restrictions for development == EMERGING = Lo #, SUSTAINABLE KZ NORTHERN DEVELOPMENT CORRIDOR, HAITI 44 LX gs &BIDB ERM [page 52] Table 12 - Summary of the main restriction factors LS Topography Hydrology ve F4 CR A S Oh. — mes € 3 de. | LS a S z | , ls WU A a un ou ee ue un CE = See ï me x Li L LL Based upon the hazard maps generated in Section 4.4, A restriction layer was prepared based upon gradient of Restrictions were placed around key water features principally for coastal and inland flooding and seismic, a | slopes, where no restrictions were applied for slopes un- including reservoirs and watersheds for principal and restriction layer was developed, recognizing the higher der 12%, moderate restriction were designated for slopes secondary rivers. A moderate level of restriction was risk areas identified along the coast due to coastal between 12-25%, stricter restriction between 25-50% and applied to the inferred Plaine du Nord/Massacre aqui- flooding and seismic, and the river corridors. a complete restriction for slopes over 50%. fer area. | Agricultural and Land Use Re SE res S— À À ve » PT Ÿ ÿ 2 x À »; { > x ED LS C0 à A nn Gas, = ou L Es Key strategic ecosystems considerations including the The cultural heritage variable implies that the coastal are- Using the agricultural soils classifications, areas of high Three Bays Marine Park, the coastal mangrove areas as are the most sensitive, as well as recognizing the ver- quality soils were identified for protection, as well as and the highlands ecosystem, which has been identified | nacular architecture areas and the sites identified in each considering other land use factors such as mining con- for reforestation and watershed protection, were used of the in or near the urban cores. A restrictions map of cessions. to build a restriction layer. these issues was constructed. Note: Further details of the restriction layers and the full maps are contained in Appendix 7 = EMERGING » KE 2 se GIDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 45 initiative ERM [page 53] 6.4.2 Synthesis Map of the Restrictions Sub- for the relative influence of the different restriction the most restricted areas and as the green becomes model factors was a ‘'conservative' or 'balanced' one, lighter, the less restricted the area becomes. . . which was to assign an equal influence to each ele- Trestmoeel conPnesthe eve of ana of ment. The sub-model results are observed spatially This map shows the strong restrictions along the Seb M nthe ea yzec. ee ifferent e eme he through a map of ten restriction levels. The map in coastal areas and also in the mountainous areas in scribed in the previous point were process together ; in he modell N Keaton The Ni roach en Figure 33 show the final restrictions levels included the south of the study area. In between are inter- 8 app ° PP in the suitability model. The darkest green shows mitted restricted areas, but also many areas of lim- Est à di RC Es & Bord de Mer de - EF” } Limonade L ra /4 Etracoh Éé di 4 Jacquezy, Pen ON “4 | sg A. ETS oi pr LA d'a D. LS ; ë EE | à ER FA; Se D a À LR TNA TR Moins restreint C1 Zone d'étude Récif Corallien Océan Atlantique = EH = = M Complètement restreint Figure 33 - Map of the restrictions sub-model: composite of maximum restrictions Map source: Results from the restrictions sub-model. Variables built from geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. = EMERGING = Lo ce te GIDB : NORTHERN DEVELOPMENT CORRIDOR, HAITI 46 initiative ERM [page 54] ited restriction. These clearly imply where would be more convenient or suitable for development to be Sous-modèle: Facteurs d'attractivité located. 6.5 Attractions Factors Sub-Model ed to existing agglomeration, access to roads, public facilities, social services and access to employment. Empreinte Routes Distribution nr Concessions This sub-model accounts for the attraction factors urbaine 2013 principales An Santé Industriel minières that will be considered by an individual, household or firm when locating within the area of study, as n . shown in Figure 34. While the importance of public I EST ÉAbIITÉ RUES Electricité Education GanmrereEl Autres lands and land value is noted, the absence of an CAES) SNS Ses ; y available information on this topic has resulted in its exclusion from the model. The selected attrac- Déchets solides Metititionnel tiveness factors are discussed as follows. Eaux usées 6.5.1 Individual Attraction Factors Figure 34 - Topics and elements considered to be attractions for development Having defined the elements and parameters that will be considered as attraction factors, then they are analyzed and processed using the geo-spatial application. The analysis of each element enables an understanding of the degree with which it should be considered in the model, as well as ideas about the more suitable areas for human settle- ment. This is explained further for each key attrac- tion factor in Table 13, and the individual attraction maps are presented in detail in Appendix 8. It should be noted that “public land” and “land pric- ing” typically are important attraction factors, how- ever this information was not available in a practical and usable form. == EMERGING = Lo et * SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 47 EX EH *IDB ERM [page 55] Table 13 - Summary of the main attraction factors Public Utilities ‘M Le re sr | à CR ALT 77 : } L ET ar EE 3 LR: ) 4 le ; ] PRE j pl 1 % F2 ! i + 7 ds à C7 vf LI ER NS KE tien All the cities, towns, hamlets and even smaller group- The modeling exercise included primary and secondary Public utilities included areas with access to water and ings of houses were considered as agglomerations, roads. The primary road was modelled to exercise recog- sewer systems, water points, and areas with electricity. which are an attraction feature for individuals and nize its greater influence as an attractor compared to sec- Each element, whether a line or a point, was modeled to families. ondary roads. exercise its influence Employment and Economic Activities nn ER Se = TER NS i VS Social services included schools, universities, hospitals, | Economic activities that were brought together in the and similar facilities, and the attractiveness also con- model included industries, mines, banks, agro industrial sidered the relevant township or community in which operations, retail centers, and others, recognizing both they serve local and regional attractiveness. Note: Further details of the resection layers and the full maps are contained in Appendix 8 = EMERGING » « NORTHERN DEVELOPMENT CORRIDOR, HAITI 48 initiative ERM [page 56] 6.5.2 Synthesis Map of the Attractiveness Sub- The different elements described in the previous the final sub-model results. model point were also processed together in the model- ling application. The approach taken was a 'con- Having defined th e restrictions and analyzed their 8 PP no pp : : 7 servative' or 'balanced' one, which was to assign an components, each variable was processed, analyzed equal influence to each element. Figure 35 shows and consolidated into the attractions sub-model. 4 18 acquezy} »., É 3 j _ 2 | “RS” ARE | ‘ Z = ë a 4 m ; L a a a ré RS * in : ra Te 5 ù PA 0 £ 25 dd 10 Moins attrayant [2] Zone d'étude Récif Corallien bn Océan Atlantique Plus attractif Figure 35 - Map of the attractions sub-model: composite of maximum attractiveness factors Map source: Results from the attractiveness sub-model. Variables built from geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. = EMERGING = Lo e## GIDB |: NORTHERN DEVELOPMENT CORRIDOR, HAITI 49 initiative. ERM [page 57] 6.6 Future Development Projects Sub- Model È - The future development projects sub-model in- d DR cludes on-going and planned development projects —— as described in Section 5.1. Four main layers were developed for existing or on-going development projects: Caracol industrial park (PIC), National Uni- " versity of Haiti Roi Henri Christophe Campus in £ Limonade (UHN-RHC), housing (EKAM in Caracol: ê 750 Housing units and 535 plots, feed the poor pro- El jects along RN6 and fishing villages) and an existing Ë quarry. Four additional layers were developed for planned + development projects: Sea Port expansion in Cap n Haïtien, mining concessions, infrastructure projects me À (thermic plants, water networks, solid waste treat- Moins attrayant D Zone d'auce ment plants and wastewater treatment plants) and BE Océan Atlantique projected housing (PIC resettlement called Calles or Faias): 572 housing units and food for the poor in Terrier rouge: 242 housing plots). Plus attractif Figure 36 - Attraction factors: Development Projects Appendix 9 shows the results of each one of the variables analyzed to compose the development Map source: Results from the attractiveness sub-model. Variables built from geographic information layers by projects sub-model. Each image shows the attrac- AlA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), tiveness levels of one or a group of development OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satel- ñ : so : lite imagery 1986, 2010 and 2013 projects according to their impact and geographic location. The sub-model results are observed spa- tially through a map of nine attractive-ness levels as the restrictions (in order to protect key ecosystems The model results of combining the three compo- seen in Figure 36. and resources) and maximizing the attractiveness nents are shown in Figure 37 with levels of attrac- factors recognizing the importance they plan in the tion and restriction that range from the most at- 6.7 Suitability Analysis development area. tractive areas for development, appropriate for To determine the areas that should to be consid . urbanization, to the most restricted for develop- k à Combining the three sub-models presented above: ment, and therefore adequate for the protection of ered for sustainable human settlement in the fu- restrictions, attractiveness factors and development natural areas. ture, a geo-spatial model has been used that simu- projects, the geo-spatial model combines all varia- lates the interaction of attraction and restriction bles relevant to identify areas potentially suitable As highlighted at the beginning of this section, this factors. This study's focus has utilized a proposed for urban development as well as rural and natural analysis provides an indication only of potential sustainable growth scenario through maximizing land use. land suitability, and cannot be used as a definitive = EMERGING = Lo 6 SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 50 EX EH % IDB ERM [page 58] result with respect to where development can and University of Limonade node is also beginning to cannot occur. create a pole of attraction, or where in the PIC area will new settlement likely occur. What this map is depicting is what the regional and urban planners, as well as decision makers, should The map also indicates how, if a new settlement bear in mind when establishing where to locate the pole was to be considered apart from the existing developments that would come in the future as townships, the above mentioned areas ought to be well as the services that the growing and migrating considered first and foremost. Interestingly, the population would demand. presence of a light green zone in between the two roads connecting Trou-du-Nord and the PIC tells us On one end, the red areas are those in which de- that these are amongst the more valuable agricul- velopment would be more attractive because they tural lands of the area, from the points of view ag- would be those areas in which a settling family or rological quality of soils, vegetation health as well business would have the greatest levels of access to as the fact that they are being utilized precisely for services, infrastructure, the economy of the ag- the uses for which they have a vocation. glomerations, transport services and others. And they would also be the places in which the natural Finally, the map is clear in defining an inverted ‘arc’ resources would be less affected. On the other end, of lands that could be useful for development be- the green areas would be those that would need tween Terrier Rouge, Grande Bassin, Perches and the greatest protection, for they are the ones in Ouanaminthe. These areas would be attractive for which most of the elements considered as re- development not only because of their agrological strictions are operating and in the greatest degree. and soil conditions (less good for agriculture) but also because they would be least exposed to natu- Consequently, the areas in between the dark green ral phenomena, would traverse almost no valuable and the dark red zones, would be the threshold in ecological asset areas, and therefore would provide which the planner and decision maker would have the ideal environment for developing roads, infra- to base in order to define what areas could be allo- structure and human settlement. cated for development and the degree to which they would affect natural resources or be affected by natural hazards. For example, when and if con- sidering that Terrier Rouge should expand because it does not have the area to provide housing for future generations inside its urban setting, it is clear from this map that this should be to the south of the urban area. The map is also key at indicating how the city of Limonade should be looking at planning the areas to the East and Southeast, where a major agglom- eration is beginning to appear, or how the EKAM - “#4 GIDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 51 as ERM [page 59] + Pa en < Dngtte F 25 # D res F LA ; NT :. à r EAST E VV D NET Ch Ve 2 1 PAPA CAE a | en EE Pr (À 5 P POS TER. F. (0 i PF = É vi Cr e RE … : 3 [ | 74 ; à 4 7 Le PME D: . LAS ei BEL ë (re PS # ue s® À Eh _ À à PE. < 4 Fa L Li 1 A e w Ÿ Lo # " ER: Cr, ie: di "A r É , »" We S à à NET DC CS PS PSS DO, 7 S Plus attractif C2] Zone d'étude Récif Corallien In Océan Atlantique = Plus restreint Figure 37 - Land Suitability Model Based on Attractions and Restrictions Map source: Results from the attractiveness sub-model. Variables built from geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. Le EMERGING Lo «ES GIDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 52 lnitiative ERM [page 60] 7. DEVELOPMENT OF A SUSTAINABLE 7.1 Land Suitability exploitation should continue inside the park, but should be regulated to ensure compatibility with GROWTH SCENARIO As presented in Section 6 and Figure 37, the results ecosystem recovery and conservations efforts. Sus- Based upon the results presented from the baseline ofthe geo-spatial moine provides insight on the tainable tourism focused on ecological conservation analysis and the geo-spatial modelling, the prefera- most attractive areas for development (appropriate and heritage preservation could become the main ble options for future development and growth of s coantaten) through she most rstitee for economic activity if the conservation efforts are the NDC can be explored, with the aim of providing evelopment (appropriate for the protection o k successful and water supply and energy networks insights on where and how should future human rural es In general, ne ras can be a are consolidated. settlement occur. Building in sustainability consid- & e " le de the The Fi on an erations, future development should seek to con- ï er ed ighlands that ought be mn Ne 7.1.2 Southern Highlands : : : edicated to ecosystem recovery; and the centra tribute to and make use of the economic opportuni- vi Y Y The highlands ecosystems to the south of the study ies of the region as a whole: not compromise op- plains, where most of the areas suitable for devel- ties o gl w ’ P | p ; area should be restored as part of a watershed ities in oth h cultural de- opment are concentrated. The hydric system, run- k . portunities in other areas Such as agricultural de à : : management priority. The highlands have potential | t i tal preservation: and not ning from the southern highlands to the Atlantic velopment, environmental preservation; ; : for watershed management, focused on reforesta- compromise the lives of people who could be ex- Ocean, works like a transversal element connecting L ? d à tion and the restoration of the hydric system, both osed to serious natural phenomena the three zones. ; Di: p pl ° crucial for the sustainability of the NDC. ; : di : This is illustrated in Figure 38, which shows an initial This section presents the key findings from this ificati N a Urban development in the highlands should be re- work and specifically covers the following: classification of the appropriateness of land use, | ' ' and these broad areas are discussed further below. stricted and agriculture should be restricted to fo- key findi d dati f th cus on ecosystem recovery. Charcoal and construc- . natal modellne orecentedin SceHon ë 7.1.1 Northern Coast Protection tion materials production should be replaced by Ë d land suit an . . . silviculture and sustainable forestry. However, the ase on and suitability; Marine and coastal ecosystems and cultural herit- success of these measures may depend on the ca- e the existing form and trends of human set- age along the coast should be protected, while de- pacity to generate and distribute energy to the tlement in the region (presented in Section velopment should be restricted. Bord de Mer de communities across the NDC, in order to reduce the 3.1), and the issue of densification, which can Limonade, En Bas Saline, Caracol, Jaquezy and demand of charcoal give insight on the capacity areas proposed Phaéton, and in general coastal areas in the north, for development and settlement; fall into the area where many of the restrictions e the main townships and hamlets’ capacity to overlap, such as the Three Bays Marine Park, accommodate new development; coastal flooding zones, cultural heritage areas, ma- ° approaches to settlement under the ‘slow’ rine and coastal strategic ecosystems and high agro- and ‘high’ population growth scenarios; logical soil classes. Urbanization in the northern ° details of the area of influence or ‘neighbor- coast should be restricted, ideally limited to the hood’ of the PIC and the potential for a new current footprints. development; and * a discussion of the Three Bays Marine Park Marine and coastal ecosystems and cultural herit- and the associated protection opportunities age areas should be protected as part of the Three and development restrictions. Bays Park management structure. Traditional eco- nomic activities, such as fishing, agriculture and salt > EMERGING TRS SUSTAINABLE IDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 53 initiative ERM [page 61] DE F D. el FN = \4 d y Ç & F; " 4 É A FN | L É q - # | + y LE » { à 0 à 4, . je f ee r. ” ë IG 1 » - s : & œ A + + J a + ne) pe ä on) “4 , & TE, er , à Ee. 1} F1 N 7 É TA } E ‘ - t 4 N 2. =, Ken ? CPR et L CES Î K Ê ÿ —_— BE; à pe + LE : L ATANS TO CD. “me ‘à 1 Zone d'étude En Forèt MM Développé à haute intensité ER Plan d'eau mm Développé à moyenne intensité Les zones humides boisées Développé faible intensité M Les zones humides émergeant Espace ouvert développé Marais salants ER Cuitivé Récif corallien Pâturage ' EM Océan Atlantique EN Marécages a vegetation herbacee Figure 38 — Optimized land use map Map source: Results from the attractiveness sub-model. Variables built from geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. = EMERGING » Lo US GIDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 54 initiative ERM [page 62] 7.13 The Central Plains hamilets, farms, ‘linear’ settlements along roads, However, as will also be seen in the township anal- Development should concentrate in the central and planned settlements. Section 3.1 has already yses that follow, there are clearly visible empty VI uIda conci l n : : : p Lu . provided more details on these patterns as part of lands throughout the townships. plains around existing settlements, balanced with : : : : ot the recovery of agriculture. Suitable areas for de the baseline discussion. Given the land suitability VV Icuiture. Sul - Jens a : ; t Y _ teinth f Li analysis, with the results supporting land protection Based on these characteristics, in this study has PAde nd Terror Rou : D La lune degree (for both natural resource protection and agricul- made two important planning assumptions: Trou-du-Nord and Cancel The best d'esofor r ture promotion), understanding the existing town- : b “ “ th of Li ° det dB pu . ships and urban areas, and their capacity to absorb + The first is that as societal effort would be betueen the u ban Le CÉLimonade and the UNA future demand for land based on the growth pro- undertaken to promote the densification of W u Lel [ nl n n a : . i = RHC and EKAM; southwest of the PIC, between Ter- jections in population becomes a very important sIreeay built plots of land: towever, ‘or " ’ ° , consideration. main in the conservative side, it has been as- rier Rouge and Gran Bassin and east of Trou-du- sumed that throughout the 25 year period of Nord. From social and economic perspectives, the most analysis, this effort will yield an additional The central plains are also the area best suited for convenient approach to providing settlement op- 20% homes in those areas. ieult P d lit logical soil portunities for the growing and the migrating popu- ° The secondis that the empty lands within the Le Ne 890 os ! Pare ne sors cal 8 lation of the NDC is to seek them in the human set- urban setting and in the areas immediately a ban Noos es Of the NDC ehoc id n V, tlements that already exist. Existing agglomerations adjacent would be developed to the highest “ d e % P mn isti ban f n ! tt provide many attractions such as access to greater possible density under the constraints posed crease ensities In ne exBstng un an ootprint to opportunities for exchanging goods, services and by the local market. optimize the use of land, balanced with urban and knowledge, as well as better public services of wa- rural nd coul à activities open space, public facil ter, sanitation, education and health, than those To this end, the Zorange Housing Expo has been ties and social services. that would be found in more rural areas. referenced. This was an effort undertaken following 714 The Hydric Syst the 2010 disaster by GOH in collaboration with the ne € Fyeric system 7.2.2 Current Density Patterns IDB, the Clinton Foundation and other organiza- The hydric system, including its riparian forests, The general pattern of urbanization in the main tions, with the goal of presenting good examples of should be restored to connect all areas from the existing urban areas is made by à single house in a housing that could be used in the re-construction. highlands to the coast. The hydric system, running 8 . DY 8 The project was not only intended at offering such : ! small parcel in which approximately 50% of the plot : from the southern highlands to the Atlantic Ocean is occupied by the dwelling. The number of parcels examples, but also to be configured as a new com- in the north, works like a transversal element con- that may be PA in one Éctare of land evicslly munity. Several multi-housing proposals were built necting the three zones, its protection is crucial. ranges between 50 and 60. except the case of that could be thought for the NDC (see Figure 39), Li E de. in which the d 17 0e dweli . and a few were dwellings raised from the ground, a 7.2 Densification Imonage, In WAIC È € density is Weng units solution that would be interesting for flood-prone per hectare. (This will be demonstrated in the de- : : . tailed | fthe diff tt hips). With areas. Interestingly, in the Zorange community, 7.2.1 Settlement Patterns and Growth alle ana yses of the different towns IPS : vrn an where the expo took place, there are multi-dwelling average 4.56 persons per household (as is the case : . La : Le . . k . . complexes that are inhabited, as can be clearly visi- While the majority of the Study Area can be consid- for Limonade), this equates to a population density ble from the lower center image ered rural, the urbanization and agglomeration between 228 and 456 inhabitants per hectare. ° patterns are important to understand in order to provide insights on future growth opportunities. These patterns are characterized by townships, > EMERGING «CE GIDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 55 initiative ERM [page 63] : TRE GC è ue t a, Mat | = ; + | !ssancrA A ” : , CR PNE ge Source: http://archrecord.construction.com/news/2012/01/Haiti-Communities-Expo-slideshow.asp?slide=42. Photo credits: Jenna M. McKnight = EMERGING » 7 0] Lo et GIDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 56 Initiative 7 ERM [page 64] According to the promoters of the Expo (see refer- 7.3.2 Approach These two analyses are further elaborated on below ence noted on photograph), each unit had to be To answer this question, two analyses were under- in Sections 7.3.3 and 7.3.4. developed within a budget that allowed either its taken for each one of the main townships in the : : : purchase by donor organizations focused in provid- 7.3.3 Net Land Available for Residential Devel- ing shelter to Haïitians such as USAID, or the pur- area as follows: opment chase by working families with capacity to consti- 1. Establishing the net available land for building In order to establish the availability of land for each tute a loan. Althaugh the Expo was successful in new residential developments inside the ur- town, the following analyses took place, which are F6 AIO MR CRUTe an see VAGCDNE ban setting, taking into account two factors: also illustrated in Figure 40: vented the Expo from becoming a living community. o The demand generated by envisioning that 1. Definition of a polygon that could be regarded k the activities taking place inside areas of as the ‘urbanized area’ or ‘urban perimeter’ These experiences demonstrate that not only tech- high risk would relocate to non-risk areas; of the township. This was traced following the nically but also financially there Isa possibility to and existing built form at the periphery of the develop multi-dwelling solutions in the country. À o A distribution of land uses reflective of town, including what could be defined as solution such as the one depicted in the lower left ‘good practice’ quantitative distribution of property lines of the built units, as defined D ou ahibe of the are eh ue result land uses. from remote imagery plus other sources of in- ’ formation such as Google Earth and the Open the density of 100 to 150 dwelling units per gross 2. The calculation, in terms of number of house- Street Map project. . ? hectare of land, or 150 — 200 dwellings per net hec- holds that could fit inside the urban setting. tare of land, that is factoring areas for roads, parks This, in turn, is comprised by two analyses: 2. Land use distribution, including areas occu- and community services. pied by roads, woods, public open spaces, in- . . © The capacity inside the net available land stitutional facilities, commercial facilities, in- Consequently, an aspiration of a density of about defined in the previous point. dustrial operations, and residential settings. 150-200 dwelling units per hectare of net existing © The capacity inside parcels of land that As a result a more precise measure could be urban areas would be a reasonable measure. currently exhibit residential buildings and achieved of the areas that could be consid- : _. . therefore have one or more households ered as ‘open’ or ‘not developed’ within the 7.3 Capacity of Existing Townships currently living inside the premises. urban setting. These areas could be consid- . ered as ‘prime developable land’, since they 7347 Introduction It was assumed that this will be the case be- are located inside the setting with the most A key question in the pursuit of sustainability in the cause of the predominant pattern of land complete system of public utilities, social ser- northern development corridor is whether the ex- subdivision and residential settlement inside vices, commercial and employment opportu- isting townships can afford, and to what extent, the urban areas and the fact that when a family nities that the region has to offer, that is, the increased population that organic growth and mi- grows into an additional household, the latter urban setting. gration are expected to bring to the region. This usually develops its living space through con- should also be understood in terms of the two pop- struction of additional rooms or levels of the 3. Land use break down of the areas that would ulation scenarios developed in this study, the ‘slow’ original house, or a new building on the origi- be affected by floods according to the risk and and ‘fast’ growth scenarios. nal parcel. vulnerability analyses presented in Section 4. This allows a determination of the demand in NORTHERN DEVELOPMENT CORRIDOR, HAITI 57 EH ERM [page 65] Urban perimeter. Defined by the farthest eo per ess including terms of these high risk areas that can then tablishing how many households could fit inside the be added to the growth demand from future net available residential areas and how many could Area defined as currently urbanized including growth, should a program for progressive re- do so inside areas currently exhibiting residential buildings and some garden / orchard activity. location be implemented for settlements in buildings. As discussed in Section 7.2.2, examples Areas defined as ‘empty’ showing independ- high-risk areas, of multi-dwelling and raised dwelling prototypes ent parcels with no construction (such as those developed at the Zorange Expo) that Area defined as expansion, selected on the 4. Land use outside flood area. This is the result a reasonable measure would be 175 dwelling units —© basis of proximity and on the results of the of subtracting from the gross areas calculated per hectare. This is the density that is applied to the modeling in the second step described above, the areas net available lands. defined as prone to flooding as determined in | FI Sections 4.4.3 and 4.4.4. This is very im- In order to determine the capacity of areas current- y / portant because it gives a measure of the ly exhibiting residential buildings, the following Z TZ ‘true’ area that is available for future devel- have been calculated for each township: // W opment by avoiding risk areas. 4 4 ° The average number of dwellings per hectare. > Ÿ 5. An approximation to the distribution of urban This was done by selecting two to three sam- Z 4 land uses within the ‘true’ area previously ple areas in each town, one near orin the & 8 \ mentioned, under a ‘good practice’ scenario. center, and the others in the periphery. The SZ LA C This considers that roads, wooded areas and exact number of buildings inside each sample CZ A V4 ZA public open space account for 45% of the ar- area was counted and divided into units of AA ea, solely institutional and solely commercial one hectare to arrive to the number of dwell- 4. LT FRS SI uses account for 10% of the area, industrial ings per hectare. CA UN for 3% of the area, and 42% for residential ac- e The number of dwellings per hectare for each SZ ee RQ tivities, which would be mixed with commer- sample area was added and divided by the 2 D: 2 DS cial and other complementary uses. These number of sample areas, to obtain the aver- ) 6 S & 44 approximate percentages are based on pro- age number of dwellings per hectare for the C2 s YO fessional judgment and experience from cities town. The average number of dwellings per fs, SE x IX in Latin America. hectare in the different townships ranges be- JA nes LA. Z : a. tween 60 and 105 dwelling units. Conse- F7, GA F 4 By applying these percentages to the true availa- quently, in calculating the number of houses ? S D LS f thin the urb _ Picreferrede for each town was applied. T° R ee 2 or wi “obI e euiUes dla which is referred to as ° From the same polygons used as sample to ER ê € AS €; net available residential areas. derermine ne average rumber of residential s GT, DS D, AAQS 4 . . . units, the total built surface as well as the av- 2 à HD, LS 2%: 734 nsinenne Capacity for Additional erage built surface was calculated. This al- Figure 40 - Detail of elements analyzed for each one of lowed the average foatprint size of one hec- the townships in the study area. The capacity of each township to hold additional tare of developed residential land and the av- residential units has been defined based upon es- erage open or unbuilt land within the same == EMERGING Lo NORTHERN DEVELOPMENT CORRIDOR, HAITI 58 initiative ERM [page 66] hectare to be calculated. These two were 7.4 Capacity in Trou-du-Nord public open space with 1% each. found to range between 3,000 and 6,000 m° in the case of the built footprint and 7,000 to 74.1 Total area and land use distribution 7.4.2 Area and land uses under high risk condi- 4,000 m° for the open or unbuilt area. . Le . tions Asillustrated in Figure 41 and supporting Table 15, A conservative stance on the amount of houses that the township of Trou-du-Nord comprises an urban However, 38 demonstrated in Table 14, 58.9 hec- exist and will emerge inside this setting was taken, area of 212.91 hectares of land. In terms of surface, tares, which equal 28% of the total urban area are assuming one household per building, and that only it is divided in order of size by empty lands compris- located inside the high risk flood area as defined by 20% additional houses would likely appear in the 25 ing 44% of the total, followed by residential areas the studies conducted as presented in Section 4.4.4. years through to 2040 covered by this study. For with 39%, roads with 8%, institutional services with These areas are comprised largely of unoccupied illustration purposes, in a township in which the 4%, wooded areas with 2% and commercial and lands with 48% of the total, residential areas with average number of homes per hectare is 60 inside the areas already built, it was assumed that there 2 are 60 families (1 family per home) and that in the \ , EL next 25 years a total of 12 new families and homes "3 LA (20% of 60) will likely be formed in the same parcels ÉD AN, of land in which these are located, for a total densi- 7 Pa ? d ty of 72 dwelling homes at the end of the period. CRE LA In addition to development within the existing built EN N D footprint described above, development potential is + CRN £ 72 also calculated and assessed (assuming 175 dwell- 0) CE NAN ing units per hectare) for the defined undeveloped Gr ce Ne areas, or prime developable land, that exists within 2 DPTR Ÿ, ANS the urban perimeter. This stance preferentially À “4 * RRQ focuses on the ‘net available residential areas’ de- ENS > 1€ UN fined by these two components. CALID MS AN 7.3.5 Expansion Areas 74 MARCZN Once the capacity of both these areas was reached, \ VE the number of families whose houses would have 2 - 4). . n—— to be developed on additional lands, the expansion Es : © Zone urbaines MN Habitation —— Route secondaire areas, was calculated using the measure of 175 Zone inondable/Forêt riveraine BB Commerciale —— Routes tertiaires dwelling units per hectare. The resulting hectares — Rivière principale MM Espace public of land per the calculation indicated in the previous = RNCS secs) — mois) point were distributed around the entire perimeter jetutonnes of the township, especially into areas defined as Plus restreint se - pas constructible more attractive for development per the results of the modeling process. Figure 41 - Main land uses identified in the township of Trou-du-Nord. == EMERGING Lo 6 SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 59 EX EU % IDB ERM [page 67] Table 14 - Total area and current land use distribution as the area in which future development could be referred to as ‘net available residential land’. in the township of Trou-du-Nord planned within. Based on the data in Table 14 for the total land outside the flood zone, the urban 7.4.4 Capacity to accommodate residential ithi -du- ï developments CATEGORY AREA (Ha) % area within Trou dur Nord would be able to receive velopi new developments inside the identified 66 ha of In order to determine the capacity of already built Rout. 17.82 8% land PACIy y Espace Boise 413 2% empty lands. areas, the two areas that appear illustrated in Fig- Pos ® ., n Espace Public 211 1% While the specific land use breakdown of a city is ure 42 (a cena city area ane us peripherat Institutionnel 893 4% closely associated to its economy, experience in one) Were use lo easure lhe num co units ve Commerciale 123 1% several cities in Latin America suggests a sustaina- E y co ? 18 28 pe Industriel 077 0% ble urbanization model for land distribution would hectare. Using the previously defined conservative Vide 94.71 44% comprise: 45% for roads, public spaces, natural or approach to the growth of areas that are currently Habitation 83.71 39% . a £. . De . built (20% of the growth of the average number of naturalized areas; 42% for residential uses (in which É ? TOTAL 21291 Il scal d'oth l + houses over a 25 year period), a density of 71 dwell- 100% Sma Scae commerce anc other comp'ementary ing units per hectare was derived, and when ap- uses are included); 10% for institutional and com- lied to the 57 hectares of residential land in which : : l l l in wi 45%, and some institutional facilities with 7% of the mercial uses and 3% for other, industry-related pi j , uses this could take place, a capacity of 4,024 homes was area. ° obtained. In addition, for the 28 hectares of net In an ideal scenario none of these areas would be If this model was applied in Trou-du-Nord, the 66 résident ane defined re 7e sppying occupied by buildings, since they are classified as ha that constitute the ‘gross’ available land should € reasonable censity o wering units per k . . Me ‘ : hectare generates a further 4,868 dwelling units. high risk, in which mitigation measures would likely be broken into the land uses and areas that appear Based on these two factors, the total capacity of not be sufficient to protect life. in Table 14 under ‘availability for development’. ? 'ese tw ’ a! capacity Consequently, not more than 28 hectares ought to dwelling units inside the urban setting would be 74.3 Available land be destined for residential developments. This is 8,892. This leaves 72% of the urban setting or 153 hectares Table 15 - Trou-du-Nord - Urban land uses inside and outside the high risk flood areas, and ‘true’ available land. URBAN LAND USE INSIDE FLOOD AREA URBAN LAND USE OUTSIDE FLOOD AREA AVAILABILITY FOR DEVELOPMENT (NET VIDE) CATEGORY AREA (Ha) % CATEGORY AREA (Ha) % CATEGORY PRACTICE AREA (Ha) RELOCATED NET AREA Routes 0.00 0% Routes 17.82 11.65% Routes 15% 9.93 0 9.93 Espace Boise 0.00 0% Espace Boise 413 2.70% Espace Boise 15% 9.93 0 9.93 Espace Public 021 0% Espace Public 1.90 124% Espace Public 15% 9.93 0.04 9.89 Institutionnel 4.18 7% Institutionnel 4.74 3.10% Institutionnel 5% 331 0.82 249 Commerciale 025 0% Commerciale 0.98 0.64% Commerciale 5% 3.31 0.53 2.78 Industriel 0.01 0% Industriel 026 0.17% Industriel 3% 1.99 0.29 170 Vide 28.48 48% Vide 66.23 43.29% Vide 0% 0.00 0 0.00 À | Habitation 26.77 45% Habitation 5694 37.21% Habitation 42% 27.82 1122 16.60 TOTAL 59.91 TOTAL 153.00 TOTAL 6623 129 28% 72% == EMERGING «CE GIDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 60 initiative ERM [page 68] g 77 . = —— . Table 16 - Trou du Nord - Capacity for residential 7 AU 4 IN 7-2 ai ; \ Ca D 7/47 TN 7 7 D. VE? developments inside the urban setting and areas 7, \ LD «<&,? Xe NICE ECS required for expansion in the 2040 fast growth 7” LA Vase à 4 Ÿ KZ scenario y = SZ De A À (E,//N D p T 72 pe. NS TD PONT : 4 GR 4 OT PULLS | — NPA ET LAPS Se 7 L'ù No. of homes per ha in sample area 1 ei C La 2 ? NN P* T: se PNY No. of homes per ha in sample area 2 57 WU LE d Up à w 4 Z 4 LA IN ñ ÿ 7 Average number of homes per ha 5 TD Ke AS ) 4° 7/4 à >, “Reasonable" density in areas currently PO QC PLAN # A 2 é 24 AU occupied by residential use (current homes 71 2 \ ns Cp “ W 7”, / 4 e, LD APN AVAILABLE OPEN LANDS ZA s 6 2 # À oo C < . * » SLT se 4 à NS. er OL Capacity in # of home: developed IX > NE VARIE NC }. C4 Des 0.085 0.0 lands at reasonable enty (275) n 4868 © Zone urbaines DM institutionnel PEVEROPED ZONES QUTRIDE RISK ARERS Plus attractif Vide = Route secondaire Area (Ha) 57 Plus restreint D, ce Es — Capacity in # of homes on already developed Æ Habitation ones d'analyse de la densité de logement zones at reasonable density (71) 4024 EM Commerciale I Espace public TOTAL SUPPLY IN URBAN AREA 8892 Figure 42 - Trou du Nord - Areas selected for calculating the building density. Demand from relocation in Ha 27 . Demand from relocation in # of homes (at Considering that by 2040, in the fast growth scenar- Table 16 summarizes these calculations for Trou- average density) #7 io Trou-du-Nord would reach a total demand of du-Nord. Total households by 2040 in the fast growth 17083 18,660 housing units including existing households scenano" plus those that might be considered for relocation, Based on these calculations, and as illustrated in TOTAL DEMAND BY 2040 18660 and that 8,892 of those could be housed inside the Figure 43, a series of areas for future urban devel- current urban area, a total of 9,768 households opment around the perimeter of the city have been Total supply in urban area 8892 would have to be located in expansion areas. At the identified, which amount to 78 hectares. This pro- Total housing needs in expansion areas 9768 reasonable density of 175 dwelling units per hec- vides an extra buffer against the 56 hectares identi- tare, Trou-du-Nord would have to incorporate an fied above, and acknowledges that the area could Area required for expansion in ha 56 additional 56 hectares of land to its perimeter. be developed with varying density parameters that + According to ERM population study and projections == EMERGING + Lo «Es GIDB | NORTHERN DEVELOPMENT CORRIDOR, HAITI 61 initiative ERM [page 69] reduced the dwelling units per hectare parameter. These correspond with the areas surrounding the town and those that extend along the main roads that were classified as more attractive for develop- ment based on the modelling process presented in Section 6.7. 74.5 Next Step: Developing an Urban Design Vision Having established the areas in Trou-du-Nord that should be considered for future urbanization inside and outside the urban setting, the next step would be to develop an urban design vision. This should be based on applying, for both the ‘true’ available land and the expansion areas, the distribution of land uses that is proposed under the ‘good practice” model discussed previously. Because ofits size and location farthest from areas of high risk, the quadrant comprised by polygon D in Figure 43 should be thought of as the one where the main institutional, commercial and recreational activities of the township ought to be developed. Mixed with reasonable density residential devel- opments, this could provide the balance that the township is requiring in terms of public open spaces and other elements of a quality civic life. == EMERGING Lo NORTHERN DEVELOPMENT CORRIDOR, HAITI 62 initiative ERM [page 70] @ 2 Trou-du-Nord J' x SSN > (de ü Ô Q 0,8 22 De CN À C1 Zone urbaines MN Habitation C1 Poilygones d'extension Zone inondable/Forêt riveraine B Commerciale —— Route secondaire — Rivière principale Mn Espaces public —— Routes tertiaires [A | 13.81] _—— Rivière secondaire MM Espace boise BB | 7] Lpitcslau: | fers utonnel CE Vide = Plus restreint mu Vide - pas constructible 1 78.5 Figure 43 — Trou-du-Nord - Current land uses, areas for densification within the urban setting and proposed expansion areas. = EMERGING = Lo ct # GIDB L NORTHERN DEVELOPMENT CORRIDOR, HAITI 63 EU ERM [page 71] 7.5 Capacity in Limonade space with 1% each. Only one industrial operation the total, residential areas with 38%, roads with 9%, was identified, with less than 1%. wooded areas with 5% and some institutional, 7.5.1 Total area and land use distribution commercial and industrial facilities adding up to 6% : . 7.5.2 Area and land uses under high risk condi- of the area. As mentioned at the beginning of this Asillustrated in Figure 44 and supporting Table 1 - F : : : . k _. tions Chapter, in an ideal scenario none of these areas 4, the township of Limonade comprises an urban ld b ied by buildi ince th | area of 155 hectares of land. In terms of surface, it As demonstrated in Table 18, 29.49 hectares, which Jfied bre 2 ich INg5, SICe they are c'as- is divided in order of size by empty lands comprising equal 21% of the total urban area are located inside Would lets se son ARS 55% of the total, followed by residential areas with the high risk flood area as defined by the studies y p ° 26%, roads with 10%, wooded areas with 6%, public conducted as presented in Section 4.4.4. These are spaces with 2%, and commercial and institutional comprised largely by unoccupied areas with 42% of Table 17 - Total area and current land use distri bution in the township of Limonade , " 7 T al NZ - = ; CATEGORY AREA(Ha) % É à ST Z à ; ! ñ Espace Boise 913 6% à L 2 RU t me 4 Espace Public 273 2% CSS _ LONGER Ne en, Hoor U Institutionnel 2.06 1% 4 6 TAN PALENS ZZS Ze Industriel 029 0% # LOZERE CU Vide 84.36 55% on ce ZSS CAE IN 2 is ne s Habitation 39.46 26% Mic LANG BIRTE SNL EEE ste TOTAL 154.66 is 8 AÉRLIGE NE F 7.5.3 Available land LA 7 DÉS LIN : FFÉNRES The above leaves 81% of the urban setting or 125 4 222 CRU A : * Se Le hectares as the area in which future development 4 CSN ' L be “n could to be planned and fostered. As also indicated NS ) à... 0 04 03 CO À in Table 18, of the total land outside the flood zone LS | ‘ = 1 (125 ha), 72.07 ha are empty (58%), followed by 28 2 Zone urbaine EM Commerciale —— Route secondaire ï ji D : Zone inondable/Forêt riveraine M Espace public —— Route tertiare ha ofresidential areas (23%), 13 ha of areas occu —— Rivière MM Espace boise pied by roads (10%) 7.55 areas occupied by wooded FA airaour _ lue areas (6%), 3 ha of public open space (2%), and Phi 72 Me pe other land uses representing less than 2% of the 2 Habitation RAS Éhcbie area. Figure 44 - Main land uses identified in the township of Limonade Based on these data, the urban area of Limonade would be able to receive new developments inside the 72 ha of empty lands previously mentioned. > EMERGING = Lo CHE GIDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 64 initiative ERM [page 72] Table 18 - Limonade - Urban land uses inside and outside the high risk flood areas, and ‘true’ available land. URBAN LAND USE INSIDE FLOOD AREA URBAN LAND USE OUTSIDE FLOOD AREA AVAILABILITY FOR DEVELOPMENT (NET VIDE) CATEGORY AREA (Ha) % CATEGORY AREA (Ha) % CATEGORY RATE AREA (Ha) RELOCATED NET AREA Routes 2.72 9% Routes 1254 10.02% Routes 15% 10.81 Le 10.81 Espace Boise 158 5% Espace Boise 755 6.03% Espace Boise 15% 10.81 o 1081 Espace Public 0.04 0% Espace Public 2.69 215% Espace Public 15% 10.81 0.04 10.77 Institutionnel 082 3% Institutionnel 124 0.99% Institutionnel 5% 3.60 082 2.78 Commerciale 0.53 2% Commerciale 0.84 0.67% [Commerciale 5% 3.60 053 3.07 Industriel 029 1% Industriel 0.00 0.00% Industriel 3% 2.16 0.29 187 Vide 12.29 42% Vide 72.07 57.58% Vide 0% 0.00 o 0.00! | | Habitation 1122 38% Habitation 28.24 22.56% Habitation 42% 30.27 11.22 19.05 TOTAL 2949 TOTAL 125.17 TOTAL 72.07 129 19% 81% Ifthe same model discussed in Section 7.4 is applied in Limonade, the 72 ha that constitutes the ‘gross’ . : = . ES : d ZZNTENT / | ES À LE EE KR available land should be broken into the land uses _ N 7 d "= ; RE he —4 ZT | K RSS and areas that also appear in Table 18. Consequent- 22 S 2e LT, RE 2 Le Le SKK ly, not more than 30 hectares should be destined CC 7 2 1 LS CE," LE ÆRR RC ! , É , D = 22 © Dh LE JS F for residential developments. # À | nn CL; & L x 7 74 A NS RUE 7.5.4 cas to accommodate residential » 2 4 Œ rz à < LE A K N 7 evelopments La y D > pe Cale KR KR L 4 Zz y L 220 cn À & KR Z In order to determine the capacity of already built LV ' > 2 LA 27 7 RRQ SK 7 areas, the two areas that appear illustrated in Fig Z CZ 17 Z TA 77 ZT ZT = ESS NI QU ure 45 (a central city area and a more peripheral 2 C (lg ee Fa à # a à © ER RSS RKÇQ St P / A NS II ÉSNIKRKKRK & one) were measured, finding an average of 105 _ y | # ne. LD: ENXIT RS RÙ >. units per hectare. Applying the conservative density CN ” ATX ? PR 4 g- = ER 7 7" (20% increase), a density of 126 dwelling units per EN A > 1% Vd ” A, NRRSS SSZ 07 LS J D o re » op 22 2. RRKKKiKkKkKIET hectare is derived, and applying this to the 28 hec- Z 22 À Z @ 7 É_ s ZA RQ NS tares of residential land area, a future capacity of PEN LE ê ZZ ENS C//; 3,558 homes is obtained. In addition, applying the ET JDN GT Cr 3 Lin PR SN $ SSÈ RENNES 4 175 dwelling units per hectare to the 30 hectares 7, D VLC A CDD GE LS SK 4 Ne) 5 N ==] ZA : PNTE : Zone urbaine Vide identified in Section 7.5.3, a further 5,297 homes Zone inondable/Forêt riveraine 4 Vide - pas constructible can be accommodated, giving a total capacity of —— Hu 1 Zones d'analyse de la densité de logement dwelling units inside the urban setting of 8,855. = ré ce de _ pere publ In striel Considering that by 2040, in the fast growth scenar- M institutionnel io Limonade would reach a total demand of 17,307 Figure 45 - Limonade - Areas selected for calculating the building density housing units including existing households plus = EMERGING = Lo 24 GIDB à NORTHERN DEVELOPMENT CORRIDOR, HAITI 65 initiative ERM [page 73] those that might be considered for relocation, and A major feature of Limonade visible in the figure is Table 19 - Limonade - Capacity for residential devel- that 8,855 of those could be housed inside the cur- the presence of healthy, wooded areas surrounding opments inside the urban setting and areas required rent urban area, a total of 8,452 households would the township. Considering that these could be ma- for expansion in the 2040 fast growth scenario have to be located in expansion areas. At the rea- jor green assets in the area, the recommendable sonable density of 175 dwelling units per hectare, action would be to declare them as public space, BASEUNE « Limonade would have to incorporate an additional integrating the western side of the city to this sys- 48 hectares of land to its perimeter. Table 18 sum- tem. This should be thought of as the area where »honmprismenienss "= marizes these calculations for Limonade. the main institutional, commercial and recreational fon int amiens z = activities of the township ought to be developed. Based on these calculations, and asillustrated in Mixed with reasonable density residential devel- rage rarmber of homes per Da 10 Figure 46, a series of areas for future urban devel- opments, this could provide not only the balance “Ressonable"densRy in areas aarentiy opment around the perimeter of the city have been that the township is requiring in terms of public va DFI ns Dene ” identified. These correspond with the areas sur- open spaces but also become a major element of Mt: rounding the town and those that extend along the attraction to this township for new migrants or set- PR ORNE 1e main roads that were classified as more attractive tlers. AVAILABLE OPEN LANDS for development based on the modelling process presented in Section 6.7. (a) 3 city in & of homes on non developed 5297 Because of the presence of a significant process of ds at reascmable density (175) settlement on the east side of Limonade, as well as DEVELOPED ZONES OUTSIDE RISK AREAS a valuable wooded area in the same area, a planned expansion area that incorporates controls over the fa) La wooded area (polygon ‘C’ on the map) should be y In 3 of homes en arendy devsoped implemented. Coupled with providing expansion msn An areas in all directions, this would yield a larger ex- TOTAL CAPACITY 1 URBAN AREA ses pansion area, which has been estimated at 111 n Te en dé hectares. Demand tros relocation in # of homes (at : : pré 178 7.5.5 Next step: developing an urban design “3 by 20401 the fut vision o* 2» Having established the areas in Limonade that boushelts 17307 ought to be destined for future urbanization inside dd à mn = and outside the urban setting, the next step would be to develop an urban design vision. This vision fenal housing needs in expansion areas 8452 should be based on applying, for both the ‘gross’ e di jé available land and the expansion areas, the distribu- tion of land uses that is proposed under the ‘good ss nn uns sd pou practice’ model discussed previously. == EMERGING es GIDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 66 initiative ERM [page 74] — " = — : i da | . : : i Limonade < Te ? ts … RE _ Pa y 4 4 S &. | C2 È ARR (F) L ! AT L CS eg : INT, 1 TRES 4 55 L | nn, : LES 4. 2 = RS V4 UE > 4 NES eur DS = 4, 4 f x % 2) } a MSN RS 3 ] F SS L de : N # ‘ * sl cos y L 2 Q 1 nf l DS : 3 , | ! me Km C1 Zone urbaine M Commerciale — Route principale Zone inondable/Forêt riveraine BN Espace public — Route secondaire D —— Rivière MM Espace boise —— Route tertiare EE MRETE heal ie EM institutionnel [D | 2101 Vide ; = . . [E | 5.12] el sm Vide - pas constructible CE HET L__] Polygones d'extension Figure 46 - Limonade - Current land uses, areas for densification within the urban setting and proposed expansion areas. = EMERGING = Lo “8 SIDB |: NORTHERN DEVELOPMENT CORRIDOR, HAITI 67 Mate ERM [page 75] 7.6 Capacity in Terrier Rouge 7.6.2 Area and land uses under high risk condi- Table 20 - Total area and current land use distribution tions in the township of Terrier Rouge 761 Total area and land use distribution As demonstrated in Table 21, just 2.56 hectares, Asillustrated in Figure 47 - Main land uses identi- which equal 2% of the total urban area are located CATEGORY AREA (Ha) % fied in the township of Terrier RougeFigure 47 and inside the high risk flood area as defined by the Routes 1535 12% accompanying Table 20, the township of Terrier studies conducted as presented in Section 4.4.4. Espace Boise 086 1% Rouge comprises an urban area of 130 hectares of The small flood zone is comprised by 1.12 ha of Espace Public 401 3% land. In terms of surface, it is divided in order of size empty lands accounting for 44%, 0.78 ha of residen- Institutionnel 5.63 4% by empty lands comprising 45% of the total, fol- tial activity areas representing 31%, and some insti- Coneniie 055 0% lowed by residential areas with 34%, roads with tutional facilities with 13% of the area. With these Industriel 0.00 0% 12%, institutional services with 4%, public open indicators, Terrier Rouge is clearly the least exposed Vide | 5885 45% : so : : : Habitation 44.74 34% spaces with 3% and very limited commercial areas. township to flooding. TOTAL 12998 100% | LL 14 au D A Relocating, if selected as a measure, and imple- S FA ei re * menting adaptation measures inside the areas . 1 7 nr Et would also be the least complex task to undertake <, ZE < a in this town, compared to the other towns in the ENITIAA | PARC \ area of study. VAT L'AILE EE 7.6.3 Available land SE Al, nPZT 1 oÿ Ca 2 NT Fr A “ZX The factors expressed above leave 98% of the urban E--- ZA Al SL AN Si = setting or 127 hectares as the area in which future T4 nr ÉLIRE Le | PCR development could be planned. As indicated in the t- PP D 0 — "4 =? - k og: Le Fate \T : Table 21, of the total land outside the flood zone à GEL L= N + ” (127 ha), 58 are empty lands (45%), followed by 44 ne 7 \ SR ha of residential areas (35%), 15 ha of roads (12%), # PRE À | institutional areas occupying 5 ha (4%), public open Gr per SP Se \ 2777 É | space occupying 4 ha (3%) and other land uses rep- %. ë o dois 03 de resenting not more than 2% of the total area. 1 Zone urbaines EM Commerciale —— Route secondaire Based on these data, the urban area of Terrier SE — Enies Sr == Dpiesintio Les Rouge would be able to receive new developments Plus attractif Ææ institutionnel inside the 58 ha of empty lands previously men- Vide . d = Plus restreint he ru tioned. Figure 47 - Main land uses identified in the township of Terrier Rouge > EMERGING = Lo NORTHERN DEVELOPMENT CORRIDOR, HAITI 68 initiative ERM [page 76] Table 21 - Terrier Rouge - Urban land uses inside and outside the high risk flood areas, and ‘true’ available land. URBAN LAND USE INSIDE FLOOO AREA URBAN LAND USE OUTSIDE FLOOD AREA AVAILABIUTY FOR DEVELOPMENT (NET VIDE] CATEGORY AREA Ma) % CATEGORY AREA (Hal “ CATEGORY are AREA (Me) RELOCATED NET AREA Routes oz 15% Routes 1502 11798 Routes 15% 26 © 866 Espace Bose 000 Lu] Espece Botse os 067% Espace Bose 15% s66 o 8.66 space Public 0.00 o Espace Public 401 s15% Espace Public 15% s6 004 8.62 mation 032 13% mattutionnæl sx 417% RTE 5% 29 o#2 207 omenerciaie ou L,7 Commercisie 05 022% ciate sx 2s9 L'ET] 236 industriel 0.00 ox msn Cr 0.00% Industriel 3% 173 oz 143 Vise 1x sx Vide s273 45.50% vide o% 000 ° 0.00 Habastion 078 31% Habtanion 4396 34.50% Habitation 42% 2425 12 13.03 TOTAL 25 TOTAL 12743 TOTAL s773 129 2% 28% Applying the same approach for Terrier Rouge as : SE = = a CASE u TR 2 F7 re has been applied to Limonade and Trou-du-Nord, 27 L. * er — CZ. 4 A ) the 58 ha that constitutes the ‘gross’ available land Z 77 L. pÿ] 4 T° PA 1] dl f ON should be broken into the land uses and areas that É _ 17 LA NZ) EN (D appear in the availability for development part of Le 2 ZA [2 22 « é 2 | k LZ Table 21. This results in 24 hectares identified for ’ Ne en us D LE er L+ 4 1Z 22 eu 7] AT rt AY f a x residential developments. D LEZ RE: VD FE / 22; D 7.6.4 Capacity to accommodate residential Ÿ + Z: ve 2 a | 2 C7 . developments LA \ 1/7 7 > 1 | 27 As in the previous townships, the determined the D ee > À de ['\ AT NE _ capacity of Terrier Rouge to accommodate new BA S LP ns. aus PT D. households was on the basis of a conservative ap- AX Ce là L / Ë 22, > proach to the growth (20% growth), together with \ D À — 4 _ . 2 NE jo! LH the application of a ‘reasonable’ density of 175 ‘à b = ï AA | [ ï i 7 £ _….S à 4 0,07 014 dwelling units per hectare to the net available lands DAT = = us pa mm Within the urban setting. The 24 net hectares that C1 Zone urbaines M Commerciale == Rues de la ville could be utilized for residential activities would Zone inondable/Forêt riveraine DM Espace boise [= Zones d'analyse de la densité de logement —— Rivière M Espace public therefore fit a total of 4,243 dwelling units. Plus attractif _ Menton EM Vide - pas constructible Plus restreint In order to determine the capacity of already built x Habitation ep areas, the two areas that appear illustrated in Fig- : . : : ure 48 (a central city area and a more peripheral Figure 48 - Terrier Rouge - Areas selected for calculating the building density one) were used to measure the number of dwell- ings they contain, finding an average 53 units per = EMERGING = Lo 8 GIDB L NORTHERN DEVELOPMENT CORRIDOR, HAITI 69 A initiative ERM [page 77] hectare. Consequently, by applying the conserva- Table 22 - Terrier Rouge - Capacity for residen- Terrier Rouge is not significantly exposed to floods, tive density of 64 dwelling units per hectare to the tial developments inside the urban setting and andit is also a surrounded by very important natu- 44 hectares of residential land uses in which this areas required for expansion in the 2040 fast ral wooded areas that should be protected. could take place, a capacity of 2,796 homes is ob- growth scenario tained. Based on these two factors, the total capac- Because of this factor, the expansion of the town- ity of dwelling units inside the urban setting would BASEUNE ° ship should be thought of more towards the south be 7,039. No. afhames per ha in sample ares 1 2 (polvgons € and D in Figure 49), which is also where the suitability analysis yielded the areas that should Considering that by 2040, in the fast growth scenar- No. of homes per ha in sample area 2 54 be developed. These should be the areas where the io Terrier Rouge would reach a total demand of main institutional, commercial and recreational 11,190 housing units including existing households Average mumber of homes per ha 5 activities of the township ought to be developed. plus those that would have to be relocated, and lReasonable”density in areas currentiy This proposition departs from the AIA Study's pro- that 7,039 of those could be built inside the current ns 2e (eurent homes “ posal to create a by-pass of RN6 towards the north urban area, a total of 4,152 households would have -essonsbie density in non developed lanés Ds of the township. That said, the south of Terrier to be located in expansion areas. At the reasonable Rouge and its connection with Grande Bassin is an density of 175 dwelling units per hectare, Terrier AVAILABLE OPEN LANDS area that should be looked for sustainable future Rouge would have to incorporate an additional 24 development. hectares of land to its perimeter. Table Area (Ha) 24 22summarizes these calculations for Limonade Capacity in # of homez on non-developed 2e lands at reszonable density (175) Based on these calculations, and asillustrated in DEVELOPED ZONES OUTSIDE RISK AREAS Figure 49, a series of areas for future urban devel- opment around the perimeter of the city have been Ares (Ha) #4 traced. These correspond with the areas surround- Capacity in # of homez on alrezdy developed 3706 ing the town and those that extend along the main zonez at reasonable density (64) roads that were classified as more attractive for TOTAL CAPACITY IN URBAN AREA 7039 development based on the modelling process pre- sented in Section 6.7. Demand from relaeation in Ha 1 Demand from relocation in # of homez {at « 7.6.5 Next step: urban design vision average densèy} Total households by 2040 in the fast growth Having established the areas in Terrier Rouge that scenario" ne ought to be destined for future urbanization inside Total houzeholdz by 2040 plu relocated 11190 and outside the urban setting, the next step would ES be to develop an urban design vision. This vision [Fotslsepeir in min ares 7039 should be based on applying, for both the ‘gross’ Total housing needs in expansion ares as available land and the expansion areas, the distribu- tion of land uses that is proposed under the ‘good Ares required for expansion in ha 24 practice’ model discussed previously. + According to ERM population Audy and projections == EMERGING Lo sl SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 70 EX EU % IDB ERM [page 78] =. E FF YO Mi “@ À _. . . A hi , / Terrier Rouge PES MES 2e PES RÉSEAU = Le TEE. ira $ red: - CZ 2 Ne, 5} - " rLA BALA 2 Le _ 8 A ' LT SR. "(172 Gr edl > XLR à \ — CAE Ci LA TTL LA RS; LES 2 LADA E esse DL IZZZLE GA À ts Zz LL DLL >>K + * pue N # % 00,15 0,3 0,6 Pa S Ÿ == — Km À C1 Zone urbaines EM Commerciale — Route principale Zone inondable/Forêt riveraine DM Espace boise —— Route secondaire —— Rivière BR Espace public —— Routes tertiaires [A | 0.92] Plus attractif EM institutionnel B | 215 en Vide : EN Vide - pas constructible = Plus restreint 2 : [D | 25.74 EM Habitation Polygones d'extension Taa 37.53 Figure 49 - Terrier Rouge - Current land uses, areas for densification within the urban setting and proposed expansion areas = EMERGING = Lo cs GIDB |: NORTHERN DEVELOPMENT CORRIDOR, HAITI 71 Initiative ERM [page 79] 7.7. Capacity in Bord de Mer de Limonade areas with 20%, roads with 7%, institutional areas high risk flood area as defined by the studies con- with 3%, and commercial and public open space ducted as presented in Section 4.4.4. 7.7.1 Total area and land use distribution with 1% each. . on . In an ideal scenario none of these areas would be As illustrated in Figure 50 and supporting Table 23, 7.7.2 Area and land uses under high risk condi- occupied by buildings, since they are classified as the township of Bord de Mer de Limonade compris- tions high risk, where mitigation effects could have lim- cut ee dk ai . re Ni mener The situation in Bord de Mer is critical because, as ited impacts. However, it is very untikely that a pro- surface, I ; dde in order of ire by ue et demonstrated in Table 24, 47 hectares, which e ual gram to relocate 30% of the township will happen, comprising 51% of the total, followed by residential , } Wni q and it is not possible to accommodate the current 90% of the total urban area are located inside the : : and future population demands in the 10% area à. Table 23 - Total area and current land use distribution >, in the township of Bord de Mer de Limonade « 4 URBAN LAND USE IN BDM de Limonade NE 77 . 4 à CATEGORY AREA (Ha) % Ÿ SNA \ > Routes 3.72 7% F LÉO 6 à /) | Espace Public 0.66 1% à < Ne Wie” > Pr À NN | Institutionnel 141 3% 10% A RRQ NUE hd RSS, Commerciale 0.36 1% Fo p ARNAT 10% D Sp 214 // Industriel 0.00 0% À SNA ÿ = x Vide 26.77 51% D D VAN E Habitation 1957 37% “ 4 that is not in situation of high risk. re l As a consequence, the options for responses would C1 Zone urbaines EM hnstitutionnel — Route tertiaire be as follows: [SS Zone inondable élevé Vide D Océan Atlantique 7° Zone inondable très élevé DM Vide Zone inondable élevé Plus attractif 1. Barr any expansion of the township and im- Zone inondable très élevé plement policies and incentives for new set- = Habitation - : EM Commerciale - Plus restreint tlers to seek location elsewhere. BI Espace public —— Route secondaire Figure 50 - Main land uses identified in the township of Bord de Mer de Limonade USSR EMERGING= oi Lo NORTHERN DEVELOPMENT CORRIDOR, HAITI 72 1nitiative ERM [page 80] Table 24 - Bord de Mer de Limonade - Urban land uses inside and outside the high risk flood areas, and ‘true’ available land. URBAN LAND USE INSIDE FLOOD AREA URBAN LAND USE OUTSIDE FLOOD AREA AVAILABILITY FOR DEVELOPMENT (NET VIDE) CATEGORY AREA (Ha) % CATEGORY AREA (Ha) % CATEGORY RRCUCE AREA (Ha) RELOCATED NET AREA È 337 7% Ë 0.34 6.29% È 15% 0.52 0 0.52 Espace Boise 0.00 0% Espace Boise 0.00 0.00% Espace Boise 15% 0.52 0 052 Espace Public 0.66 1% Espace Public 0.00 0.00% Espace Public 15% 0.52 0.04 0.48 Institutionnel 140 3% Institutionnel 0.01 0.14% Institutionnel 5% 0.17 0.82 0.65 Commerciale 023 0% Commerciale 0.13 231% Commerciale 5% 0.17 0.53 0.36 Industriel 0.00 0% Industriel 0.00 0.00% Industriel 3% 010 0.29 019 Vide 23.30 50% Vide 347 63.31% Vide 0% 0.00 ( 0.00 | | Habitation 18.04 38% Habitation 153 27.96% Habitation 42% 1.46 1122 -9.76 TOTAL 47.01 TOTAL 5.48 TOTAL 347 129 90% 10% 2. Consider the entire township for the applica- gram for changing the existing housing stock 7: tion of the ‘good practice’ land use distribu- to these kinds of models. À tion model, which would be representative of a sustainable community in terms of econom- 4. Develop multi-family housing projects that : ic activities. This would entail, as demonstrat- maximize the density that solutions such as ed in Table 24, doubling the surface occupied the one depicted in Figure 50 could afford. by roads, creating a public space realm of Based on the analyses of the Zorange pilotis- open spaces and wooded areas of 15.6 hec- supported structure and the EKAM project, tares in a place that just has 0.66 ha of this the density in this case that could be reached kind of space, doubling the area for institu- would be 80 homes per hectare of net area. tional activities, and setting aside 4 hectares of land for a commercial operation. As a re- 5. Design and develop incentives for future sult, in terms of residential surface the town- households to settle outside the township, on Figure 51 - À pilotis - supported house developed for ship could only allocate 22 ha. areas that are less exposed to natural phe- the Zorange Housing Expo nomena. 3. Develop a housing solution that adapts as maximum as possible to the potential haz- 7.7.3 Capacity to accommodate residential ards, in this case coastal floods and hurri- developments 196 units less than the 1,644 homes that would be canes. This could be possible, through housing According to the analyses presented, and is demon- expected to have arrived in the township by 2040. models such as the one depicted in Figure 50. strated in Table 25, in Bord de Mer de Limonade This is illustrated in Table 23 These should be implemented in the 22 ha there is a net availability of land for residential uses that would be allocated for residential uses. of approximately 23 hectares. This would yield a This should be coupled with an intense pro- total 1,840 homes, which would be > EMERGING = Lo ce SUSTAINABLE S NORTHERN DEVELOPMENT CORRIDOR, HAITI 73 EX EH % IDB ERM [page 81] Table 25 - Bord de Mer de Limonade - Capacity for the total, followed by residential areas with 34%, de Mer, but still a concern, with 49% of the area of residential developments inside the urban setting roads with 10%, institutional areas and public open the township at risk of flooding (see Table 28). space with 3% each, wooded areas with 1%. There Again, in an ideal scenario none of these areas is only a minimal commercial activity near the town would be occupied by buildings. However, it is un- center. likely that a program to relocate 49% of the town- ship will happen. Additionally, here it is also not Net land available for 33 7.8.2 Area and land uses under high risk condi- possible to accommodate the current and future residential land uses tions population demandés in the half of the township Multifamily density that could 80l The situation in Caracol is not as critical as in Bord that Is not at risk of flooding. Furthermore, as will be reached Total homes 1840 Total households by 2040 in the > Le 1644| 2 : fast growth scenario SA TR | &] … ue: | Remaining capacity by 2040 196 ne JE mi bé \c = AT | + According to ERM's population projections nu a. { a ai ÿl IV 4 7.74 Next step: developing an urban design $ . _ AIT FE LL vision ù an y : AU ÈS à 72 QD ANT EE == BEM 7 A re-thinking of the future of this township should \.. Er VAE CCE ai OT TS > CURE Le Re be considered. However, this should be coupled fs n7 ZE FD 2 a Le A + with a program aimed at offering alternative set- 4 7 2 2 FE F7 tlement areas near the employment centers where } à CA 27/22 12 * its inhabitants work. For in the long term, a process | , ! 1L J &s of growth reflective of the current construction ! ! ! À È habits and trends would expose this township and | ï ë (1 F its population to even more severe situations of Ë k\ A H risk. i \ “, ee mm ti C1 Zone urbaines Plus attractif Vide 7.8 Capacity in Caracol Zone inondable EM Vide - pas constructible Zone inondable élevé Plus restreint 7 Route secondaire 7.8.1 Total area and land use distribution M Zone inondable très élevé EM Commerciale == Routes tertiaires | Zone inondable/Forêt riveraine MM Espace public I Océan Atlantique Asillustrated in Figure 52 and supporting Table 27, — Rivière principale MM Espace boise : : —— Rivière secondaire mars the township of Caracol comprises an urban area of Em Habitation M institutionnel 43 hectares of land. In terms of surface, it is divided FI 52- Main land identified in the t hi L ' in order of size by empty areas comprising 48% of igure 52 - Main land uses identified in the township of Caraco Be EMERGING = K «Es GIDB & NORTHERN DEVELOPMENT CORRIDOR, HAITI 74 Initiative ERM [page 82] Table 27 - Total area and current land use distribu- Table 26 - Bord de Mer de Limonade - Distribution of tion in the township of Caracol demand issue, and would comprise: urban land uses under a ‘good practice” scenario. 1. Seek to prevent any expansion of the town Routes s81 15%] CATEGORY AREA (Ha) % ship and implement policies and incentives for pue Boise 551 5% space Public 6.51 15% Routes 444 10% new settlers to seek locations elsewhere. The institutionnel 217 x Espace Boise 0.44 1% proximity of the PIC and the fact that this is Commerciele 217 x Espace Public 127 3% attracting workers from the entire region, Industriel 130 3% Institutionnel 140 3% calling for a local housing solution to be seri- vide 000 0% Commerciale 0.14 0% ously thought, is an opportunity in this case. Habitation 18.23 42% Industriel 0.00 0% TOTAL 4339 Vide 20.36 48% 2. Consider the entire township for the applica- ss pu 34% tion ofthe ‘good practice’ land use distribu- for other productive activities. As a result, in 100% tion model, which would be representative of terms of residential surface the township a sustainable community in terms of econom- could only allocate 18 ha. be shown in this section, due to the expected ic activities. With respect to the existing dis- | | growth in the demand for housing in Caracol in the tribution of land uses (see Table 27), this 3. Develop a housing solution that adapts as fast population growth scenario, the township will would result (as shown in Table 26) in adding maximum as possible to the potential haz- not have the necessary area to cover that demand if two more hectares of roads, creating a public ards, in this case coastal floods and hurri- restricted areas are respected and adhered to. space realm of open spaces and wooded are- canes. These should be implemented in the as of 13 hectares in a place that just has 2 ha 18 ha that would be allocated for residential As a result, the responses would have to the same of this kind of space, doubling the area for in- uses. This should be coupled with an intense as those proposed in Bord de Mer de Limonade, stitutional and commercial activities, and set- program for changing the existing housing with an additional sense of urgency because of the ting aside 1 or 2 additional hectares of land stock to these kinds of models. Table 28 - Caracol - Urban land uses inside and outside the high risk flood areas, and ‘true’ available land. URBAN LAND USE INSIDE FLOOD AREA URBAN LAND USE OUTSIDE FLOOD AREA AVAILABILITY FOR DEVELOPMENT (NET VIDE) GO0D CATEGORY AREA (Ha) % CATEGORY AREA (Ha) % CATEGORY practice AREA (Ha) RELOCATED NET AREA Routes 161 7% Routes 284 13.20% Routes 15% 122 0 122 Espace Boise 0.00 0% Espace Boise 0.44 205% Espace Boise 15% 1.22 0 1.22 Espace Public 0.06 0% Espace Public 120 5.61% Espace Public 15% 122 004 118 Institutionnel 026 1% Institutionnel 114 531% Institutionnel 5% 0.41 0.82 041 Commerciale 005 0% Commerciale 010 045% Commerciale 5% 0.41 0.53 0.12 Industriel 0.00 0% Industriel 0.00 0.00% Industriel 3% 0.24 029 0.05 Vide 1285 59% Vide 811 37.76% Vide 0% 0.00 0 0.00 L.. … Habitation 7.09 32% Habitation 7.64 35.60% Habitation 42% 341 1122 -7.81 TOTAL 2192 TOTAL 2147 TOTAL 8.11 129 51% 4% == EMERGING Lo 6 SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 75 EX EH % IDB ERM [page 83] 4. Develop multi-family housing projects that Table 29 - Caracol - Capacity for residential devel- maximize the density. Based on the analyses, opments inside the urban setting the density in this case that could reached would be 80 homes per hectare of net area. 5. Design and develop incentives for future Net land available for households to settle outside the township, on residential land uses 21 areas that are less exposed to natural phe- Muitifamily density that could nomena. be reached 80 7.8.3 Capacity to accommodate residential Total homes 1680 developments According to the analyses performed, and as Total housenolds by 2040 inthe 2528 fast growth scenario demonstrated in Table 27, in Caracol there is a net availability of land for residential uses of approxi- Remaining capacity by 2040 _ga8 mately 21 hectares. This would yield a total 1,680 homes, which would be 848 houses shy of the Additional area required to number of homes that would be expected to have house the 2040 housing 106 arrived in the township by 2040. Figure demand at multifamily density This means that, should the fast population growth this town in the next 25 years. However, in this case scenario become a reality, the township of Caracol this should be coupled with a program aimed at would not be in capacity to receive the population offering alternative settlement in the vicinity of the that is projected to arrive. Assuming that the densi- PIC. fication program was successful, an additional 11 hectares of land would have to be found some- For in the long term, a process of growth reflective where adjacent to Caracol (see Table 29). In a of the current construction habits and trends would township of 43 hectares, this represents 25% of the expose this township and its population to even surface. However, this would be contrary to the more severe situations of risk, and continue to ex- policies that are being sought through the imple- pand the township into areas that are now protect- mentation of the Three Bays National Park, which ed by Law, threating the sustainability of the Three include zero land expansion of the urbanized areas. Bays National Park. 7.8.4 Next steps As expressed in the previous paragraphs, a re-think of the future of this township should be carried out. An aggressive program to change the construction habits and to implement the density increases dis- cussed would contribute to ease the pressures of NORTHERN DEVELOPMENT CORRIDOR, HAITI 76 EH ERM [page 84] 7.9 Capacity in Jacquezy center. Table 30 - Total area and current land use distribu- : Le : tion in the township of Jacquezy Asillustrated in Figure 53 and supporting Table 30, 7.9.1 Area and land uses under high risk condi- the township of Jacquezy comprises an urban area tions CATEGORY AREA (Ha) % of 18 hectares of land. In terms of surface, it is di- . k vided in order of size by empty residential areas Asa coastal town, Jacquezy is less at risk from | Routes 444 10% comprising 45% of the total, followed by empty flooding, with only 8% of the area of the township Espace Boise 0.44 1% lands with 34%, roads with 6%, public open space affected (see Table 31). Characteristics like the type Espace Public 127 3% with 3% and institutional areas with 2%. There is of dwelling developed, the street and public open Institutionnel 140 3% only minimal commercial activity near the town spaces scarcity are the same as Caracol and Bord de Commerciale 0.14 0% Industriel 0.00 0% Vide 20.96 48% NA Habitation 14.73 34% TOTAL 43.39 | y 100% E? Z 22 A Mer de Limonade. With the creation of the Three 44 2 Bays National Park, the ideal situation would be Z MA that this township did not expand, accommodating ZAËKAE 2 rowth within its boundaries. The recommended 1 ZA Z responses would be the same as those proposed for À 17 Bord de Mer de Limonade and Caracol including À . Z # avoiding expansion of the TR CA 1. Barr any expansion of the township and im- DE 47 plement policies and incentives for new set- 7 2) tlers to seek location elsewhere. The proximi- 77 11 ty of eastern exit of the Caracol Industrial Park | Z Gi and the fact that this is attracting workers = 4 457 _ from the entire region is also opportunity in Fe en | 4 7 this case. Furthermore, the Eastern entrance T. TT; J cu | to the park could be planned more as the Davies a iition front’ entrance with the Western one being Zone inondable M Commerciale the ‘service’ entrance. Together with the im- 1 Zone inondable élevé MM Espace public pressive landscape views at this side of the — EE très élevé + much PIC, this could create an attraction for higher D Vide - pas constructible end residential solutions that will be required F = Route secondaire ï Plus restreint — Routes brtialtes at some point. Figure 53 - Main land uses identified in the township of Jacquezy Be EMERGING = Lo NORTHERN DEVELOPMENT CORRIDOR, HAITI 77 Initiative ERM [page 85] Table 31 - Jacquezy - Urban land uses inside and outside the high risk flood areas, and ‘true’ available land. URBAN LAND USE INSIDE FLOOD AREA URBAN LAND USE OUTSIDE FLOOD AREA AVAILABILITY FOR DEVELOPMENT (NET VIDE) CATEGORY AREA(Ha) % CATEGORY AREA (Ha) % CATEGORY PRIE AREA (Ha) RELOCATED NET AREA 'È 161 7% E 284 13.20% E 15% 122 0 122 Espace Boise 0.00 0% Espace Boise 0.44 2.05% Espace Boise 15% 122 0 122 Espace Public 0.06 0% Espace Public 120 5.61% Espace Public 15% 122 0.04 118 Institutionnel 026 1% Institutionnel 114 5.31% Institutionnel 5% 0.41 0.82 041 Commerciale 0.05 0% Commerciale 0.10 045% [Commerciale 5% 0.41 0.53 0.12 Industriel 0.00 0% Industriel 0.00 0.00% Industriel 3% 0.24 029 -0.05 Vide 1285 59% Vide 8.11 37.76% Vide 0% 0.00 0 0.00 À | Habitation 7.09 32% Habitation 7.64 35.60% Habitation 42% 341 1122 -781 TOTAL 2192 TOTAL 2147 TOTAL 811 129 51% 49% 2. Consider the entire township for the applica- 4. Develop multi-family housing projects that that today is expected to arrive under the assump- tion of the ‘good practice’ land use distribu- maximize the density, which analysis suggests tion that the township does not expand. Assuming tion model, which would be representative of could reach would be 80 homes per hectare that the densification program was successful, an a sustainable community in terms of econom- of net area. additional 5 hectares of land would have to be ic activities. With respect to the existing dis- found somewhere adjacent to Jacquezy. In a town- tribution of land uses (see Table 30), this 5. Design and develop incentives for future ship of 18 hectares, this represents 27% of the sur- would entail, as demonstrated in Table 32, households to settle outside the township, on face. However, as mentioned above this would be doubling the area covered by roads, creating areas that are less exposed to natural phe- contrary to the policies that are being sought with a public space realm of open spaces and nomena. the implementation of the Three Bays National wooded areas of 5 hectares in a place that just has half a hectare this kind of space, tri- 7.9.2 Capacity to accommodate residential Table 32 - Jacquezy - Distribution of urban land uses pling the area for institutional and commer- developments under a ‘good practice’ scenario cial activities, and setting aside 1 or 2 addi- According to the analyses presented, and as tional hectares of land for other productive demonstrated in Table 31, in Jacquezy there is a net Routes 271 15% activities. As a result, in terms of residential availability of land for residential uses of approxi- Espace Boise 271 15% surface the township could only allocate 8 ha. mately 8 hectares. This would yield a total 640 Espace Puolic 271 15% : : homes, which would be 418 houses shy of the Insütutionnel 0.20 5% ue 2 in the other cases, develop a housing solu- number of homes that would be expected to have Commerciale 0.30 5% tion that adapts as maximum as possible to arrived in the township by 2040. Industriel 0.54 3% the potential hazards, in this case coastal Vide U.uu U%0 floods and hurricanes. These should be im- This means that, should the fast population growth Hakitation 759 42% plemented in the 8 ha that would be allocated scenario become a reality, the township of Jacquezy TOTAL 18.07 for residential uses as per the application of would not be in capacity to receive the population the model discussed previously. = EMERGING Lo 6 SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 78 EX EU hd IDB ERM [page 86] Park, which include zero land expansion of the ur- study. As illustrated in Table 33, a total of 174 hec- the different townships sooner rather than later. banized areas. tares would have to be incorporated into the urban The elements defined in this study - size and loca- setting and developed with infrastructure to ac- tion of expansion areas, distribution of land uses, 7.9.3 Next steps commodate the demand for housing in the fast should serve as suitable reference points for tracing A re-think of the future of this township should be growth scenario. This allocation would more than detailed urban designs for the future situations. carried out. An aggressive program to change the adequately cover the needs of the slow growth . construction habits and to increment the density scenario (note that this does not include Jacquezy, 7.10 The Neighborhood of the Caracol In- could contribute to ease the pressures of this town as dissggregated baseline and projected infar- dustrial Park in the next 25 years. However, in this case this SD. for this township is not provided by the An important consideration for this study, and as should be coupled with a program aimed at offering ° raised by both the AIA Study and the CIAT Strategic alternative settlement in the vicinities of the PIC, Under both scenarios, the existing urban areas are Plan, is whether or not a ‘new city’ is required to vie Fr less SM ave to daalep insufficient and the allocated expansion areas support future growth of the NDC, as well as high quality settlement including residences for would be needed: The existing urbanized setting nat stoutbe Petedinihe negrhorhood middle and upper middle income workers on the would not be sufficient to accommodate the x include th n d " h of # k PIC, students attending the University of Limonade pected demand of either scenario, especially if the nee ei e presence Ana grow ni “ne PIC ne and others d future of these townships was planned following University of Limonade, large agro-industrial opera- ° the model of sustainable distribution of land uses tions that are beginning or expected to appear, and 7.9.4 “Fast versus ‘slow’ growth scenarios. that has been proposed in this study, in which resi- the aim of reducing as a minimum the urban growth dential development covers only 42% of the availa- of the coastal townships with the establishment of The previous analyses were conducted using the ble areas outside the risk zones. the Three Bays National Park. housing demand that the townships would face in the ‘fast growth’ scenario as defined in the popula- Consequently, it is appropriate and urgent to adopt From a purely quantitative point of view, and as tion and socioeconomic section (Section 5.2) of this an integrated policy of re-design and planning of presented in the above sections, it would be possi- ble to accommodate inside the area townships and Table 33 - Total areas of expansion that would be required to accommodate the housing demand expected by 2040 adjacent lands, the growing population at least until in the fast’ population growth scenario 2040. However, in Haiti there are two particular phenomena that must also be considered: one is the ‘volatility” or ‘ease’ with which land uses change. This is largely associated with the absence of a secure, formal, system of property rights and the fact that the State is the owner of large exten- [Limonade |" 22289 | 83 [15616 | 6673 |" 38 | sions of land that are usually designated for differ- [Terrier Rouge [11149 [24 [7811 [3338 [19 ent uses based on the priorities and projects of the [BOM de Limonage [1644] 0 [1152 [492] 0 | government in turn. For example this is how set- (Caracol "|" 2528 | 106 À" 1771 | 757 | 4 | tlements like that of EKAM or those promoted by [Total [| 54693 [1736 [38318 [16375 [91 | aid organizations appear suddenly, in locations ** Calculated using a density of 175 dwellings per net hectare of land. work. NORTHERN DEVELOPMENT CORRIDOR, HAITI 79 as ERM [page 87] The second phenomenon is also the volatility with A. The settlement that appears along RN6 before the D. More scattered settlements along road that enters which population movements happen in the coun- municipal limits of Limonade and Trou-du-Nord. Trou-du-Nord at the Wie Gr Lierels; and try. This is largely because the poverty conditions in B. The settlement located at the very intersection be- Eh The prototownshipitratiisbeginnineltoppeamatthe : : : " tween RNG6 and the roads that enters to Caracol. Jesus junction, where the old road between Trou-du- which many live creates in families the need to go C. Settlement along same road, after PIC entrance. Nord and Terrier Rouge connects back to RN6. where employment is, regardless of this being tem- porary or permanent. So when a development such as the PIC appears, it is likely that many families will L'4 pursue settling in the area. In consequence, the ‘neighborhood” of the PIC and ic ESS "1 , VV Caracol, the University of Limonade and EKAM, Cox pr | 4 f {2 ’ Trou du Nord, and Terrier Rouge, referred to here- # r nb ] & l # { inafter as the ‘diamond’, is an area that should to a « Le. / - v d! be considered for implementation of a planned ls & / > x a process of human settlement. D LA / LÉ | Both the AIA and the CIAT call for the creation of b } nc , S . 3 such new center. This is proposed in the area of pe 7: #4 XSS à : + known as Champin, located to the south of the PIC TPE K {iN \. k : / se and along the road that connects Caracol to Trou- 5H 4 = ? à ÿ À Ÿ, || du-Nord. While the CIAT document does not specif- * LP 4 sŸ NN @ 4 ? ically indicate a location, the AIA does propose that ro 2 #\ SSSR 1 à F / it be built as a complementary and extended devel- jt > Sy” À D. & #- * opment to the community located half way be- € Ÿ {& NÉ y cg: : à tween Champin and Trou-du-Nord. 2 LR Ÿ ab # P ÿ y Sy However, based on the analyses carried out for this Éd 1 PME Ÿ r 52 À-.mmmms study, the following factors should be taken into = Parc des Trois Baies EX Zone de agriculture durable Routes tertiaires consideration when defining the location of a po- Zone urbaines £=>] Plantation de banana : Zone inondable/Forêt riveraine Zone de peuplement très contrôlés tential new town: Bâtiments __ Polygones d'extension En Plus attractif — Rivière principale i Existing developments = —_ tre Pre M Pius restreint pen ca Les ct Asillustrated in Figure 54, the area exhibits numer- ous settlements such as the one where the AIA Figure 54 - The ‘neighborhood’ of the Caracol Industrial Park proposes its new town (see lettered boxes on map below): Legend: The land use classifications ‘Plantation de banana', ‘Zone de peuplement très contrôlés’ and ‘Zone de agriculture durable were established during conversations senior consultants from the Ministry of Tourism and the IDB. Map source: From geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. == EMERGING = Lo et * SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 80 EX EH hd IDB ERM [page 88] In this context, there are numerous alternatives to consider for future settlements; the convenience of 7 RN6, as well as proximity to the PIC and the Univer- Z sity of Limonade, are considered to be the strongest drivers for growth. L RE > ii. New plantations in the neighborhood ERA om GE 1) ASS , a Alsoillustrated in Figure 54 is a banana plantation Pa dtin. RNA Le : of 1,000 that hectares is being set up in two major Don /’ D AT Y parcels: the first one is North of the RN6 junction 4 À! - LS LEE that leads to the Champin area passed the Universi- À fe à, + { ty of Limonade, right in front of the EKAM project; L | te D “ \ à le the second one is East of the same junction extend- } RARE S NS à } 4 £ 4 ing approximately 1.5 km along the access road to | F— à SSS Ÿ/ # 4 < # Trou-du-Nord. This plantation will likely be demand- F D ? RS NS f ) FN k ing labor from area residents, in which case the sh é PS S Ÿ Age 4 ñ vicinity to the EKAM area as well as that of the set- ei; À > Ÿ + INK Si Fee: Ve RE £ tlement at the Champing crossing would have a K FANS CZ É F4 » «Ai major advantage. à Q° à NS Ésptn”, 6 $ $ 2 3 a ii. The results of the geo-spatial modeling en PEN PT % } LS Ed à — Finally, in this setting it is recommended that the Ex ere Baies Lnchohaigentdonds - Routes tertiaires results of the geo-spatial modeling process for land [__] Zone inondable/Forêt riveraine {772 Zone de peuplement très contrôlés suitability presented in Section 6 also be consid- [__] Bâtiments . Polygones d'extension ered, which indicate the area of the University of = FRA DA ee Loin Eos Limonade, the area of the Champin crossing and — Route principale the area of Jesus to be the most attractive for de- _— —— Route secondaire velopment based on the maximization of the attrac- Figure 55 - Areas that should be considered for future development tions and the maximization of the restrictions. The modeling process also yields a highly valuable area Map source: From geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID- : OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis that should to be preserved and restricted between from satellite imagery 1986, 2010 and 2013, the two roads that link Trou-du-Nord with the Champin and Jesus crossings. As a result, the areas To expand further, based upon the following con- develop integrated human settlements to absorb that appear highlighted in yellow in Figure 55, siderations, it is considered that the areas that are future population: should be the ones in which more detailed urban depicted in the grey mesh pattern in Figure 56 planning analyses should be carried out. would be, together, the most convenient ones to . EMERGING »« F Lo #1, SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 81 Lx gs IDB ERM [page 89] e The policy recommended with regards to cre- Asillustrated in Figure 56, these three elements, dation of these nuclei of urban settlement and thus ating a buffer zone for the Three Bay National the PIC, the Champin and the Jesus settlements, prevent further scattering of development the ad- Park between its South limit and RN6, with would serve as the basic articulating elements of a aptation of the lands with road, water and sanita- activities of all types that could absorb all the gridded system of roads and pathways that would tion infrastructure should be undertaken. This could pressures from outside as well as provide the structure a mixed use, mixed dwelling planned begin a process in which different organizations necessary services and opportunities for eco- community. To successfully guarantee the consoli- could locate their individual efforts and projects in nomic exploitation associated to the Park; ° The clear presence of three highly attractive 12 places human settlement as a result of the modeling, all of them inside or adjacent to the SS : buffer zone; "ho SS Re L'4 7 ° The significant demand that the University of UN 7 Fe à y 1 = r / Limonade will create in its immediate zone, "+ {y h. SI | dl and ) di SU : NZ NN = a 14 e The very significant demand that the PIC is bare S , NZ 2 creating and will continue to create. * or À 1 EL ji In the case of the settlement areas in the neighbor- b / À RS S . NO hood of the University of Limonade the proposal à } CR SN à ÿ would configure a linear continuum that not neces- O7 CE ANS }| NS De ” sarily benefits the flow of goods and services that ê TA NE PA ù NN SNS V L \ }, utilize RN6. It is also clear that under the present AL 54 : S INNNSSS 1 4 7 circumstances in Haïti it is difficult to determine if, TX & 2 NN lé à , - in the near future, new road infrastructure will be FA PAR IST IR 2 & a À ? ° , SEEN A 2 2 Me developed. Therefore, a better approach is to de- 7 | N' RSA NS ( le] » velop a good plan that maximizes the opportunities 182 RARE D F ein LNZ MAT. | Al of proximity and location while minimizing potential TES of SEP L m rrè d_ JA negative effects on traffic. This hypothesis will nev- —— — — — Se . ne a ertheless be confirmed with a mobility study that is C1 Parc des Trois Baies Zone de développement urbain planifié === Route principale : . C1] Zone urbaines Zone de agriculture durable —— Route secondaire currently in progress for the same region. [1 Zone inondable/Forêt riveraine 7772 Zones de peuplement très contrôlés —— Routes tertiaires 2] Bâtiments : Plantation de banana In the area surrounding the PIC, and given the fact = Plus restreint _ Polygones d'extension hat thi jor facility has t intsatth = —— Rivière principale that this major facility has two access points at the — FE — Rite: socondalté east and west, it is very likely and recommendable that a functional loop be formed between the in- Figure 56 - Preferred locations for consolidating new urban settlements in the PIC area ternal road of the facility and RN6. This calls, there- ce: From geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), fore, to the consolidation of both the Champin and {saip-oFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing the Jesus areas. analysis from satellite imagery 1986, 2010 and 2013 > EMERGING = Lo Es GIDB \ NORTHERN DEVELOPMENT CORRIDOR, HAITI 82 initiative ERM [page 90] the neighborhood. by the areas depicted in yellow between EKAM, would have to be rapidly destined to agro-industrial Trou-du-Nord and Champin, which considers that, activities of the scale of the banana and sisal planta- For the a proper evolution of the entire setting, a in order to prevent the consolidation of a linear tions that are being set up nearby. buffer zone surrounding the PIC should be imple- urbanization along RN6 as well as the roads leading mented. In addition, a critical role would be played to Trou-du-Nord and Terrier Rouge, those areas The protection of the green zones that appear be- tween the two roads connecting Trou-du-Nord with TP the PIC and Jesus would also be essential for the & success of this scheme. These interventions would ensure a balanced, sustainable becoming of this : SSL region. ia Ÿ | KV « / 7.11 The Three Bays Marine Park 7 SN K À | KR AZ EN In December 2013, the Government of Haïti created Ro NE Ne ) < ; KR k the Parc National des Trois Baies or Three Bays Ma- 1 N À Y 4 RSS F ë rine Park. This covers an area of approximately \ % AN F NN KR S ù 90,000 hectares that includes the bays of Limonade, |? Ÿ lux À ER ; LS S&. Caracol and Fort Liberté, as well as the Lagon aux ) [2 RS : } h VA N Boeufs to the east of Fort Liberté. à Æ 27 D | NS À ME $s Va 3 Je }} À 1 a This newly established marine protection area will 7. ke: € SS 4 Fr ; help protect the mangroves, eel grass beds, reefs É S 74 7 # ÿ ; and habitats housing important fisheries that are a as QI À ’' ä #: s crucial for providing livelihoods to nearby commu- dE 1 4 ! re SJ FR SJ La Lee “> 4 nities. It will also help protect the area from storm #2 LA Le à cet F7, K (S > we + surges and provide local communities with ecosys- : DA LR / | P\ ÿ ee > d Fr Maé su tem services such as carbon sequestration, tourism oh NE CE - - + ——— value and more. The area is also home to numerous = Parc des Trois Baies Zone de développement urbain planifié == Route principale threatened species, including sea turtles, whales Zone urbaines EX Zone de agriculture durable — Route secondaire d mi L bird ! É [1 Zone inondable/Forêt riveraine {77 Zones de peuplement très contrôlés — Routes tertiaires manatees and migratory birds. [_] Bâtiments 21 Plantation de banana —— Rues proposées Dm Pius restreint _ Polygones d'extension es panoramique et piste cyclable To recognize and protect the Three Bays Marine ES “Es Étrte principale pue de Roienane ci PIC Park, a series of ‘preemptive zoning’ classes has EE Pius attractif ivière secondaire : : - . , been developed in conjunction with the IDB's spe- Figure 57 - Creating a planned, integrated community with the PIC as pivot. cialists who are supporting the Three Bays Marine Park development. Figure 58illustrates the pro- Map source: From geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID- osed zoning as follows: OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis p 8 L from satellite imagery 1986, 2010 and 2013 . . . 1. Coral reefs. This a marine ecosystem of high economic value derived from the services it Æ EMERGING = Lo NORTHERN DEVELOPMENT CORRIDOR, HAITI 83 initiative ERM [page 91] provides including i) fish, crustaceans and de Mer de Limonade, La Chappelle, Borony, other marine species that sustain large por- Monto-lon and La Genevré areas in the mu- tions of the local population and ex-ports; ii) nicipality of Limonade:; in the municipality of shoreline protection, providing a barrier to Caracol, they include the Southwest areas of storm surge and impacts of hurricanes on the En Bas Saline, Car-acol, Jackezy and the Cara- shore. Coral reef health is closely linked to col Industrial Park. Additional areas not visible that of the sea grass and mangrove ecosys- in the map in Paulette and Phaeton. tems, and üi) very important carbon sink, se- questering atmospheric carbon. Currently threatened by overfishing and sediment load from eroded watersheds. 2. See grasses. Critical habitat for juvenile stage of fish species, turtles, manatees and other species. They play and important role in se- questering carbon. (not visible in map) 3. High flood zones. This category covers the sea grasses zone, the man-grove areas, and ex- tends beyond the latter into the mainland. The man-groves have been declared for com- plete protection, providing critical ecosystem services such as i) nurseries for commercially important fish species as well as crustaceans, mollusks and bird nesting habitat; ii) shoreline protection from storm surge, sea level rise and hurricanes. Currently threatened by char- coal production, firewood collection, salt pro- duction techniques and overfishing. Commu- nities such as Jaquezy, Caracol, and Madrasse have been established in this area and are se- riously threatened by increasing risk of flood- ing and storm damage. The risk is greatly ex- acerbated by illegal destruction of mangroves. 4. Zones where the population process needs to be highly controlled. This corresponds to the areas identified as currently exhibiting any of the kinds of settlements discussed in the next Chapter. They correspond, largely, to the Bord NORTHERN DEVELOPMENT CORRIDOR, HAITI 84 EH ERM [page 92] f Mt. ==. EN + / \ = | k Ps an, : L sa «G LES } a es _ ; D = É +4 ; * n L . £. / … 4 à À $ p= / | ] : x ( $ ;] k x = . | A P Exf À >" P ‘ 0 25 $ 10 Em Océan Atlantique EM Zone de agriculture durable Ge | no — Féci coralien mm ones uroanes RE C2 Zone d'étude — Zone tampon C1 Parc National des Trois Baies SM Plantation de banana D Ligne de terre ferme ER Future plantation de sisal RE OE EM Zone innodation élevé —— Route principale CEE de paume res conte] An | [uerens Pareaten | Em Zone de peuplement très contrôlés ——— mn Zone de patrimoine culturel et tourisme durable A 2 AE LEE Figure 58 - Preemptive zoning classes proposed for the Three Bays Marine Park Sources: Map source: From geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. = EMERGING » Lo NORTHERN DEVELOPMENT CORRIDOR, HAITI 85 initiative. ERM [page 93] 5. Zones for the protection of cultural heritage locate population outside the boundaries of To facilitate a prioritization of hazards, a simple and the promotion of sustainable tourism. the park. The urban areas outside the pro- methodology has been developed based on a com- These correspond, largely, to the entire West tected area, but nevertheless exerting direct parison of the maximum probable losses for each coast of the Bay of Fort Liberté and the North and indirect impacts on it, require particularly hazard. À matrix is presented in that identifies the shores of the area of study between the strong planning and control structures in re- relative area of concern based on probability (ex- straight of the former bay and the point gard to solid waste management, water use pressed by the return period) and estimated impact where the coral reefs begin. Inland, these ar- and effluents. Finally, (expected losses). eas extend between 500 m and 1kminland, which is where a large number pre-Columbian 8. A buffer zone to absorb and provide the nec- Hazards that tend to occupy the top left-hand and colonial era fortifications and other herit- essary economic processes and settlements quadrant of the chart, Quadrant A, have a high age elements are still visible. Additional areas that would support and be supported by the probability (occur with the most frequency) and of smaller size not visible in this map but ac- park. À band of land and water immediately have a potential high impact (high damage). There- counted for in the modeling process would al- external to, but contiguous with, the park, fore, these are likely to be of greatest concern to so be part of this class. Appropriate uses in- where development is strictly managed to be stakeholders, and consequently should be a focus clude restoration and man-aged visitation, low intensity and compatible with the conser- for risk reduction planning efforts. Areas of sec- guided walks etc. of cultural sites and small vation objectives of PN3B. Plantations of sisal, ondary concern are those hazards identified in scale sustainable tourism infrastructure. banana and others in this zone would have to Quadrants B and C. certify production to be free from use of 6. Zones of sustainable agriculture. A large ex- chemical inputs which negatively affect biodi- Damage Recurrence Comparison tension of the terrestrial area of the park con- versity in the Park such as pesticides and ferti- l sists of land that is appropriate for ecotour- lizers. The park’s regulations refer to a 5 km | B ism, culture-focused tourism and other forms south of the boundary. This could be attained È er l : : : £ High Impact, $ of sustainable tourism as well as sustainable to the east of the park, but to the West, that £ L agricultural practices which are compatible is in the NDC area, this should be extended to è Ton | eee with the conservation of natural resources i.e. the limits of RN6. É C | D that conserve soil, water, biodiversity and | | À H Par | ee ecological cycles and prevent run off into the 7.12 Risk Reduction Recommendations bays. Plantations such as a banana or sisal or : : RecurrenceMRP, years any other kind could take place in the former 712.1 Risk Ranking D areas oftthe Dauphin plantation. Section hazard and risk assessment studies4 pre- Figure 59 - Framework for Relative Risk Evaluation 7. Urban areas. Inside the park there would sents the hazard and risk assessment for the priori- : : have to be a highly controlled mechanism of tized natural hazards, and Table 7 summarizes the Hazards in Quadrant B have alow probability of hazard losses for each hazard. This section builds occurrence but have a potentially high impact, urban growth. The settlements located on ar- . : : . Le eas affected by flooding ought to be re- upon these results and compares and prioritizes the while hazards in Quadrant C have a high probability located. The remaining areas ought to be re- hazards and presents a series of general risk reduc- of occurrence but low impact. Hazards categorized develoged with an adavtive architecture. such tion recommendations. In addition, five mitigation in Quadrant D are likely to be of lowest relative P p , : : : as pilotis. No expansion ought to be allowed, strategies have been explored In more detail concemn because they are predicted to have both 2 " . through a cost-benefit analysis. low probability and a low impact; however, this which would therefore require efforts to re- framework does not necessarily negate the im- == EMERGING Lo sé * SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 86 EX EU % IDB ERM [page 94] portance of addressing hazards that fall into Quad- and damage/loss estimates were calculated in this losses divided by the Mean Return Period, repre- rant D (low probability, low impact). risk assessment: earthquake, hurricane, flooding, senting the amount of capital the local govern- and coastal flooding. It provides a systematic ments would have to set aside to cover the damag- Ranking hazards that fall into Quadrants À, B, and C framework from which to compare and prioritize es for such an event. shows a comparison of losses depends on the level of risk tolerance or im- hazards. It is important to keep in mind that the for different hazards based on the aggregate losses portance a specific hazard might have to the com- comparison does not represent an absolute ranking for a specific return period and the losses per year. munity. of hazards, but has been developed to assist in For example, earthquake and hurricane hazards : : evaluation of the results so as to help decision show highest losses. The coastal flood and inland Figure 60 provides a plot of the losses (general 9c- makers prioritize mitigation measures. floods have comparatively lower losses. When cupancy and infrastructure losses) vs. return period compared in terms of loss per 1000 USD/per year, for each hazard and is an effective way to evaluate The comparison is also presented in tabular form in coastal flooding and hurricane hazard are the top the relative risks. and is based on an expected loss per year for each two hazards and followed by earthquake and inland hazard, simply calculated as the total expected floods. Figure 60 maps hazards for which both recurrence Table 34 - Comparison of Hazards for the study area Damage/Loss Comparison Total Loss Loss 1800 MRP L (1076 (us$)/ 1 A (NE) US$) ear 1600 I y _ 1400 1 £ 1 S 1200 i Coasta FoodIng 2 1000 L a = ————————— Z 800 y * 7.12.2 General Risk Reduction Interventions Ed Ê 600 L - In Haiti, the most urgent risk reduction measure is 8 00 1 to increase economic opportunity and alleviate 1 poverty and improve living conditions to reduce risk 200 L to natural disasters. While increased wealth is not a o Ë 1 save all for reducing the impact of natural hazards, ï : it is often the poor that are less able to afford to (e] 500 1000 1500 2000 2500 3000 : relocate or resettle in areas that are less prone to Recurrence (Years) natural hazards. The poor are also less prone to . . : rebuild or reinvest after a disaster. This section lays Coastal Flooding —®—Flooding —#— Earthquake —e— Hurricane out some general sustainability interventions to increase disaster resilience and reduce losses. Figure 60 - Standardizing loss damage recurrence comparison for the study area == EMERGING «CE GIDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 87 initiative ERM [page 95] Earthquake Hazard lesser degree. Restricted areas should include loca- Several sustainability options for reducing or miti- The section below highlights some high level inter- tions in coastal areas and very steep hillsides. Again, gating risk associated with in-land flooding hazards ventions and mitigation measures, as well as rec- building design standards and construction practic- include: ommendations, to improve the characterization of es take precedence, however, there are a few spe- the earthquake hazard, should be considered. cific and design considerations for hurricanes which ° Developmentin the high hazard flood area include: should be restricted especially within identi- The fact that most of the study area is comprised of fied floodway. High risk lands should not be deep, unconsolidated alluvial sediments, where the ° Orientation of the building relative to site to- considered for future development. If devel- ground shaking hazard is high, the most important pography; opment occurs within high hazard flood areas, mitigation recommendation for reducing seismic ° Ensure that there are stabilizing measures in- there should be detailed planning and devel- risk is to develop a more detailed understanding of corporated into the design and construction opment criteria that ensures that the proper- the hazard itself. Development a detailed seismic of building, especially the connections be- ty is built above base flood elevation. risk or zonation map for the entire study area, tween building parts; ° In high hazard areas where there is existing which details maximum accelerations and allow for e Debris removal in vacant lands to reduce fly- development (formal and informal settle- the identification of high risk districts, will be critical ing projectiles, securing urban furniture, and ments), bank stabilization should be pursued for identifying specific risk areas and to prioritize the burying of utility lines, etc. are simple to help reduce flooding and erosion so as to risk reduction activities. measures that can work to reduce damages help contain river flooding to watercourses. during wind storms; and The re-vegetation of river banks is an im- Another recommendation is to ensure that new ° Reducing the amount of clear cutting as vege- portant consideration for binding silt and soil construction is designed and built in accordance to tation stabilizes soil and trees provide friction to reduce water erosion due to flooding. international building standards and codes (residen- and stabilization from winds. ° In areas where there is urban flooding due to tial, commercial, institutional and industrial). AI overland flows, drainage infrastructure should construction should ensure that there is proper The general poor construction characterizing much be improved to facilitate the flow of water reinforcement in walls and that building connec- of the study area is again tied to Haiti’s poor econ- away from settlements and back to water tions including roof connections to the walls, wall omy and limited building code enforcement. Build- courses. In rural areas, special attention connections to each other and the connection of ing codes have little relevance if they are not appli- should be given to dredging irrigation chan- walls to a strong foundation are present in all new cable to local construction practices, do not support nels and in some case surfacing with concrete construction. known engineering solutions, and have not been bases so as to increase the flow of water to : tested. farmland and out of farmland. For large construction projects and infrastructure : . : fie coiemi e By far the most important measure for river- developments, site specific seismic assessments Inland Flooding ine flooding during intense rainfall events is should be undertaken to identify geologic con The foremost consideration for flood hazard should awareness and education. The development straints for the construction of critical facilities and be given to building a careful record of precipitation of public awareness and public education major infrastructure. information within and surrounding the study re- campaigns should be targeted for high risk Hurricane gion. The lack of reliable precipitation data has groups to increase hazard knowledge, im- Hurricanes tend not to be tied to a specific location, Bi regime ENTRER etat the loc prove TK perception and foster risk avoid especially in such a small geographic study area as ° the study area. Nevertheless, there are areas where development should be restricted to a greater or > EMERGING Lo sl * SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 88 EX gs F IDB ERM [page 96] Coastal Flooding Other adaptive responses may include preventing more stable hydrological cycle. For such approach- There are several sustainability options for reducing hazard impacts by building strengthening protective es, financial or in-kind incentives to involve com- or mitigating risk associated with coastal flooding structures. munities should be explored. hazards. The ability to relocate settlements or adapt existing infrastructure for protection is lim- Drought The construction of reservoirs and the revitalization ited by resources, but the physical expansion and The hydrological assessment has indicated that of irrigation systems may also be another option for the growth of settlements is not. Therefore, in demand for water will increase and the water po- stabilizing the water supply in communities. While coastal high hazard areas, the expansion of settle- tential will diminish due to increased population the construction of reservoirs for each community ments should be discouraged and the construction and development pressures. Therefore, it will be does not help restore natural resource area, they of private and public infrastructure, including roads, paramount that policy makers consider actions that can be effective mechanisms for the storage of energy sub-stations, drains and housing settlements will help reduce the impact of water deficits in the rainwater which can be used during dry seasons should be limited. The location and nature of study area, especially during dry periods. (June to October). Such investments will improve planned infrastructure should draw on forecasted access to water for agriculture and rural house- inundation maps developed under this study. In view future demands, a watershed management holds. If reservoirs are designed correctly, they can approach to mitigating the effects of drought be linked to new or integrated into existing irriga- The protection and expansion of coastal wetlands should be pursued. Integrated watershed manage- tion distribution systems. Attention should be given and estuaries is also an important measure that ment should be incorporated into more diversified to the revitalization of a series of historic canals should be considered. Wetlands that are linked to development planning for the region so as to ad- that are found throughout the study area. the coast and estuaries serve as a natural buffer dress problems related to land degradation and against storm surges, sea-level rise and wave action unsustainable land use practices (i.e. deforesta- Finally, more education and outreach is required to in particular. Natural features help to absorb large tion). increase the efficiency of water utilization in the volumes of advancing water, and as a result, have a area. In this regard, effective institutional support dissipating effect on wave energy. Further reduc- The focus of a watershed management planning {including assistance from international donors) will tion to the size of wetlands and estuaries, is to dis- approach should be to increase ground infiltration be paramount. Projects, particularly technical assis- count their importance that these natural resources so as to increase the water stock and reduce the tance focusing on increasing efficiency water utiliza- play in reducing the impact of coastal hazards. impact of flooding. Many studies have shown that tion, must take into consideration local land use when local water catchment areas are protected, practices and farming methods so as to identify The lack of an effective storm water system for the vulnerability of local agriculture is reduced. local adaptation measures that have a chance of discharging high volumes of water has hindered Such approaches will help restore functions natural being implemented. International donors can play a development meaningful responses to flooding and drainage areas and increase the supply of water role in funding the required research needed to coastal flooding hazards. Drainage infrastructure and have an additional impact in reducing flooding develop such local level approaches, as well as fund needs to be improved to relieve coastal inundation impacts in low lying areas. Reforestation is a critical government institutions to disseminate information caused by coastal storms or heavy rainfall events. component in any watershed conservation program on innovative locally-based interventions that ad- in Haïti. The planting of forest lands, even for tree dress drought and climate change. The coastal flooding hazard should be incorporated crops (coffee) agroforestry (teak) purposes, will into disaster management prevention programs. help developing sustainable water resources in the 7.12.3 Risk Reduction Case Studies There is a need to take into consideration coordi- study area. The reforestation of upland areas Le : : nate responses for the evacuation of high risk popu- should be pursued, along with a series of filtration Specific recommendations and projects that are lations, which will mean focusing resources toward ditches and natural walls to help in reducing the assessed mn this section, were identified based upon : : : : field observations, stakeholder discussions and also prevention as opposed to response and recovery. loss of land by erosion and will help in develop > EMERGING Lo sé * SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 89 EX gs F IDB ERM [page 97] from ensuring tangible and practical projects could ° _A:takes into consideration the targeted de- nerability parameter used for situation A is be implemented. Given the intensity and frequency velopment projects and utilizes a linear ex- based on an assumption that the develop- of flooding (both inland and coastal), these hazards trapolation of population growth and land use ment pressures will continue on the current have been given priority for risk reduction assess- to project spatial development and growth on trajectory and that urbanized areas will in- ment. its current trajectory and assumes that there crease by 7.45% and that little improvements are not any interventions that are put in place will be made to existing infrastructure, while A Cost Benefit Analysis (CBA) model has been used to limit urban expansion and/or the further the situation B, assumes the same rate of to assess the likely costs and benefits of identified deterioration of the natural resource base population increase, but urban expansion will mitigation measures, focusing on the following mit- within and around the study area; and be limited and only increase by 2.6% due to igation strategies in high risk areas: + B:alltargeted development projects for the increased density requirements for future de- study areas are accelerated, but where velopment. B also considers that future de- + Upgrade Urban Drainage Infrastructure; growth is controlled by taking into considera- velopment in hazard prone areas will be lim- + Rural Drainage Infrastructure Implementa- tion sustainability opportunities and con- ited and/or reduced. Both include assump- tion; straints as projected this report. tions that construction practices, in terms of e Revitalize Historical Canal System To Alleviate workmanship and materials, will improve in- Flooding; By comparing future risk for a growth projection crementally over time. °__ Upland Reforestation; and without the consideration of land use planning rec- e Exposure—population growth estimates is e__ Mangrove Protection In the Three Bays Ma- ommendations (A) to a fast growth development used to predict future exposure (value of rine Park. scenario that takes into consideration land use buildings and infrastructure) across the study planning recommendations (B), the impact of intro- region and used a linear regression analysis to Appendix 10 contains the details of the five strate- ducing sustainable land use planning as a risk re- estimate the increased value of assets (i.e. gies explored and the methodology applied, and the duction measure is clearly demonstrated. For this buildings infrastructure, etc.). Therefore, the results are summarized below in Table 35. comparison, a risk projection model was used, model assumes that exposure values will in- : which combines three different components to crease proportional to population growth and 7.12.4 Land Use Planning to Manage Future understand the potential future losses for each will be uniform across different land use cate- Risks hazard for a projected time period to 2040: gories in the study region as defined in for dif- The section above has elaborated on specific ap- . | ferent development scenarios. proaches to reducing risk by introducing and analyz- ° Hazard - The hazard intensity/frequency rela- ing specific management options (structural and tionship is assumed to increase due to climate Table 35 provides a benchmark for decision makers nonstructural). This section looks at the potential change. To provide a consistent basis from to understand the implications of implementing the impacts of the introduction of land use planning as which to compare hazards a 100-vear return urban land use planning recommendations, which a method to reduce risks in the future. period is used. incorporates mitigation measures and introduces a e _ Vulnerability—The general characteristics of sensible utilization of land as a way of reducing the A comparative assessment of future risks has been the built environment are expected to change negative consequences of reducing risks. undertaken considering two situations so as to illus- over time due to the introduction of better trate how decisions regarding the utilization of land building practices and improvements in con- The total aggregated loss estimates for 2013 may impact future losses in the study area. The two struction materials. A vulnerability multiplier amounted to 2.61B USD (i.e. a summation of the situations are as follows: was used to update/modify the building per- aggregate losses for each hazard). Under situation formance from the present to 2040. The vul- À, aggregate losses for each hazard are expected to > EMERGING Lo sl * SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 90 EX See F IDB ERM [page 98] are considered and factored into future growth Table 35 - Potential Losses, 10*6USD models, the future losses can be significantly re- duced and citizens can meet the following goals: CET OT = | Future Loss Future Loss Aggregate Expo- Future Loss gregate Exposure ers who make land use decisions to channel sure Value Value development to low hazard areas and/or flag Seismic Hazard 1694.47 2947.67 51.69% 1784.13 31.29% development proposed in high hazard areas. Flooding Hazard | 1097 14.86 0.26% ges | o16% ° Recommend sustainable locations for major - developments projects and/or infrastructure Hurricane Hazard 815.81 1023.62 17.95% 615.89 10.80% . : . : Le: : 2 projects, including public facilities, and resi- ment. increase substantially. However, if the urban land ysis. The mangrove protection in the Three Bays e Support the conservation of natural re- use planning and sustainability recommendations Marine Park, urban drainage infrastructure up- sources, and the designation of critical areas, from this study are incorporated, which have uti- grade, and upland reforestation are three most agricultural land, or historical resources. lized hazard and risk maps generated as part of this promising interventions. The mangrove protection study to minimize future growth in more vulnerable and upland reforestation, being non-structural To avoid losses occasioned by these hazards, there areas, the projected potential aggregate losses for measures, will also start providing benefits to the is a need for the new approaches and strategies to most hazards is greatly reduced from the loss esti- wider environment in terms of supporting environ- be put in place. The International Strategy for Dis- mates for 2013 and represent approximately 44% of mental protection, aquifer recharge and biodiversi- aster Reduction (ISDR -http://www.unisdr.org) rec- the projected total exposure values in 2040 (2.48B ty enhancement, in addition to the flood manage- ognizes this and emphasizes the importance of un- USD). ment and hazard protections. The urban and rural derstanding local risk patterns, developing strate- drainage upgrade will be beneficial immediately gies to decentralize responsibilities at the relevant Therefore, a substantial amount of potential future after implementation. It is necessary to note that sub-national or local levels, and supports and inte- losses can be avoided if the land use recommenda- while the canal revitalization appears to be low gration of risk reduction, as appropriate, into de- tions are implemented. yielding investment due to the lesser beneficiary velopment and planning policies. areas (sparse settlements and low exposure at risk), 712.5 Risk Reduction Summary it does not preclude conducting such analysis in The information outlined in this section should be areas with denser populations, which may yield used to inform citizens and decision makers about different results. Considering the satisfactory bene- hazards and the risks. This section in particular pro- fit cost ratio of top four scenarios, these options vides a basis for understanding the hazard impacts should be taken forward for possible planning and and prioritizing actions by laying out general sus- pre-feasibility studies. tainability interventions that should be considered . . k . . for building more sustainably in the study area. Finally the analysis provides a basis for examining how risks may increase significantly if growth con- These analyses are preliminary and they indicate tinues in an unplanned fashion, and how through options to be taken forward for more detailed anal- the smart growth scenarios, where the hazard maps == EMERGING Lo sl * SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 91 EX EU % IDB ERM [page 99] Table 36 - Summary of risk mitigation measures Upgrade Key Urban Drainage Rural Drainage Infrastructure Revitalize Historical Canal Sys- Upland Reforestation of the Mangrove Reforestation in the Infrastructure Implementation tem to Alleviate Flooding Trou Du Nord Watershed Parc National Trois Baies e Limonade used as a case study e Example of land drain west e The revitalization of the ca- e Sustainable mitigation meas- e Serve as a natural buffer ° Upgrade drainage to cope of Caracol nals in northern Haiti ure to increase the intercep- against storm surges, sea- with flooding during heavy + Address overland flows in + Comprise dredging to re- tion of water in the upper level rise and wave action. rain (increase capacity) rural areas move silt and increasing the reaches of watershed e Proposed that a 5km strip of e Increase in open drainage (as- e_Install new drainage to cope cross-sectional area of the e Will also work to prevent soil mangrove be reforested sumed 18km) with overland flooding dur- canals erosion and contribute to along coast line. ing heavy rain e _Alength of 3.8 km is evalu- forest conservation ° This would mean about 300 ated (near Jacquezil) e Benefits will be tied to a sq.km of area would be longer time horizon planted. Over 20 years: Over 20 years: Over 20 years: Over 20 years: Over 20 years: © Costs: $3m © Costs: $0.9m e Costs: $0.2m e Costs: $1.2m © Costs: $6.7m e Benefits: S7.7m e Benefits: S1.1m e Benefits: S0.1m e Benefits: S2.8m e Benefits: S60.4m ° CBA: 2.5, which represents a ° CBA: 0.7, which represents ° CBA: 0.7, which represents ° CBA: 2.3, which represents a e CBA:9, which represents an good investment an okay investment an okay investment good investment excellent investment == EMERGING Lo 16 * SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 92 EX EH % IDB ERM [page 100] 8. CONCLUSIONS AND RECOMMENDA- tion quality index, and other land use deci- In terms of human settlement, as illustrated there- TIONS: À SMART GROWTH SCENAR- sions of similar nature. in, the scenario is that in which the urban footprints of the main townships that comprise the region 10 + Understanding and defining the ‘pre-emptive’ outside the Three Bays National Park are expanded zoning classes that should be considered in between 5 and 100 hectares each, in all cases on 81 Study Focus the Three Bays National Park. surrounding areas identified as the most suitable This study seeks to address the complex question of + Seeking to define the ‘geography of what for sanaon nos seenario the urban re : rints of Bord de Mer de Limonade, Caracol, an NDC of WI special eme on hou bunian could be called ‘human settlement for a sus- Jacquezy, are kept with their current dimensions in ( inli j tainable future’ of the region. This has been response to the Three Bays National Park intention Ana an for chat ul fine CDs UOn see o) sarrstancaine snste ee to limit and if possible reduce the urban footprint jeration i ; uilding efforts, levels of urban density that inside its territory. In all three cases, a program of Fe to rares de aLeon Er un a could realistically be pursued in the different re-densification base on an architecture GE pilotis ent angles that included the following: townships of the region, and most important- {which is also applied to the non or low risk flooding ly, testing these parameters in each one of areas in Caracol and Bord de Mer de Limonade) is e _ Understanding the context from physical, so- the townships to determine how would they also proposed in order to establish a culture of cio-spatial and socio-economic points of view. Pare de prrecred fan demon de te edification that is more resilient. ec . e _ Analyzing recent planning efforts that have cases in which areas of expansion would be In this scenario, in which it is assumed that the been key in tracing orientations with regards required for the townships, the exercise in- Government of Haiti will be able to control the ex- to the area; contrasting, comparing and build- cluded tracing them in accordance with the pansion of Caracol and Bord de Mer de Limonade, ing on their conclusions. results of the modeling. and has acquired resources to carry out a gradual, integrated socio-economically equitable and crea- e Conducting additional supplementary popula- + Attempting to define the areas in which new, tive process of resettlement, two new areas have tion analyses and projections with regards to planned settlements should be pursued, as a emerged as contemporary towns. One of them is in the residential land demand. This was under- result of the significant impact that the PIC is the Champin and Jesus areas, and the other in the taken for a slow and a fast growth scenario having in the region, the increased economic University of Limonade — EKAM area. These con- projected to 2040, based on the patterns that activities associated to the University of temporary towns are well designed, with a variety would be seen in the region following the im- Limonade, the large plantations that are ap- of mid to high density dwelling solutions, connected plementation development projects that are pearing in the area and the need to provide between themselves and with the townships in the being planned. alternatives for the coastal townships. area through a system of bicycle and landscaped paths. ° _ Determining, through geo-spatial suitability 8.2 Smart Development Scenario modelling applied with a ‘balanced approach” The results of putting these pieces together are of restrictions and attractions, what lands presented in Figure 61 which outlines recommen- would be more suitable for urbanization, dations for the Northern Development Corridor where should the existing patterns of agricul- Smart Development Scenario. ture be preserved, which areas should be pre- served for their bio-diversity and/or vegeta- EMERGING = Lo 6 \ SUSTAINABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 93 EX gs F IDB ERM [page 101] $ La \L ere LES — L 4 2 ACa Li PP &. 4 ; Re. f RS 2 C- 4 f S À | 0 1] me ! Fa ST Bee ) He UE - ee ; / A > PANNE. | # Fe “ ) 14] È ;. UY. N f) FE Les NO = 14 / é à 4 ÿ + { A 8 p il A ‘4 k d D, nt \ | =: . | é Uk 4 { f f $ > - Rs ?. ; \ S | £ « À aa & L a Fa f à ? 4) À L4 "4 1 4 K » ÿ dd à VER SF ga INA \ F “ | 7 À à a 4 1 £ [,)ç PURES 3 19 À Le: C1 Zone d'étude EN Marécages a vegetation herbacee Marais salants MM Développé à haute intensité EM Plan d'eau D Future plantation de sisal MM Développé à moyenne intensité Les zones humides boisées EN Plantation de banana Développé faible intensité I Les zones humides émergeant Récif corallien Espace ouvert développé SK Zone de développement urbain planifié IN Océan Atlantique I Cuitivé D Zone de patrimoine culturel et tourisme durable nn Pâturage M Zone de agriculture durable EN Forêt M Zones de peuplement très contrôlés Figure 61 - Smart Development Scenario for Haïti's Northern Development Corridor Map source: From geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013 = EMERGING == Lo «EF GIDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 94 Initiative ERM [page 102] Inside the Three Bays National Park, the mangrove tively addressed by the relevant governmental, aid extracted with negative environmental con- areas so critical for the survival of this biodiverse organizations and key economic stakeholders in the sequences) as well as a good array of engi- Caribbean hotspot are healthy and are not suffering region: neering and construction firms? with good ca- the pressures posed by those seeking their liveli- pacity to build complete urbanizations. Should hood in their exploitation for charcoal. This is in + Address the land property rights system in the these factors be put at play in a transparent part thanks to the emergence of alternative sense of creating a cadaster formalizing parcel and private entrepreneurship context such as sources. Consequently, the mudflats next to the boundaries, land uses, owner information, as- the one discussed in the previous point, there mangroves are preserved, which are followed by a sessed value, and similar elements that are would be no need for the Government or aid continuous protection forest. essential for planners and decision makers to organizations to directly provide housing solu- properly understand the situation. This ele- tions, at least for the lower middle and supe- The elements described above manage to coexist in ment is also fundamental for creating a visi- rior income levels. The limitations for a more a balanced way, the system of zones of sustainable ble, transparent and effective land and real varied housing and socially diverse setting ap- agriculture, of cultural heritage and sustainable estate market, which would be a key driver of pear to be the lack of infrastructure and social tourism will have greater opportunity for thriving, a ‘better’ or more ‘sustainable’ territorial or- services that could drive families and individ- providing new and better alternatives for native der, as well a major contributor to raising uals to establish in a given setting. Conse- settlers and farmers. This includes the renewed, peoples and families from poverty. À cadaster quently, efforts to build large housing com- massive sisal plantations that have given more would also bring clarity with respect to the plexes could be put to a more effective and prime matter for the industries in the PIC, as well as lands belonging to the public realm, which sustainable use by enhancing the infrastruc- the banana plantations. would play a key role in defining where would ture and social services in cities and town- it be less costly for society as a whole to plan ships, as well as creating two new infrastruc- Outside the park, the agricultural lands whose voca- and program interventions such as the ‘new tion is for this activity, together with those that city’ that has been discussed for this region. imilar kinds of bioine. Th bservati : have been traditionally used as such continue to do Finally, the cadaster should be implemented similar <Inds of plping. These observations were a So so, with no additional scattered settlements ap- . ’ corroborated through interviews with people in the 1. k in equal terms for both the rural and the ur- area that significant amounts of the stone materials pearing on this realm thanks to the efforts and at- ban settings . : . : : are extracted from riverbeds and quarries along the traction created by the new, integrated planned : | R. roads on the more mountainous areas. settlements in the PIC and University areas. In all e Focus governmental and aid work on infra- three realms, the coast and Park, the valley, and the structure, social services and productive op- 2h ; de mountainous areas all the elements are protected erations such as the PIC. An important discov- A list of approximatelÿ 30 engineering and construc- that reauire this as a result of their environmental É ne p k _ tion companies was collated with the support of a ve l V c u “+ N n nl ery was to find that there is an active and sig- local industry expert. Telephone interviews were held or ecological valie. Consequently, fresh water wi nificantiv large market of construction mate- with several of these companies, and these revealed flow from peaks, through natural channels, before rials” (although some of them are produced or that there is as an industry appropriate capacity (in reaching the wonderful biodiversity area that the ——— terms of technology, experience and labor) to support mangroves form. ? The assertion on the active and significantly large urban development, however the barriers are more construction materials market was based on expert associated to the financial leverages and instruments 8.3 Challenges to be Addressed judgment and observations during field visits to the needed to develop a viable and self-sustaining con- : : area. A visible phenomenon is the large number of struction/development industry (including access to For a proper implementation of the proposed sus- . : : 2 : : : stores along the roads and in the urban areas that are mortgages by middle income families), and to the tainable development scenario, there are a series of A : ? : Lee fund li h Id dtob dedicated to the sale of cement-based bricks and oth- question of property rights and the difficult land mar- > EMERGING TRS SUSTAINABLE IDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 95 iathe ERM [page 103] ture and social services ready-areas around e The long term sustainability of this region also o The criteria used to calculate the capaci- the University of Limonade and around the depends on the pursuit of alternative routes ty of existing townships to accommo- PIC. connecting Cap Haïtien, Ouanaminthe and the date future housing demand was very Dominican Republic. This scenario shows the conservative in that, (i) it includes the ° The efforts of the many aid organizations and path that one of such routes could take, in assumption that all the land that is the IDB could be put together for the creation which many a benefit would be accrued: re- available within the urban settings of a sustainable setting in the region. Instead duced impact on environmentally or agricul- would be divided into an array of uses of dividing the actions of different organiza- turally productive lands, better access to mar- (recreation, commercial, residential, in- tions by sector and within that by locale, a kets by mountain communities, and lesser use stitutional, public open space, forest) concerted plan acting on existing townships of the flood, hurricane and other hazard ex- that, in the process of designing specific and future areas of integrated development posed areas along the coast. area plans, could be arranged in differ- such as the ones proposed in this study, ent ways so as to privilege one land use would yield much better results. e The statements and proposals with regards to over the rest — without affecting the the capacity of existing townships to receive overall mix; and (ii), it establishes the + This study has demonstrated the unused op- and accommodate future population are dimensions of the expansion polygons portunities and capacity that existing cities based on official data and formulas from the that are proposed based on the quanti- and townships have to accommodate growth IHSI. During the process, several stakeholders tative needs of the population expected and its demands. Consequently, there should discussed possibilities in the future such as a at the end of the period of study, under be a proactive move at intervening in these diminished household size, which would in an assumed density. But these areas are places, not only through the land regulariza- turn yield larger numbers of households and a drawn from larger stocks of land sur- tion and formalization process discussed greater demand for housing than that which is rounding each township that fall under above, but also on the different mechanisms assumed in this study. This would affect the the same criteria of ‘suitable for devel- that are needed to unlock the urban land and results or projections in terms of the capacity opment’ as defined through the model- real estate markets. of existing townships. ing process. Consequently, should larger areas be required for expansion be- + Developing one or two integrated planned + There were also discussions with regards to cause the urbanized lands reached satu- settings in the neighborhood of the PIC and the levels of density that the ESCI Growth ration, these could be drawn from a the University of Limonade is essential for re- Study is considering in this study as the basis stock of lands with the same levels of ducing the pressures on the coastal town- for calculating the demand for land. suitability. ships, because of the proximity to the oppor- o Given the previous argument, the mat- tunities for employment and services that + In response to the important questions raised ter of a different demand for housing these facilities create, which could deter the in the previous two points, the following would indeed exercise different pres- pursuit of those townships as places to live or should be considered: sures than those assessed in this study. even induce migration from them. It is im- However, pressures would be related portant to base these planned settings on the o The only population data with rigorous more to the question of ‘when’ would land suitability analyses presented in this re- analyses and projections that is availa- the townships reach saturation and re- port to protect and enhance environmental ble is the one belonging to the IHSI. quire expanding their urbanized setting, resources. Consequently, adopting or basing the rather than to the question of ‘where’ study on other data could not be sup- should the future demand be allocated. ported. = EMERGING = Lo sel SUSTAÏNABLE NORTHERN DEVELOPMENT CORRIDOR, HAITI 96 EX gs F IDB ERM [page 104] o It is agreed that varying densities could many cases the latter seems to occur because significantly alter the picture, and to this of informal processes that seem to be the re- end the ESCI Growth Study has adopted sult of weak governance, what is evident in a reasonable, medium point between this case is that ‘formal’, donor or state- high densities such as those considered promoted organizations are contributing to under the AIA study (which could reach the phenomenon. It is strongly recommended in excess of 350 dwelling units per net that a collaborative and cooperative process hectare of land) and the existing low be implemented between all organizations to density parameters of 50 — 60 dwelling develop integral human settlement settings units per net hectare that the vernacu- on or near areas already developed. lar process of urban development is cur- rently producing. In addition, it is clear + Notwithstanding the latter, the ESCI Growth that under a market of land and real es- Study also recommends, as indicated previ- tate in which more private as opposed ously, focusing on creating two new integrat- to public or donor organizations partici- ed, planned human settlements. One on the pated, densities of about twice the ver- surroundings of the PIC and the other on the nacular measure could be easily surroundings of the University of Limonade. reached. Together with the townships of Limonade, Trou du Nord, and Terrier Rouge principally, + Achallenge that remains unresolved and these could create a system or network or could alter the sustainable scenario devel- loop of townships that could limit the irregu- oped in this study is the question of availabil- lar, informal settlements appearing in the re- ity of water to support urbanization. But, gion. again, a possible scarcity of water would seem to pose greater difficulties for the implemen- tation of distant or isolated housing develop- ments than would to areas already urbanized. This is because water transport and/or instal- lation costs would be far greater larger in re- mote or isolated areas. Consequently, it is necessary and urgent to develop a potable water master study to determine the capacity to provide the resource to the future and un- der-served population. + _Inthis region, as in any other, urbanization occurs through a combination of densification of urban settings and the transformation of rural or rustic lands in the periphery. While in “#4 GIDB NORTHERN DEVELOPMENT CORRIDOR, HAITI 97 as ERM [page 105] BIBLIOGRAPHY Greener, Healthier and Happier. New York, NY: The Joerin and Thériault (2001). Joerin, Florent and Penguin Press, 2011. Thériault, Marius. Using GIS and outranking mul- AIA Legacy, American Institute of Architects. Cap ticriteria analysis for land use suitability assess- Haïtien - Ouanaminthe Development Corridor Re- IHSI 2012. Total Population, Population 18 years ment. International Journal on Geographic Infor- gional Comprehensive Plan. Vol. 1. 3 vols. Port-Au- and older households and Densities Are times in mation Science, 2001, VOL. 15, No. 2, 153-174 Prince: AIA - IADB - USAID, 2012. 2012, IHSI, Departments of Vital Statistics and So- cial, January 2012 NATHAT 2010, Analysis of Multiple Natural Hazards —. Cap Haïtien - Ouanaminthe Development Corri- in Haiti. (NATHAT). Port-au-Prince,Haiti. March 26, dor Regional Comprehensive Plan. Vol. 2. 3 vols. IHSI 20093. Total Population, Population 18 years 2010. Government of Haiti. Port-Au-Prince: AIA - IADB - USAID, 2012. and older households and Densities Are times in 2009, IHSI, Departments of Vital Statistics and So- Open Street Map (OSM) 2013 - Geographically Centre National de l'Information Géographique et cial, January 2009 referenced information on numerous physical and Spatiale (CNGIS) - Multiple layers of geographic spatial elements inside the study area. Database information utilized for developing the maps IHSI 2009b. Trends and Prospects of Population in consulted throughout the study period. associated to this project. Information dated 2012. Haïti at the Departments and Commons 2000-2015, IHSI, Directions of Demographic and Social Statis- UN Office for the Coordination of Humanitarian Comité Interministériel d'Aménagement du tics, February 2009 Affairs (OCHA) - Several layers of geographic Territoire. La Gestion Intégrée Des Bassins Versants information utilized for developing the maps en Haïti: - Méthodologie de délimitation IHSI and CELADE / ECLAC 2008. Estimates and Pro- associated to this project. Information dated 2013. cartographique des bassins versants. Rapport final, jections of the Total Population, Urban and Rural Port-au-Prince: CIAT, 2010. and economically active, IHSI - Census Bureau and Office of Post Disaster Needs Assessment (PDNA) - Latin American Demographic Centre, CELADE / Several layers and geographically referenced —. Plan d'Aménagement du Nord / Nord-Est: Cou- ECLAC, May 2008 information utilized for developing the maps loir Cap - Ouanaminthe. Port au Prince: CIAT, 2012. associated to this project. Information dated 2010 IHSI 2009c. Socio-demographic Large Lessons DTM (2013). IADB - Digital Terrain Model of the learned from 4th RGPH, IHSI - Bureau of the Census, USAID Office of US Foreign Disaster Assistance study area 2013 February 2009 (OFDA) - Several layers and geographically referenced information utilized for developing the Famine Early Warning Systems Network, HAITI Food IHSI 2004. Results of the Fourth General Census of maps associated to this project. Information dated Security Outlook Update, June 2012 Population and Housing, IHSI 2004 2010 2010 Food and Agriculture Organization of the United IPCC AR4, 2007, IPCC Fourth Assessment Report of Nations, 2004. Technical Cooperation Programme. the Intergovernmental Panel on Climate Change Project Title: Assistance to improve Local Agricul- tural Emergency Preparedness in Caribbean coun- IPCC AR5, 2014, IPCC Fifth Assessment Report of tries highly prone to hurricane related disasters. the Intergovernmental Panel on Climate Change Glaeser, Edward. Triumph of the City: How Our IPCC SRES. A Special Report of IPCC Working Group Greatest Invention Makes Us Richer, Smarter, IIl-Special Report on Emissions Scenarios, SRES, IPCC, 2000). ISBN: 92-9169-113-5 > EMERGING Lo TRS SUSTAINABLE ad IDB ; NORTHERN DEVELOPMENT CORRIDOR, HAITI 98 iathe ERM [page 106] APPENDIX 1: Individual GIS Maps for the Ecological System + Ka | #6 BIDB : NORTHERN DEVELOPMENT CORRIDOR, HAITI Appendix tie ERM [page 107] Figure A1 — Topography of the NDC de he - ; dE À 4 Le TS Fa pe Cp à #4 à EN ( 5 DS SE a 0 ei # ir w ISBD TE V@\ A SD: PS a VV 2e; À Sa Ah SNS VEN 3 RW "© JC, 1 Q En \ m ec sé + eh à Na ee te AR ; Re # Me Ps - Loi n 2j Hi #7 EN D NU AS ARS LR 3 k el PRÉSENTS Al SEAL LT AE Ar NEA LL NV EN: 2 RS Dr 4 AV 2 QUES NE SAN A: ESS EEE X ANR NET EC Ca NAT ua) FA Me LT ( Hi AE ES ENT TRES EN RS PTE LRU 38 pi F MT 47 Ne, DRE UN TRS Sa bee a EE JS rs NA ENS Lo 2 RE. CE, a Re ET Ge A A : OONNPES 2 RO Pie OU NCA et AL APR ecrire A C1 Zone d'étude Récif corallien M Zones urbaines BMM Océan Atlantique -— 0-200m 201-400 m = 401-600 m —— 601-840 m —— Route principale —— Route secondaire Map source: From geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. [page 108] Figure A2 — NCD Hydric System: Superficial water re 7 \ es 2e 5 LA #2] Es Fa D *« Le EURE € - { ÿ D AS M RE Rat À s fa € RL le ANR | HE FD L' Gr) É Zi à Lx WS ) l % PAR Er F ) SA # rs 0 + 7. : e# ÿ \. D GNT, LS ES À 0)" De un Que M L 3 EH À ad ee MORE D NRC on de UT CNET RAT ARE à y | | dt CN Ad Rte nes HN SU ARENA CAN AT PS 1 Pvc UN PES à CS AE . EN ARTS RS SA CAES SN À EE + Ne M à at ON AS DURS en … cé AR ON Lee Pr 7m AO C1 Zone d'étude Récif corallien M Zones urbaines DM Océan Atlantique —— Rivière principale —— Rivière secondaire Forêt riveraine EM Zone humide M Lac-étang e _Réservoirs Map source: From geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. [page 109] Figure A3 - NCD Hydric System: Watersheds ff: fe j et +. NL. AE * 4 DE si t PA HS. L F1 mn, a À D œ Ÿ =: À d . ë = A | j'a! E & Fa 4 4 Lg) 7 NS ; NL 3 FA ; 7 Ad EE #2 Sy Cm: S Br DT ENS : LR Se 0ù À es a ir aol C1 Zone d étude Bassin versant 6 M Zones urbaines Bassin versant 7 —— Route principale Bassin versant 8 Route secondaire 1 Bassin versant 9 —— Rivière principale _ Récif corallien —— Rivière secondaire BM Océan Atlantique Bassin versant 0 Bassin versant 5 Map source: From geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. [page 110] Figure A4 — NCD Hydric System: Superficial and underground water | Ce + DE à | \ SP nu Z DT RE Clx ( net À 5 M7 ai € PA. NE É H LA at F ! A at Me | .. \N7#. | à Ti >). D à NN ANTS < LS 1 1 ë @ wr4 , # 4 \ À Fe g' L à AE D VAALD AT) à ‘ Ÿ & > Aû Ÿ lè 4 LEA PORN RER C1 Zone d'étude Aquifère !__] Limite commune Récif corallien Section communale BSM Océan Atlantique MM Zones urbaines —— Rivière principale —— Rivière secondaire —— Route principale —— Route secondaire Map source: ERM (2014) from geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. [page 111] Figure A5 — NCD Strategic Ecosystems LA jé Hi dl D Lu x à S en 7, x s % "4 A * | } LA rnb ait ss ; 4 SW nn < ae l'i - Dr À 4700 à PA 0 PO Le Ve. £ E À a T ; so + QUE 7 su vo PPS PT ZM ë Fe ie 27 70 LS *. {hi Dr à E >» © Fi AA de PR di à #5 1 N “A Ë SA >: < pa La pa g 6 Un, 4 CR EE Le , à ts 19 él re IN À S HE =" AS > NN :) Cu" 2 re / AE 4 € Le LR... (ES < EAPANS EN %- à fes # À > : mo) A 4 de \.Z DE à re eh FINS PA a ù << RS Pa +, By AP TS hi A Fr PA Sn à à p._« LES bu que vie) (ee 4 4 ri Bs L RU CE 4 nn AVES PAT -DP PE] ù A L ( re PT OT ét À \ DUT VOLE jh, C1 Zone d'étude EM L'écosystème des régions montagneuses Parc des Trois Baies = Lits fluviaux et alluvions récentes M Zones urbaines —— Route principale D Espace boisé Route secondaire DM Lac - étang Récif corallien Manglier IN Océan Atlantique Plages et dunes M Zone humide Map source: From geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. [page 112] Figure A6 - NDVI: Normalized difference vegetation index A 3 te À À w + 0. ll 1, mi Th . ! pe $ *# ALLEZ 1 Lens j y TAC A Ca AE 2 LG 82 ) ea SE | F. fu D VAN +2 } / > u ; FX DA fu ARE - ÉoL< sue Z LT 5: mt L.…{ OMS $ 54 cd) 4 Rd + l ; ‘e É 2, 0 É PANNES. ce CA : RE" 4 3 4"0à FAN: w\ és C1 PRES Clan 41 RAA - AS À ee Éje Ï D 25 SA? ‘0 À .# ot ER: D % EYE LÉ car li | C1 Zone d'étude —— Route secondaire Périmètre irrigué Récif corallien M Zones urbaines EM Océan Atlantique Terre aride Les terres dégradées La terre moins saine IBM La terre saine —— Route principale Map source: From geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), FAO(c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. [page 113] Figure A7 - Parc Marin des Trois Baies and Main Ecological Structure à _/: or. Ya D ET re de = ere NET à KPATT ITA AU # à sa Med v = ON a. à DS ER: : LIRE: DR S à 4 a ST FA k Sd \r NET ae / = ù D Fi Le < Je) ae à { où À: -(8 CRT a 4 ( “ES Ÿ CRD, C1 Zone d'étude Forêt riveraine Océan Atlantique Parc des Trois Baies —— Rivière principale M Zones urbaines —— Rivière secondaire IN Espace boisé —— Route principale cE M Lac - étang -— Route secondaire dE U"" Manglier Récif corallien É À > Plages et dunes D l'écosystème des régions montagneuses | | ET F À Zone humide … Lits fluviaux et alluvions récentes s Legend: From geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. [page 114] Figure A8 — Agrological Quality of Soils Classification | ? AB. “4 sf PS # j M à 24 (. a à + mm À L [N bu, = ? à “ [/ d 2 “ Ne 4 el C3 j D : à ei : | à | Œ | Ne à % » 4 * GE] A L 08. w n 2 Î Vi / Fi Pré] Fr + < » À 25 5 10 | x t ; se Km À Ê NN Ê * à À CT A ©] Study area nl EM Océan BE Es | !….: Limite commune IV EE 128% Section communale V —+ 55 | —— — Route principale En VI dj Route secondaire DM VII RE 2571 || MN Eau # Viil = | I MN Urbain wsx 13 Il Recif corallien L_49:391 1 T 45391 ] 18% ] Map source: From geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. [page 115] APPENDIX 2: Individual GIS Maps for Urban and Infrastructure Development EMERGING — c<# GIDB À NORTHERN DEVELOPMENT CORRIDOR, HAITI Appendix LE ERM [page 116] Figure B1 — Mining Concessions 'e ; | » S 4 # ls : F7 \ . + ne 4 hf F - ra FA Nr À 7 È If l DE PAT Xe: LS 1% LE +3 “ ILE d € F7, À { 17 # PEER £ # 11 = Fe D : 4 ff, nee | hs ; F: À C2 Zone d'étude EM Ccéen Atlantique !.: Umite commune Section communale mn Zones vrosines — Route principale —— Route secondaire + Concessions minières Récif corallien [page 117] Figure B2 - Road Network Hierarchy F à. HP CU ee (pfa 2 BAPE NP. FT Fe \ A TRS # à = ZT E_IStudy area © Port os — mm tr À Limite commune m Pont — Route principale mem Océan _——— Route secondaire —— Route tertiaire _——— Routes urbaines @e Aéroport Map source: Variables built from geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. [page 118] Figure B3 — Solid Waste and Waste Water System LIMONADE . | dès cut CARACOL Ps =. : $ A na de. 7 CS 24 LES Fe e ? _ x « : æ TROU DU NORD * TERRIER ROUGE 4° F4 . : : - gs Lt * . ÿ À : N é 0 04 08 16 EM À Section communale I Zones urbaines e Systéme de traitement d'eau en projet e Centre de traitement des déchets + Evacuation des eaux usées EM Océan Atlantique Map source: Variables built from geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), [page 119] Figure B4 - Water Supply System [ONADE CARACOL | j' ÿ _!Le .. / SAS a } | | & TROU DU NORD à FA TERRIER ROUGE > LZ LS NC ” \ Ô Er : \ 4 £ é | L] \ “ ; L] \ x Î û 025 0s U ‘ k | = mm À C7] Study area —_ Tubages !: Limite commune BS Océan Section communale EM Zones urbaines e Station pompage ® Inventaire captage + Reservoirs * Puit Map source: Variables built from geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. [page 120] Figure B5 — Waste Water System ONADE CARACOL \ | PA € ! | * LI TROU DU NORD L "4 TERRIER ROUGE Ê f Et \, à Si, caf r \ * k \ . L2 \ At À \ È } © 025 as 1 1 : | = mm: À C1 Study area !.: Limite commune Section communale Em Zones urbaines * Waste/Storm water drainage nm \Waste water treatment plant — Tubages I Océan Map source: Variables built from geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. [page 121] Figure B6 - Health Facilities System _ te : _—— È - à Fo FUN _— Fe PL NL : AT ] { SE \ # à OM /) { Q À À { ) i ue ri + ». + | À f/ € 1 F D 2 à 4 À L 2 } ) h 2 î . pue 4 —— Pi Le 7. L f Fo © } FN : 0 " { #” 8 ! FT FA L 1 L H m . j . LAS e LS { \ | \ nr Er J * } Ÿ. Îm à à 1" : re ESA % mon) 0! 25 5 10 À 22 ee d i C1 Study area EM Océan Atlantique ! 7: Limite commune Section communale M Zones urbaines + Hôpitaux de premier niveau + Hôpitaux intermédiaires “ Poste de santé Recif corallien Map source: Variables built from geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. [page 122] Figure B7 - Health Facilities System, urban core mosaic ApE CARACOL À . ; Fo L2 LA TROU DU NORD " TERRIER ROUGE LI . ‘ : # N o os 05 ' = mm" À C1 Study area !__.! Limite commune Section communale EM Zones urbaines + Clinic * Hospital * Installations Poste de santé IN Océan Map source: Variables built from geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. [page 123] Figure B8 - Education Facilities System . j 207"... Lf ; . ae f LE . \ 2, \ | * } 4 À N / j à À .i | - LA f L : ri Ce >\ 2 / . #4 À =. 2 ? L, } cs \ Z É 3) # ST a F { 1 Gé PTS LD / { ù D + ARR ; 4 } ÿ. } . De 2. À FL . ê x € / 1 à ÿ PE s 1119 : £ k À A, Fu #° "+ M »,. h # Û ( £ D! d c Ke 72 . ni o( 25 5 , À C1 Study area EM Océan Atlantique !: Limite commune Section communale IN Zones urbaines + Primary + School * College Recif corallien Map source: ERM (2014). Results from the restrictions sub-model. Variables built from geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. [page 124] Figure B9 — Education Facilities System, urban core mosaic LIMONADE CARACOL L = . LA . ud . . . L / a . 7] TROU DU NORD > TERRIER ROUGE LL - L LA - w L è . F2 x LL œ 025 05 t mm: À C1 Study area M Océan Atlantique :_: Limite commune Section communale EN Zones urbaines + Primary + School * College EM Université Map source: Variables built from geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. [page 125] Figure B10 - Economic development projects that have been identified in the North and Northeast region of Haïti ss F” > 287. L NS e . UN Te É. h ' A : : NOUS. US | P « Lun nn) ELA bn, $ Fi SL : li #0 — RIT M ne 4 À 8 FR 0 es nt ee de: RE" PER ET rl À: FE RE 22 TS ET æ e.4 Ni = DE 7 +. NET NX ED Ve) à PSP 440 à VA St ü #7 Es Fe 4 7 5 vi aù en g” itR D re = AN Ke) nr F3 a. + (ef | HS TAN: » D Gun SM Pa M ARR EE es ne a Pc (QT TR AN . AT CR Dre VOA D PAR SNS PNA tt SLT WE) CNT PR DENT CD PT à Le TT 28 RS © pa C1 Study area e installations industrielles Recif corallien !_.2: Limite commune e Services de police EN Océan Atlantique Section communale e Marchés MM Zone urbaines ® Hôtels M Parc Industriel e Établissements de santé EM Ekam ® Stations-service à combustible EM Lhiversité — Route principale e Points financière du service —— Route secondaire Note: It is important to note, however, that this criterion does not mean that beyond the municipality of Terrier Rouge, into Fort Liberté and Ouanaminthe, there is no economic activity of significance, especially along RN6. What it means is that it is likely that land and settlement changes which may occur, will likely respond to economic factors other than those associated with the development of the Caracol Industrial Park, thus falling outside the realm of this study [page 126] APPENDIX 3: Climate Studies by the University of West Indies + Ka | at: IDB ;: NORTHERN DEVELOPMENT CORRIDOR, HAITI Appendix EU ERM [page 127] Projected Changes in 5 Atmospheric Variables for selected grid boxes over Haiti from the PRECIS RCM Prepared by Climate Studies Group, Mona The University of the West Indies February 2014 [page 128] 1. About the Model PRECIS was developed by the Hadley centre (UK) in order to help generate high-resolution climate change information for as many regions of the world as possible. PRECIS is made freely available to groups of developing countries in order that they may develop climate change scenarios at national centres of excellence, simultaneously building capacity and drawing on local climatological expertise. http://www.metoffice.gov.uk/precis/intro. PRECIS is a hydrostatic primitive equations grid point model. It contains 19 levels in the vertical and has horizontal resolutions of 0.44°x0.44° (50 km) and 0.22°x0.22° (*25 km). Initial and lateral boundary forcing are taken from reanalysis or from outputs of General circulation Models (GCM5s). The sea surface temperatures (SSTs) and sea-ice fractions surface boundary conditions are from a combination of monthly HadiSST1 dataset and weekly NCEP observed datasets. Observed values of greenhouse gases are also fed into the model. PRECIS utilises a relaxation technique across a four point buffer zone at each vertical level. Dynamical flow, the atmospheric sulphur cycle, clouds and precipitation, radiative processes, the land surface and the deep soil are also described in the model. A full description of the model’s physics is found in Jones et al. (2004). Validation of the PRECIS Model for the Caribbean is offered in a number of papers including Campbell et al. (2011) and Taylor et al. (2013). Campbell et al. (2011) ‘compared PRECIS’s modeled patterns of temperature and precipitation with reanalysis datasets and available observations. They showed the mean Caribbean climatologies to be generally captured by the model, with the relative timing of temperature and precipitation maxima and minima being reproduced. This included the model's reproduction of the Caribbean midsummer rainfall minimum, which is a significant feature of many of the larger Caribbean islands. There was, however, also a general underestimation of rainfall amounts across the main Caribbean basin during the wet season and a simulation using temperatures that were too warm over the Caribbean islands but too cold over Central America and northern South America’ (Taylor et al. 2013). 2 [page 129] 2. About the Perturbed Physics Experiments (PPESs) The first set of results presented are from the perturbed physics ensembles (PPE). PPEs are designed by varying uncertain parameters in the model’s representation of important physical and dynamical processes. PPEs are used to capture some major sources of modelling uncertainty by running each member using identical climate forcings. It provides an alternative to using GCMs developed at different modelling centres around the world (e.g. a multi-model ensemble, MME), like those in the CMIP3 (Coupled Model Intercomparison Project 3). The Hadley Centre’s PPE includes 17 members which are formulated to systematically sample parameter uncertainties under the A1B emissions scenario — this is referred to as the QUMP (Quantifying Uncertainties in Model Projections) ensemble. The QUMP ensemble was designed for use in the UK’s own climate projections and is described in detail in the UKCP report available online at http://ukclimateprojections.defra.gov.uk/content/view/944/500/. Globally, and for many regions and variables, the range of climate futures projected by the QUMP PPE is equivalent or greater than those based on the CMIP3 MME. The PPE systematically samples the parameter uncertainties, exploring a wider range of possible variation in the formulation of a single model, leading to a wider range of physically plausible future climate outcomes than the MME. It is important to remember that PPE (similarly for MME) does not account for all of the sources of model uncertainty. ! The following 6 QUMP experiments were evaluated: Q0, Q3, Q4, Q10, Q11, and Q14. AÏl were run at 25 km and from 1960 through 2100. For each experiment the deviation of a future decade e.g. 20205, 20305, 2040s from the experiments baseline (1960-1990) were determined. This gave an ensemble of 6 future changes for each decade. The ensemble results are summarised and presented in Tables below for the 2040s (as requested). 2.1. Grid Boxes The model grid boxes over Haïti are as shown in Figure 1 below. Projections for grid boxes 46, 47, 53, 54, 60 and 61 are provided as requested. ! Portions of the narrative are adapted from narrative on the PRECIS webpage http:/www.metoffice.gov.uk/precis/qump 3 [page 130] T CT [7 19-28-8217] pe | |] | 162026 33 40166 ssi00 27 34 41,47 5461 | | -1 $5 42 48 55 63 | 21-28 36,43 49 56 64 1,4 711 222937 350 5765 25/8 12 1623-3038 44 5458 3 69 1347-24,31 39455259 66 10 14 18 67 À Co D 2 on GR RE MA Figure 1: PRECIS 25 km model grid boxes over Haiti. Table 1: Coordinates for each grid box. Grid Box # Longitude Latitude 61 71.75 W 19.5 N 60 71.75 W 19.75 N 54 72W 19.5 N 53 72 19.75 N 47 72.25 W 19.5 N 46 72.25 W 19.75 N 2.2. About the Data Future change data are provided for five variables. For four of the five variables the data are provided as absolute change. These variables are: minimum temperature (°C), maximum temperature (°C), mean temperature (°C) and 10 m wind speed (m/s). Percentage change is provided for precipitation. Data is averaged for over three month seasons: November-January (NDJ), February-April (FMA), May- July (MJJ) and August-October (ASO), roughly consistent with the Caribbean dry season and wet season (Taylor et al. 2002). The mean annual change is also given. The change for each variable and for each period is calculated for the 20405 for each member of the ensemble. The minimum, maximum and mean values of the 6 member ensemble are provided. 4 [page 131] 2.2.1. Data by Variable Table 2: Projected absolute change in minimum temperature (°C) for the 2040s relative to the 1960- 1990 baseline. Data presented for minimum, maximum and mean value of a six member ensemble. Values are for 25 km grid boxes shown in Figure 1. Change in Minimum Temperature (°C) GRID BOX 61] 60] 54) 53] a7| 46 5 [page 132] Table 3: Projected absolute change in maximum temperature (°C) for the 20405 relative to the 1960- 1990 baseline. Data presented for minimum, maximum and mean value of a six member ensemble. Values are for 25 km grid boxes shown in Figure 1. Change in Maximum Temperature (°C) GRID BOX 61) 60] 54] 53| 47] «6 6 [page 133] Table 4: Projected absolute change in mean temperature (°C) for the 20405 relative to the 1960-1990 baseline. Data presented for minimum, maximum and mean value of a six member ensemble. Values are for 25 km grid boxes shown in Figure 1. Change in Mean Temperature (°C) GRID BOX 61) 60! sa] 53] a7| 7 [page 134] Table 5: Projected absolute change in 10 m wind speed (m/s) for the 20405 relative to the 1960-1990 baseline. Data presented for minimum, maximum and mean value of a six member ensemble. Values are for 25 km grid boxes shown in Figure 1. Change in Wind Speed at 10m (m/s) GRID BOX 61] 60! sa] 53] a7| 46 8 [page 135] Table 6: Projected percentage change in precipitation (%) for the 2040s relative to the 1960-1990 baseline. Data presented for minimum, maximum and mean value of a six member ensemble. Values are for 25 km grid boxes shown in Figure 1. Change in Precipitation (%) GRID BOX 61] 60] sa) 531 a7| 4 [MIN | 19.14 -28.09 -19.12 -34.16 -20.88 -25.72 MEAN 11.33 MIN | 22.51 -32.84 -24.98 -32.40 -23.32 -31.88 MEAN [MIN | 1144 -23.03 -14.42 -30.62 -14.55 -50.38 MEAN | 475) -996| -5.59) 647) 443) 10.14 [MIN | 13.98 -25.56 -17.63 -29.04 -14.20 -41.20 MEAN -11.03| -1196| 1189) -11.82 -15.72 [MIN | -13.08 -22.99 -16.23 -20.70 -12.89 -25.48 MEAN 9 [page 136] 2.2.2. Data by Grid Box Tables 7-12: Mean projected absolute change in minimum, maximum, and mean temperature and 10 m wind speed and mean projected percentage change in precipitation for the 20405 relative to the 1960-1990 baseline for each 25 km grid box. Data presented for mean value of the six member ensemble. Grid boxes are as shown in Figure 1. Table 7: Grid Box 61 Variable _| Mean Temp _| Min | MexT | wind | precip | ND | 169 179| 162| -003| 301 _FMA | sal 181] 151/ 001) 403 M] 165) 200] 190] 008| 475 _AS0 | 18] 104| 184] 049 | -1.03 _ Annual | 160] 188] 172! o06| -360| Table 8: Grid Box 60 | Variable | Mean T | Mint | MaxT | Wind | Precip | RS ER EE 7 [nn | 148 179) 255) -004| -860 [_rma | 139] 181] 14) oo! 998 [mu | 125] 294] 172) 01! -956 [aso | 160] 192] 162| o27| -1196 [Annual | 143] 187] 159) o10! -543| 10 [page 137] [Variable | Mean | Min | maxr | Wind | Precip | [__ L < ‘ %| ml *] Lu |_rma | | mu | | aso | |_ Annual | [vansbie [Mean T_ | Mint | Max [Wind |Precir 5e | nu | | ra | | mu | | aso | | annual [bn 138/h 167148 /007| 423) 11 [page 138] [vriable | Mean r | min | ex | wina | precip | ___— _w | __rwa | BTE | aso |_ annual | Durable can | Min 7 [Maur | Wind | rece 5e | ni | |_rma | |__| |_nso | |_ annual | 12 [page 139] 3. About the SRES Experiments The results presented in this section come from experiments in which the PRECIS model was forced by the ECHAMA4 Global Climate Model at its lateral boundaries. In these experiments the model was run from 1960-2100 and a baseline of 1960-1990 used to determine change for future years. The model was run at 50 km resolution and for one run each of the A2 and B2 scenarios described by the Intergovernmental Panel on Climate Change’s (IPCC) Special Report on Emissions Scenarios (Nakiéenovié et al. 2000). 3.1. Grid Boxes and Data Data is presented for one 50 km grid box centered on 72W and 19.5 N. The pattern follows that of the previous Tables. Data are only presented for the 2040s. CPE RTE TT) CL RRER EEE RER FRS C IRIS PIE Figure 2: PRECIS 50 km model grid boxes over Haiti. Box with X denotes grid box used. 3.2. Data Table 13: Projected absolute change in mean, maximum and minimum temperature (°C) for the 2040s relative to the 1960-1990 baseline. Data presented for A2 and B2 scenarios (one run each). Values are for a 50 km grid box centered on 72W and 19.5 N. D A S | | 4 B2 A2 B2 A2 B2 NDJ 2.0 1.7 2.5 2.2 1.6 1.3 FMA 2.0 1.6 2.4 1.9 1.8 1.5 MI) 2.2 1.9 2.4 2.2 2.2 1.8 ASO 2.2 2.3 2.6 2.6 2.1 2.3 Annual 2.1 1.9 2.5 2.2 1.9 1.7 13 [page 140] Table 14: Projected percentage change in precipitation (%) and absolute change in 10 m wind speed (m/s) for the 20405 relative to the 1960-1990 baseline. Data presented for A2 and B2 scenarios (one run each). Values are for a 50 km grid box centered on 72W and 19.5 N. [ Precipitation (%) Wind Speed (m/s) | A2 B2 A2 B2 NDJ 37.6 58.9 -0.11 -0.25 FMA -1.3 15.0 -0.08 0.24 MJJ -13.4 -5.6 0.17 0.18 ASO 0.8 -13.4 0.09 0.17 Annual 5.9 13.7 0.02 0.08 4. References Campbell, J. D., M. A. Taylor, T. S. Stephenson, R. A. Watson, and F. S. Whyte, 2011: Future climate of the Caribbean from a regional climate model. Int. J. Climatol., 31, 1866-1878, doi:10.1002/joc.2200 Jones, R. G., M. Noguer, D. Hassell, D. Hudson, S. Wilson, G. Jenkins, and J. Mitchell, 2003: Workbook on generating high resolution climate change scenarios using PRECIS. UNDP, GEF, and Met Office Hadley Centre Manual, 32 pp. Nakiéenovié, N., and R. Swart, Eds., 2000: Special Report on Emissions Scenarios. Cambridge University Press, 599 pp Taylor, M. A. D. B. Enfield, and A. A. Chen, 2002: The Influence of the tropical Atlantic vs. the tropical Pacific on Caribbean Rainfall. J. Geophys. Res., 107(C9) 3127, doi:10.1029/20011C001097 Taylor, M. A., and Coauthors, 2007: Glimpses of the future: A briefing from the PRECIS Caribbean Climate Change Project. Caribbean Community Climate Change Centre, 24 pp. Taylor, M. A., A. Centella, J. Charlery, A. Benzanilla, J. Campbell, |. Borrajero, T. Stephenson, and R. Nurmohamed, 2013: The PRECIS-Caribbean Story: Lessons and Legacies. Bull. Amer. Meteor. Soc doi: 10.1175/BAMS-D-11-00235. 14 [page 141] Evaluation of trends in sea levels and tropical storm intensities Prepared by Climate Studies Group, Mona The University of the West Indies February 2014 [page 142] ACKNOWLEDGEMENT The following authors contributed to the compilation of this report: Tannecia S. Stephenson Jhordanne Jones Michael A. Taylor i [page 143] ABOUT THIS DOCUMENT This report presents an assessment of current literature on current and projected trends in sea level rise and storm intensities with particularly emphasis (where possible) on future values for the Caribbean region. Specifically the document reviews: 1. Current and projected trends of mean and extreme sea levels globally and regionally. 2. Current and projected trends in storm intensities as characterized by the Accumulated Cyclone Energy (ACE) and the Power Dissipation Index (PDI) ii [page 144] Table of Contents About this document ii List of Figures iv List of Tables vi At a Glance vii 1. Background 1 1.1 Introduction 1 1.2 SRES and RCP Scenarios 2 2. Sea level rise 5 2.1 Introduction 5 2.2 Causes 5 2.3 Current Trends 6 2.3.1 Global 6 2.3.2 Caribbean 6 2.4 Projected Trends 7 2.4.1 Global and Caribbean 7 2.5 Observations and projections of sea level extremes 11 2.5.1 Observations of sea level extremes 11 2.5.2 Projections of sea level extremes 12 2.6 Uncertainties 16 3. Tropical cyclones 17 3.1 Hurricane Metrics 17 3.1.1 Accumulated Cyclone Energy (ACE) 17 3.1.2 Power Dissipation Index (PDI) 18 3.2 Current Trends in North Atlantic Intensities 20 3.3 Projected Trends in North Atlantic Intensities 20 4. Some References 23 ji [page 145] List of Figures Figure 1.1 Schematic illustration of the four SRES storylines 3 Figure 1.2 Radiative Forcing of the Representative Concentration Pathways. Taken from van Vuuren et al (2011). The light grey area captures 98% of the range in previous IAM scenarios, and dark grey represents 90% of the range. 4 Figure 2.1 Projections of global mean sea level rise over the 21st century relative to 1986-2005 from the combination of the CMIPS ensemble with process-based models, for RCP2.6 and RCP8.5. The assessed likely range is shown as a shaded band. The assessed /ikely ranges for the mean over the period 2081-2100 for all RCP scenarios are given as coloured vertical bars, with the corresponding median value given as a horizontal line. 9 Figure 2.2 (a) Ensemble mean projection of the time-averaged dynamic and steric sea level changes for the period 2081-2100 relative to the reference period 1986-2005, computed from 21 CMIP5 climate models (in m), using the RCP4.5 experiment. The figure includes the globally averaged steric sea level increase of 0.18 + 0.05 m. (b) RMS spread (deviation) of the individual model result around the ensemble mean (m). Source: IPCC ARS. 9 Figure 2.3 Projected relative sea level change (in m) from the combined global steric plus dynamic topography and glacier contributions for the RCP4.5 scenario over the period from 1986-2005 to 2081-2100 for each individual climate model used in the production of Figure 2.2 Source: IPCC ARS. 10 Figure 2.4 Estimated trends (cm per decade) in the height of a 50-year event in extreme sea level from (a) total elevation and (b) total elevation after removal of annual medians. Black dots indicate trends are not significant at the 95% confidence level. Data are from Menéndez and Woodworth (2010). 14 Figure 3.1 Showing the total ACE value per year as a percentage of the 1981-2010 median for the period 1950-2012. Years falling above 120% are considered to be above-normal, years falling in between 71 - 120% are considered to be normal activity, while years falling below an 71% are considered to have below-normal activity. Years exceeding 165% is deemed to be very active. 18 Figure 3.2 Projected changes in tropical cyclone statistics. All values represent expected change in the average over period 2081-2100 relative to 2000-2019, under an A1B-like scenario, based on expert judgement after subjective normalisation of the model projections. Four metrics were considered: the percent change in 1) the total annual frequency of tropical storms, Il) the annual frequency of Category 4 and 5 storms, III) the mean Lifetime Maximum Intensity (LMI; the maximum intensity achieved during a storm's lifetime), and IV) the precipitation rate within 200km of storm center at the time of LMI. For each metric plotted, the solid blue line is the best guess of the expected percent change, and the coloured bar provides the 67% (likely) confidence interval for this value (note that this interval ranges across -100% to +200% for the annual frequency of Category 4 and 5 storms in the North Atlantic). Where a metric is not plotted, there is insufficient data (denoted "X") available to complete an assessment. A randomly drawn (and coloured) selection of historical storm tracks are underlaid to identify regions of tropical cyclone activity. Source: IPCC ARS. 22 iv [page 146] List of Tables Table 1.1 Summary of SRES Storylines. 3 Table 1.2 Descriptions of the Representative Concentration Pathway (RCP) Scenarios. 4 Table 2.1 Rates and absolute change in mean sea level pressure. Rates are obtained from IPCC (2013). 6 Table 2.2 Observed rates of sea level rise for some Caribbean stations. Source: The State of the Jamaican Climate (2013). 7 Table 2.3 Projected increases in global mean surface temperature and global and Caribbean mean sea level from the IPCC (2007) contrasted with those of Rahmstorf (2007). Projections are by 2100 relative to 1980-1999. Source: CARIBSAVE Climate Change Risk Atlas — Jamaica (2011). 8 Table 2.4 Projected increases in global mean surface temperature and global mean sea level. Projections are taken from IPCC (2013) and are relative to 1986-2005. 8 Table 2.5 Major storms affecting the east coast of the United Sates from 1970-2005. Adapted from Irish et al. (2008) 11 Table 2.6 Summary of studies examining projections of sea level extremes. Compiled from IPCC (2013). 13 Table 2.7 Storm surge heights under different sea surface temperature and sea level rise scenarios. (Wind speed of 225 km h? corresponds to that of the April 1991 cyclone impacting Bangladesh.) Taken from Ali (1996). 15 Table 2.8 Storm surge under different sea level rise and sea surface temperature rise conditions. (Scenario | represents base condition that corresponds to wind speed and central pressure as observed during the 1991 cyclone) Derived from Emanuel (2005) and Ali (2000). 15 v [page 147] At A Glance Future Projections for 3 key variables are shown below: A. Mean sea Level Table A1 Historical rates and absolute change in global mean sea levels. Rates are obtained from IPCC (2013). [Referenced as Table 2.1 in text] Rate (mm yr!) Total sea level rise IPCC Likelihood 1901 — 2010 1.7 +0.2 0.19 + 0.02 Very likely 1971 — 2010 2.0 + 0.2 [= | Verylikely 1993 - 2010 3.2+0.4 [= | Verylikely Table A2 Projected increases in global mean surface temperature and global and Caribbean mean sea level from the IPCC (2007) contrasted with those of Rahmstorf (2007). Projections are by 2100 relative to 1980-1999. Source: CARIBSAVE Climate Change Risk Atlas — Jamaica (2011). [Referenced as Table 2.3 in text.] Scenario Global mean surface | Global mean sea level | Caribbean mean sea temperature (°C) rise (m) level rise (+0.05 m) relative to global mean IPCC B1 1.1-2.9 0.18 — 0.38 0.14 — 0.43 IPCC A1B 17-44 0.21 — 0.48 0.16 —0.53 IPCC A2 2.0-5.4 0.23-0.51 0.18 —- 0.56 Rahmstorf, 2007 [= | Uptol4m Upto1.4m Table A2 Projected increases in global mean surface temperature and global mean sea level. Projections are taken from IPCC (2013) and are relative to 1986-2005. [Referenced as Table 2.4 in text.] D Lames | [ami-20 | | | Variable "| Scenario | Mean | Likelyrange | Mean | Likely range Global Mean | RCP2.6 0.4-1.6 0.3-1.7 Surface RCP4.5 0.9 2.0 1.1-2.6 Temperature | RCPG0 08-18 14-31 Changel"c) RCPBS 14-26 26-48 Lelyrange Lily range Global Mean Sea | RCP2.6 0.17-0.32 |0.40 |0.26-0.55 Level Rise (m) RCP4.5 0.19 — 0.33 0.32 — 0.63 RCPSO [025 018-032 [045 [033-063 CPAS [030 [022-038 |063 [045-022 vi [page 148] B. Tropical cyclones Projections relating to intensities suggest the following: e__ Simulations with high resolution dynamical models and statistical-dynamical models consistently find that greenhouse warming causes tropical cyclone intensity to shift towards stronger storms by the end of the 21° century (2 to 11% increase in mean maximum wind globally). ° _Applying 21* century sea surface temperature projections to statistical relationships between local or relative SST and tropical cyclone power dissipation, suggests power dissipation increasing by about 300% in the next century for one relationship but suggesting no change in the other relationship. Both relationships can be reasonably defended based on physical arguments but it is not clear which, if either is current. ° When simulating 21* century warming under A1B, the present models and downscaling techniques suggest increases in intensity and fraction increases in the number of most intense storms. Of concern however is the limited ability of global models to accurately simulate upper- tropospheric wind which modulates vertical wind shear and tropical cyclone genesis and intensity evolution. vii [page 149] 1. BACKGROUND 1.1 Introduction The Caribbean region has been characterized as among the most vulnerable to climate change and climate extremes. This is in the context of limited natural and human resources, restricted lands, densely populated urban and coastal areas, economic dependence on international funders, and heavy reliance on fragile sectors such as tourism. In response, small island states are compelled to undertake assessments towards characterizing present and future climate trends, evaluating possible impacts and proposing possible adaptation and mitigation strategies. One of the major challenges facing island states is that posed by tropical cyclone events and sea level rise. Sea-level rise greatly impacts human activity near the coastal zone (IPCC, 2007) since in many cases the majority of human settlements, economic activity, infrastructure and services are located at or near the coast and local economies are often reliant on just a few sectors such as tourism and agriculture (Nicholls, 1998). Sea level rise therefore exacerbates the vulnerability of coastal regions to other physical processes (e.g. storm surges, storm waves). Another direct influence of sea-level rise is the inundating of low level coastal areas which is of concern. It is anticipated however that ocean waves and storm surges of the future will exhibit changes in characteristics from the present climate and are the dynamic side issue of climate change. It is necessary to seriously consider the impacts of these dynamic phenomena for coastal disaster prevention and reduction, if extreme weather events will become stronger than those in the present climate (Mori et al., 2010). This review examines the current state of knowledge of trends in (i) sea level rise, and (ii) tropical cyclone intensities. The report concentrates on these factors since the relative sea level and its rise over time influences the extent to which the surge and wave heights generated by hurricanes impact coastal areas. In the past relative sea level rise was included in coastal protection design by raising design water levels an amount equivalent to the relative sea level rise. But surge generation and propagation are nonlinear processes and linear addition of relative sea level rise to design water levels underestimates the impact in many areas. In addition to the surge elevation, wave heights also increase with water level in coastal areas where wave height is limited by water depth. 1 [page 150] 12 SRES and RCP Scenarios Long term projections discussed in this report are premised on assumptions made with respect to human activities or natural effects that could alter the climate over decades and centuries. Future anthropogenic emissions of greenhouse gases (GHG), aerosol particles and other forcing agents such as land use change are dependent on socio-economic factors, and may be affected by global geopolitical agreements to control those emissions to achieve mitigation. The Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report (AR4) made extensive use of the SRES scenarios that do not include additional climate initiatives, which means that no scenarios were included that explicitly assume implementation of the United Nations Framework Convention on Climate Change (UNFCCC) or the emissions targets of the Kyoto Protocol. However, GHG emissions are directly affected by non- climate change policies designed for a wide range of other purposes. The SRES scenarios were developed using a sequential approach, i.e., socio-economic factors fed into emissions scenarios, which were then used in simple climate models to determine concentrations of greenhouse gases, and other agents required to drive the more complex atmosphere-ocean global climate models. The SRES scenarios were labelled A1, A2, B1 and B2, describing the relationships between the forces driving greenhouse gas and aerosol emissions and their evolution during the 21st century for large world regions and globally. Each storyline represents different demographic, social, economic, technological, and environmental developments that diverge in increasingly irreversible ways (Nakicenovic et al. 2000). See Figure 1.1 and Table 1.1. In the IPCC Fifth Assessment Report (ARS), outcomes of climate simulations that use new scenarios (some of which include implied policy actions to achieve mitigation) referred to as “Representative Concentration Pathways” (RCPs) are assessed. These RCPSs represent a larger set of mitigation scenarios and were selected to have different targets in terms of radiative forcing at 2100 (about 2.6, 4.5, 6.0 and 8.5 Wm”*). They are defined by their total radiative forcing (cumulative measure of human emissions of greenhouse gases from all sources expressed in Watts per square metre) pathway and level by 2100. The scenarios should be considered plausible and illustrative, and do not have probabilities attached to them. The SRES scenarios resulted from specific socio-economic scenarios from storylines about future demographic and economic development, regionalization, energy production and use, technology, agriculture, forestry and land use (IPCC, 2000). The RCPs are new scenarios that specify concentrations and corresponding emissions, but not directly based on socio-economic storylines like the SRES 2 [page 151] scenarios. The four RCP scenarios are identified by the 21* century peak or stabilization value of the RF derived from the reference model (in Wm?). Table 1.1 provides a summary of the RCPs. SRES Scenarios Economic À Global D Regional Em tal ä ST t$ rech "2 à: D, :s tving For<* Figure 1.1: Schematic illustration of the four SRES storylines Table 1.1 Summary of SRES Storylines. SRES scenario | Storylines family A1 A future world of very rapid economic growth, global population that peaks in mid- century and declines thereafter, and rapid introduction of new and more efficient technologies. A1FI (fossil intensive), AIT (predominantly non-fossil) and A1B (balanced across energy sources). A2 A very heterogeneous world with continuously increasing global population and regionally oriented economic growth that is more fragmented and slower than in other storylines B1 A convergent world with the same global population as in the A1 storyline but with rapid changes in economic structures toward a service and information economy, with reductions in material intensity, and the introduction of clean and resource-efficient technologies. B2 A world in which the emphasis is on local solutions to economic, social, and environmental sustainability, with continuously increasing population (lower than A2) and intermediate economic development. 3 [page 152] 10- ———RCP6 —— RCP4,5 — RCP3PD/RCP2.6 8+1 —RCP8.5 E 2 Z 6 O = 5 4 FA 2 & 2 TD (ee m O0 -2 2000 2025 2050 2075 2100 Figure1.2: Radiative Forcing of the Representative Concentration Pathways. Taken from van Vuuren et al (2011). The light grey area captures 98% of the range in previous IAM scenarios, and dark grey represents 90% of the range. Table 1.2 Descriptions of the Representative Concentration Pathway (RCP) Scenarios. Radiative Forcing Behaviour RCP2.6 Peaks at 3 Wm”? and then declines to approximately 2.6 Wm ? RCP4.5 Stabilization at 4.5 Wm? RCP6 Medium-high Stabilization at 6 Wm? RCP8.5 REF of 8.5 Wm? by 2100 but implies rising RF beyond 2100 4 [page 153] 2. SEA LEVEL RISE 221 Introduction Global sea levels have risen through the 20th century. The rising levels are expected to accelerate through the 21st century and beyond because of global warming, but their magnitude remains uncertain. Key uncertainties include the possible role of the Greenland and West Antarctic ice sheets and the amplitude of regional changes in sea level. In many areas, non-climatic components of relative sea level change (mainly subsidence) can also be locally appreciable. Cooper and Pilkey (2004) suggests that factors which cause changes in the morphology of coasts are numerous and include sediment supply, variations in wave energy, tidal currents, wind action, sediment type, tidal inlet dynamics, morphological feedback, etc. Therefore isolating the influence of sea-level rise from these other factors is perhaps the biggest challenge in discerning the impact of sea level rise. 2.2 Causes Two main factors contribute to SLR: (i) thermal expansion of sea water due to ocean warming and (ii) water mass input from land ice melt and land water reservoirs. Thermal expansion is the physical response of the water mass of the oceans to atmospheric warming. Ocean temperature data collected during the past few decades indicate that ocean thermal expansion has significantly increased during the second half of the 20th century. Thermal expansion accounts for about 25% of the observed SLR since 1960 (Domingues et al., 2008) and about 50% from 1993 to 2003 (for e.g. IPCC 2007, Nicholls and Cazenave 2010). Ice sheets have the largest potential effect, because their complete melting would result in a global sea-level rise of about 70 m. Yet their dynamics are poorly understood, and the key processes that control the response of ice flow to a warming climate are not included in current ice sheet models. The interplay of these factors and their action on different timescales makes understanding of global sea level rise dynamics very difficult. The mechanism of thermal expansion can be gauged relatively accurately through an analysis of global temperatures and their rate of increase through GCMs. However, there is a great uncertainty in predicting the melting rate of the ice sheets and ice caps. Analysis of observed data suggests that the decay rate of ice sheets is non-linear. Simpson et al (2009) further highlights that rapid collapses of ice sheets have occurred previously that bear no correlation, whether contemporaneous or lagged, to any climate forcing that may have triggered it. 5 [page 154] 2.3 Current Trends 2.3.1 Global Using proxy and instrumental data, it is virtually certain (i.e. with 99-100% probability) that the rate of global mean sea level rise has accelerated during the last two centuries, marking the transition from relatively low rates of change during late Holocene (order tenths of mm yr'!) to modern rates (order mm year). Rates and absolute changes in global mean sea level are shown in Table 2.1 Table 2.1 Rates and absolute change in mean sea level. Rates are obtained from IPCC (2013). RE PE LL RE PL Tide-guage and satellite altimeter data both reflect the rate represented in the 1993-2010 period. It is likely that rates similar to this period also occurred between 1930 and 1950. It is also likely that global mean sea level has accelerated since the early 19005, with estimates ranging from 0.000 to 0.013 [-0.002 to 0.019] mm yr? (IPCC, 2013). Accelerations in the rate of increase over the 20" century are have been detected in most regions. See for example Woodworth et al. (2009), and Church and White (2006) 2.3.2 Caribbean Estimates of observed sea level rise from 1950 to 2000 suggest that sea level rise within the Caribbean appears to be near the global mean. Table 2.2 shows the rates of sea level rise for a number of locations in the Caribbean. All values suggest an upward trend. It is important to note that due to shifting surface winds, expansion of warming ocean water and the addition of melting ice, ocean currents can be altered which, in turn leads to changes in sea level that vary from place to place. Additionally more localized processes such as sediment compaction and tectonics may also contribute to additional variations in sea level. 6 [page 155] Table 2.2 Observed rates of sea level rise for some Caribbean stations. Source: The State of the Jamaican Climate (2013). 2.4 Projected Trends 2.4.1 Global and Caribbean Estimates of future global mean sea level were obtained from observations and GCM results reported by IPCC Working Group1 for IPCC Fourth and Fifth Assessment Reports (IPCC 2007, IPCC 2013). According to the Fourth Assessment Report by the end of the century, sea levels are also expected to rise by 0.21m to 0.48m under an A1B (medium emissions) scenario or by 0.26-0.59 m under the highest emissions scenario, A1F1, but the models exclude future rapid dynamical changes in ice flow. One study suggests that the rate of rise may actually double as noted for A1B (Science Daily, Feb. 12, 2008). Higher projections of sea level rise are noted in the IPCC Fifth Assessment Report (ARS) in comparison to the Fourth Assessment Report (ARS). This is considered to be primarily due to the improved modeling of land-ice contributions. There is also higher confidence in the projections of sea level rise in the latter report due to improved understanding of the components of sea level, improved agreement of process-based models with observations, and the inclusion of ice-sheet dynamical changes. Projections and Graph are shown below. In the RCP projections, thermal expansion accounts for 30 to 55% of the 21* century global mean sea level rise, and glaciers for 15 to 35%. AR5 notes that the basis for higher projections of global mean sea level was considered but it was concluded that there is currently insufficient evidence to evaluate the probability of specific levels above the assess likely rate. Finally the point is made in ARS that sea level rise will not be uniform. It is very likely that sea level will rise in more than about 95% of the ocean area. Approximately 70% of the coastlines worldwide are projection to experience sea level change within 20% of the global mean sea level change. It is useful to note that for the SRES A1B which was assessed in AR4, the likely range bases on the science assessed in the ARS is 0.60 [0.41-0.79] m by 2100 relative to 1986-2005 and 0.57 [0.40- 7 [page 156] 0.75]m by 2090-2099 relative to 1990. Compared with the AR4 projection of 0.21-0.48 m for the same scenario and period, the largest increase is from the inclusion of rapid changes in Greenland and Antarctic ice-sheet outflow. Table 2.3 Projected increases in global mean surface temperature and global and Caribbean mean sea level from the IPCC (2007) contrasted with those of Rahmstorf (2007). Projections are by 2100 relative to 1980-1999. Source: CARIBSAVE Climate Change Risk Atlas — Jamaica (2011). Global mean surface | Global mean sea level | Caribbean mean sea temperature (°C) rise (m) level rise (+0.05 m) relative to global mean IPCC B1 1.1-2.9 0.18 — 0.38 0.14 — 0.43 IPCC A1B 17-44 0.21 — 0.48 0.16 —0.53 IPCC A2 2.0-5.4 0.23-0.51 0.18 —- 0.56 Rahmstorf, 2007 Um Upto 14m Upto 14m Table 2.4 Projected increases in global mean surface temperature and global mean sea level. Projections are taken from IPCC (2013) and are relative to 1986-2005. [| 2046-2065 |" | 2081-2100 | | Variable |Scenaro [Mean |Ukelyrenge [Mean | Ukelyrange Global Mean | RCP2.6 0.4-1.6 0.3-1.7 Surface RCP4.5 0.9-—2.0 1.1-2.6 Temperature RCPG.0 08-18 14-31 Changel*c) RCPE.S 14-26 26-48 Likely range Lielyrange Global Mean Sea | RCP2.6 0.17 — 0.32 [0.40 |0.26-0.55 Level Rise (m) RCP4.5 0.19 — 0.33 0.32 —0.63 RCPSO [O2 [018-032 [045 |033-063 RCPES [030 [022-038 [063 [045-027 8 [page 157] 1.0 Mean over 2081-2100 0.8 0.6 E 0.4 8 É 4 $ 02 8 & è A 00 2000 2020 2040 2060 2080 2100 Year Figure 2.1 Projections of global mean sea level rise over the 21st century relative to 1986-2005 from the combination of the CMIPS ensemble with process-based models, for RCP2.6 and RCP8.5. The assessed likely range is shown as a shaded band. The assessed likely ranges for the mean over the period 2081-2100 for all RCP scenarios are given as coloured vertical bars, with the corresponding median value given as a horizontal line. a) 60°N 0.30 30°N 0.24 0.18 0° 0.12 o Poe 0.06 us 0.00 90°E 180° 90°W 0° (m) b) 60°N 0.30 30°N 0.24 ë 0.18 0.12 30°S 0.06 60°sS 90°E 180° 90°W 0° (m0 Figure 2.2 (a) Ensemble mean projection of the time-averaged dynamic and steric! sea level changes for the period 2081-2100 relative to the reference period 1986-2005, computed from 21 CMIPS5 climate models (in m), using the RCP4.5 experiment. The figure includes the globally averaged steric sea level increase of 0.18 + 0.05 m. (b) RMS spread (deviation) of the individual model result around the ensemble mean (m). Source: IPCC ARS. 1The term ‘steric’ pertains to the temperature, salinity and pressure dependent specific volume of the ocean. (Lauderer et al. 2006) 9 [page 158] ACCESS L0 DC CM pES M hs _ _ ec céline MIUA CM éd QûA Ge D.t 5 Lance x == . g 7 a A - 14 L A TEX M2 ce C2 a PER Ven, mn, Ha d'GEM 2 | == —— rs - ——" ps MA ME MRC - Ë HAOC-ESM ee . . F = un _— ?. — as PP A = MIROK Mt. M LA MPLESM MA l ee pu, == S mn, dom ME n Mor£Shi-ME Nas M1-M (m) -0.2 -0.1 0Q O1 0.2 0.3 0.4 Figure 2.3: Projected relative sea level change (in m) from the combined global steric plus dynamic topography and glacier contributions for the RCP4.5 scenario over the period from 1986-2005 to 2081- 2100 for each individual climate model used in the production of Figure 2.2 Source: IPCC ARS. 10 [page 159] 2.5 Observations and Projections of Sea Level Extremes 2.5.1 Observations of sea level extremes Tsunamis aside, extremes in sea level (ï.e. coastal flooding, storm surge, high water events, etc.) tend to be caused by large storms, especially when they occur at times of high tide. In fact any low pressure system off shore with associated high winds can cause a coastal flooding event depending on duration and direction of winds. Table 2.5 highlights several examples of major storms impacting the US east coast and shows the peak pressure, storm size, continental slope and storm surge measurements. The table highlights that the maximum sustained wind (represented by the category of the storm) is not necessarily a good indication of storm surge values, though some correlation exists between the two. Table 2.5: Major storms affecting the east coast of the United Sates from 1970-2005. Adapted from Irish et al. (2008) Storm Date Central Radius to Saffir- Estimated Observed (Name) pressure maximum Simpson influencing open coast (mb) wind (km) Category continental shelf surge (m) slope August 1974 943 28 3 1:2,500 (Carmen) August 1979 950 3 1:1,500-1:1,900 3.5-3.8 (Frederic) August 1992 30 5 1:750-1:1,500 2.4 (Andrew) October 1995 3 1:750-1:1,000 3.1-3.7 (Opal) August 1999 (Bret) 1800-1110 | 09-15 September 2002 28 1 1:4,000-1:7,500 3.2-3.6 (Lili) September 2004 950 19 1:500-1:1,000 2.1 (Charley) September 2004 955 56 3 1:1,500-1:1,900 3.0-3.1 (Ivan) July 2005 Dennis) 175011,500 | 17-25 August 2005 919 47 3 1:5,000- 7.5-8.5 (Katrina) 1:10,000 11 [page 160] me | | Nr rer pe | (Rita) (Wilma) The AR4 noted that highest water levels have been increasing since the 19505 in most regions of the world, caused mainly by increasing mean sea level. Higher regional extremes are also associated with climate fluctuations such as ENSO, the North Atlantic Oscillation and the Atlantic Multidecadal Oscillation, among others. Since the publication of AR4 most analyses have focused on specific regions and find that extreme values have increased since the 19505, using various statistical measures such as annual maximum surge, annual maximum surge-at-high-water, monthly mean high water level, changes in number of high storm surge events or changes in the 99 percentile events. Global analysis of tide gauge data that spans 1970s to the present, also suggests that the magnitude of extreme sea level events has increased in all regions studies. The height of a 50-year flood event has increased anywhere from 2 to more than 10 cm per decade since 1970 although some areas have seen a negative rate because of the vertical land motion is much larger than the rate of mean sea level rise. The height of a 50-year flood event has increased anywhere from 2 to more than 10 cm per decade since 1970 (Figure 2.4a), although some areas have seen a negative rate because vertical land motion is much larger than the rate of mean sea level rise. However, when the annual median height at each gauge is removed to reduce the effect of local mean sea level rise, interannual and decadal fluctuations, and vertical land motion, the rate of extreme sea level change drops in 49% of the gauges to below significance (Figure 2.4b), while at 45% it fell to less than 5 mm yr”. Only 6% of tide gauge records evaluated had a change in the amplitude of more than 5 mm yr ‘after removing mean sea level variations, mainly in the southeast United States, the western Pacific, southeast Asia, and a few locations in northern Europe. The higher rates in the southeastern United States have been linked to larger storm surge events unconnected to global sea level rise (Grinsted et al., 2012, IPCC, 2013). 2.5.2 Projections of Sea Level Extremes Lowe et al. 2010 suggest that increases in the observed sea level extremes in the 20° century and increases projected for the 21° century occur mainly through an increase in mean sea level. Projected changes in storm surges (relative to mean sea level) have been assessed by applying climate-model forcing to storm-surge models. Return periods of sea level extremes, exceeding a given 12 [page 161] threshold level, referred to as return levels, are used in quantifying projected changes. Table 2.6 summarizes the results of studies examining projections of sea level extremes. The results show that projections are sensitive to the choice of global climate model or regional climate model. Table 2.6 Summary of studies examining projections of sea level extremes. Compiled from IPCC (2013). [Too | Scemaño| Reut | Referne | Regionally A2, B2, 8-10% increase in the 99° percentile surge heights | Debernard and Roed downscaled A1B between 1961-90 and 2071-2100, mainly during | (2008) GCM the winter season along the coastlines of the eastern North Sea and the northwestern British Isles, and decreases south of Iceland. Downscaled A1B Significant increase in wintertime storm surges | Wang et al. (2008) GCM around most of Ireland between 1961-1990 and 2031-2060 3 climate A2 Changes in the 95th percentile sea level height | Colberg and Mcinnes models across the southern Australian coast in 2081-2100 | (2012) compared to 1981-2000 were small (+0.1 m), mostly negative, and despite some inter-model differences, resembled the changes in wind patterns simulated by the climate models Numerical For the tropical east coast of Australia, a 10% | e.g. Mcinnes et al. ocean model increase in tropical cyclone intensity for 2050 led to | (2013) increases in the 100-year return level (including tides) that at most locations were smaller than 0.1 m GCM Sea level rise has a greater potential than | Brown et al. (2010) meteorological changes to increase sea level | Woth et al. (2006) extremes by the end of the 21st century for | Lowe et al. (2009) southeastern coast of Australia, eastern Irish Sea, North Sea and the United Kingdom coast RCM A2 The combined effect of MSL rise of 4 mm/yr and | Unnikrishnan et al. RCM projections for winds and atmospheric | (2011) pressure gave an increase in 100-year return levels of total sea level (including tides) between 0.40- 0.67 m (about 15-20%) along the northern part of the east coast of India, except around the head of the bay for 2071-2100 compared to 1961-1990. Numerical The effect of MSL rise on simulated surges was | Smith et al. (2010) surge and wave linear. Yet in the regions of moderate surges (2-3 models m), particularly in wetland-fronted areas, the increase in surges was larger by 1-3 m than the present values. They showed that sea level rise alters the speed of propagation of surges and their amplification in different regions of the coast. 13 [page 162] Statistical The dynamic interaction of surge and sea level rise | Mousavi et al. (2011) model lowered or amplified the surge at different points Within a shallow coastal bay for the Gulf of Mexico. Higher mean sea levels can significantly decrease the return period for exceeding given threshold levels. For a network of 198 tide gauges covering much of the globe, Hunter (2012) determined the factor by which the frequency of sea levels exceeding a given height would be increased for a MSL rise of 0.5 m. The calculations were repeated in the ARS using regional relative sea level projections and their uncertainty using the RCP4.5 scenario. This multiplication factor depends exponentially on the inverse of the Gumbel scale parameter (a factor which describes the statistics of sea level extremes caused by the combination of tides and storm surges) (Coles and Tawn, 1990). The scale parameter is generally large where tides and/or storm surges are large, leading to a small multiplication factor, and vice versa. (a) F mn 4 \ù « (b) 0] L 1 L__ 1] L_] EL] L__] LI} ; LE F WE "TE En ET h Te : 1: 4 none — ms — ms — mr — mur — mr —| -12 40 8 6 -4 -2 O0 2 4 6 8 10 12 cm per decade Figure 2.4: Estimated trends (cm per decade) in the height of a 50-year event in extreme sea level from (a) total elevation and (b) total elevation after removal of annual medians. Black dots indicate trends are not significant at the 95% confidence level. Data are from Menéndez and Woodworth (2010). 14 [page 163] Another useful study is that of Ali (1996). The study showed that storm surge heights increase with an increase in wind speed and SST (Table 2.7). On the other hand, sea level rise tends to reduce the surge heights if wind speed remains constant. It is to be noted here that the model had a fixed boundary and as a result could not simulate the surge height for a moving shoreline. Sea level rise will convert the hitherto land area into a part of the sea which will then become a shallow water area where the surge will be amplified. Thus, although the sea level rise will apparently reduce the surge height in the present sea water, it will increase the surge height in the newly converted (from land to sea) sea area. Table 2.7 Storm surge heights under different sea surface temperature and sea level rise scenarios. (Wind speed of 225 km h°? corresponds to that of the April 1991 cyclone impacting Bangladesh.) Taken from Ali (1996). Surge height in m (% change) Sea level rise = 0.0 m 7.6 (0) 9.2 (21) 11.3 (49) Sea level rise = 0.3 m 7.4 (-3) 9.1 (20) 11.1 (46) Sea level rise =1.0m 7.1 (-7) 8.6 (13) 10.6 (40) Table 2.8 Storm surge under different sea level rise and sea surface temperature rise conditions. (Scenario | represents base condition that corresponds to wind speed and central pressure as observed during the 1991 cyclone) Derived from Emanuel (2005) and Ali (2000). Climate Sea level rise SST rise (°C) Wind speed Central Surge height Scenarios (m) (km/h) pressure (hPa) | (m, mean sea level) | Scenario1 100 7 [0 | 225 926 |76 | [Scenaroll_ [00 [2 [26 9% [9 ScenariollL [00 [4 |24 92 [us _ ScenarolV [03 0 25 19% 74 DSœænañow [03 [4 2% jen us _ Scenario VIE 1.0 0 12% 1% 1 Scenario VI | 10 [2 24 [924 [56 | DSœænarñoix [no [a 2% jen [ws 15 [page 164] Karim and Mimura (2008) examined different scenarios of sea surface temperature and sea level rise increases and the impact they have on storm surge height (Table 2.8). Scenario | represents the base condition, while Scenario V is considered as an average climate condition by 2050. It shows that storm surge height may increase as much as 21% if SST rises by 2 °C (Scenario 11) and 49% if SST rises by 4 °C (Scenario Ill). Another analysis based on continental shelf length and wind speed (Chowdhury, 1994) also produced similar surge heights at the coast. These predictions are relatively large compared with the results of Mitchell et al. (2006) in which they predicted 0.5-0.7m increase in surge height of a 50-year return period storm surge. It is interesting to note that surge height reduces by 7% if sea level rise by 1.0m, but SST remains unchanged (Scenario VII). The reason is that the amplification of the surge is less if the water depth is increased, due to the differences in bottom friction on the propagating waves. 2.6 Uncertainties The range of uncertainty cones from a combination of emissions uncertainty and methodology uncertainty as follows: e Emissions uncertainty arises because the future is uncertain and there are a range of plausible futures (with respect to for example technologies and energy requirement). This is addressed by presenting results for a range of projections that attempt to cover the range of plausible futures. e Methodological uncertainty arises because of the inability to create perfect models of the climate system even if future emissions were known. Additionally some processes depend on parameters that have not yet been accurately measured or one may discover are entirely missing from the models. This uncertainty is addressed by either running simple models for a range of climate sensitivities or by looking at a range of values from the few available complex climate models. 16 [page 165] 3. Tropical Cyclones 3.1 Hurricane Metrics Several publications and reports outline the damage sustained by countries due to hurricanes during the annual North Atlantic hurricane season throughout history. It is not surprising then, the initiative shown by several institutions to characterize and, thus, predict the yearly trend. Institutions such as the National Oceanic and Atmospheric Administration (NOAA) have led research into hurricane dynamics and have developed several important tools and datasets that are used today to characterize hurricane activity. This report highlights two metrics for intensity (accumulated cyclone energy [ACE], power dissipation index [PDI]). In addition to intensity, it should be noted that the structure and areal extent of the wind field in tropical cyclones is largely independent of intensity and play an important role on potential impacts, particularly from storm surge (Irish and Resio, 2010), but measures of storm size are largely absent in historical data 3.1.1 Accumulated Cyclone Energy (ACE) The ACE proposed by the National Oceanic and Atmospheric Administration (NOAA) is defined as the sum of squares of the maximum sustained wind speed in knots (v..,) measured every six hours for all named storms while they are at least tropical storm strength (Bell et al. 2000). The ACE is calculated by the formula below and is of the order of 10“ kt2. (1) NOAA classifies a season as above-normal, normal or below-normal as illustrated in Figure.3.1: e _ Above-normal: above 111x10° kt? (corresponding to 120% of the 1981-2010 median), 13 or more named storms, 4 or more hurricanes, 3 or more major hurricanes ° _Below-normal: below 66x10° kt° (corresponding to less than 71.4% of the 1981-2010 median), 9 or fewer named storms, 4 or fewer hurricanes, 1 or fewer major hurricanes. ° _Near-normal: ACE values lies in the range of 66x10° - 111x10° kt° (between 71.4%-120% of the 1981-2010 median), 10-15 named storms, 4-9 hurricanes, 1-4 major hurricanes. Klotzbach (2006) suggests that the ACE is proportional to the kinetic energy generated by the storm. 17 [page 166] 300 1981-2010 median = 98.06x10% k2 “| © 200 à | uJ Q 165 AIRNESS |A 11 = | 120 L 100 & HI URRA I L'IR [11 11A " 950 1960 1970 1980 1990 2000 2010 Figure 3.1 Showing the total ACE value per year as a percentage of the 1981-2010 median for the period 1950-2012. Years falling above 120% are considered to be above-normal, years falling in between 71 - 120% are considered to be normal activity, while years falling below an 71% are considered to have below-normal activity. Years exceeding 165% is deemed to be very active. 3.1.2 Power Dissipation Index (PDI) Emanuel (2005) introduced the power dissipation index (PDI), the cubed of the maximum sustained wind speed at the standard altitude of 10 m, integrated over the lifetime of the storm. Theory provides that a hurricane loses power at a rate per unit area of (2) where p is the air density, C is the surface drag coefficient and V is the surface wind speed (Emanuel, 1998). Assuming a circularly symmetric hurricane, the dissipative rate D is integrated over the radius (from the center to an outer limit ro) and over the lifetime t of the storm to give the total power dissipation PD, shown below. (3) The equation is further simplified due to unavailability of accurate records of hurricane radii and the variability of C, and p within the radius of maximum wind speeds. Another limitation is that there is very little relation between the peak winds and the size of the hurricane (Emanuel, 2005), and so may 18 [page 167] provide a less than accurate picture of the intensity of the storm. Taking the values for the drag coefficient and sea-level air density to be constant, the final equation arises as: (4) It is however important to note that both ACE and PDI are based on the assumption that velocity varies linearly with radius inside the radius of maximum winds. Yu et al (2009) point out that the radial wind structure within a tropical cyclone varies and that the formulae proposed for the two indices results in an overestimation of the total activity. For this reason, both indices were revised (Yu et al. 2009; Yu and Chiu, 2012). The revised ACE (RACE) and revised PDI (RPDI) indices are constructed based on a modified Rankine vortex structure proposed by Yu et al(2009). The algorithm outlining their construction is described in detail in Appendix A. The RACE index and RPDI index are defined by equations (5) and (6), respectively. TE (5) —- ——— (6) where is the cut-off radius (dimensionless) in which the wind energy is measured and is the decaying tendency of wind beyond the radius of maximum wind. Like the ACE and PDI indices, the revised versions are strongly influenced by strong tropical cyclones. This is due to the inclusion of the intensity and duration of the storms, and is thus a better indicator of the strength of the storms within a given season. Yu and Chiu (2009) concludes that the ACE, PDI, RACE and RPDI indices may not give a clear trend in frequency. These metrics have recently become a popular tool of characterizing overall hurricane activity. However, no publications as of yet outlines the use of the revised metrics in projections of future tropical cyclone activity. Additionally, Klotzbach (2006) notes that ACE and PDI correlate globally at 0.97. Therefore to examine the trend in either index at the global level should yield virtually the same results. 19 [page 168] 3.2 Current trends in North Atlantic Hurricane intensities Intensity measures in historical records are especially sensitive to changing technology and improving methodology. Over the satellite era however, increases in the intensity of the strongest storms in the Atlantic are quite robust (Kossin et al., 2007; Elsner et al. 2008). Time series of cyclone indices such as power dissipation show upward trends in the North Atlantic since the late 1970s (Emmanuel, 2007) but interpretation of longer-term trends is constrained by data quality concerns (Landsea et al., 2012). The ARS notes that evidence suggests a virtually certain increase in the frequency and intensity of the strongest cyclones in the Atlantic since the 1970s. Wu et al. (2008) suggest that the magnitude of the statistically significant linear trend in PDI over 1975-2004 is 0.024 m° s° year’. Itis further noted that the average lifetime of North Atlantic tropical cyclones show an increasing trend of 0.07 day year” for the same period which is statistically significant. The variability and trend in power dissipation can be related to sea surface temperature and other local factors such as tropopause temperature and vertical wind shear (Emanuel, 2007), but there is debate to whether local sea surface temperature or the difference between local sea surface temperature and mean tropical sea surface temperature is the more physically relevant metric (Swanson, 2008). The distinction becomes important when making projections of changes in power dissipation based on projections of SST changes particularly in the tropical Atlantic where SST has been increasing more rapidly than in the tropics as a whole (Vecchi et al., 2008). ACE has shown declines globally since reaching a high point in 2005, and is presently at a 40 year low point (Maue, 2009) (IPCC SREX 2012). 3.3 Projected trends in North Atlantic frequencies and intensities The AR4 concluded that a range of modeling studies project a likely increase in peak wind intensity and near storm precipitation in future tropical cyclones. Simulations with high resolution dynamical models (e.g. Oouchi et al., 2006; Bengtsson et al., 2007; Gualdi et al., 2008; Knutson et al. 2008; Sugi et al., 2009; Bender et al., 2010) and statistical-dynamical models (Emanuel, 2007) consistently find that greenhouse warming causes tropical cyclone intensity to shift towards stronger storms by the end of the 21* century (2 to 11% increase in mean maximum wind globally). Applying 21* century sea surface temperature projections to a relationship between local SST and tropical cyclone power dissipation (constructed by Emmanuel, 2007), power dissipation is projected to increase by about 300% in the next century (Vecchi et al., 2008; Knutson et al., 2010). Alternatively 20 [page 169] when using a similar strong relationship between power dissipation and relative SST (which represents the difference between local and tropical mean SST), projections indicate almost no change in power dissipation in the next century (Vecchi et al., 2006). Both relationships can be reasonably defended based on physical arguments but it is not clear which, if either is current (Ramsay and Sobel, 2011). When simulating 21* century warming under A1B, the present models and downscaling techniques suggest increases in intensity and fraction increases in the number of most intense storms. The frequency of the most intense storms is more likely than not to increase by more than +10% (IPCC 2013, ARS), while the annual frequency of tropical cyclones are projected to decrease or remain relatively unchanged for the North Atlantic (See Figure 3.2). Of concern however is the limited ability of global models to accurately simulate upper-tropospheric wind (Cordero and Forster, 2006) which modulates vertical wind shear and tropical cyclone genesis and intensity evolution. Knutson et al (2013) conclude that there is likely to be an increase in radius-averaged precipitation rainfall rates near the hurricane core. The conclusion follows from an expected increase in atmospheric water-vapour content, and thus an increase in moisture convergence that results in convective systems, such as hurricanes. Increases of approximately 20% are projected for radii within 100km from the hurricane core. However, an accurate quantification of positive changes for smaller radii prove to be difficult due to complex dynamics near the centre of the hurricane (Knutson et al, 2010). 21 [page 170] Western North Pacific . North Atlantic 50 \ A / A20% North Indian 8 Eastern North Pacific 1 5 80 —_ £ 0 el. : HS re 50 CN : 28 de — — 8 .e ë = Se 3 Ua Ë 0 —#—— SS à di. NON “50 UN mo W -50 1 [l LL W l ft in 12 South Indian 2%, ERP ; 5 50 : É F 0 6 2 Tropical Cyclone (TC) Metrics: Le -50 1 ANTC frequency ù IL Category 4-5 TC frequency | MN VON OM ON IL Lifetime Maximum Intensity. IV Précipitation rate SOUTHERN HEMISPHERE GLOBAL NORTHERN HEMISPHERE 50 50 50 & & ® ô 0 Ê 0 Ë 0 . _ LEA 1 " LL W ! [l LL N | {] LL IN Figure 3.2 Projected changes in tropical cyclone statistics. All values represent expected change in the average over period 2081-2100 relative to 2000-2019, under an A1B-like scenario, based on expert judgment after subjective normalisation of the model projections. Four metrics were considered: the percent change in 1) the total annual frequency of tropical storms, 11) the annual frequency of Category 4 and 5 storms, Ill) the mean Lifetime Maximum intensity (LMI; the maximum intensity achieved during a storm's lifetime), and IV) the precipitation rate within 200km of storm center at the time of LMI. For each metric plotted, the solid blue line is the best guess of the expected percent change, and the coloured bar provides the 67% (likely) confidence interval for this value (note that this interval ranges across -100% to +200% for the annual frequency of Category 4 and 5 storms in the North Atlantic). Where a metric is not plotted, there is insufficient data (denoted "X") available to complete an assessment. À randomly drawn (and coloured) selection of historical storm tracks are underlaid to identify regions of tropical cyclone activity. Source: IPCC ARS. 22 [page 171] 4. Some References Ali, À. (1996). Vulnerability of Bangladesh to climate change and sea level rise through tropical cyclones and storm surges. In Climate Change Vulnerability and Adaptation in Asia and the Pacific (pp. 171-179). Springer Netherlands. Ali, A. (2000). Climate change impacts and adaptation assessment in Bangladesh. Climate Research, 12(2-3), 109-116. Bender, M.A., Knutson, T.R., Tuleya, R.E., Sirutis, J.J., Vecchi, G.A., Garner, S.T., Held, I.M. (2010). Modeled Impact of Anthropogenic Warming on the Frequency of Intense Atlantic Hurricanes, Science 327, 454-458, doi:10.1126/science.1180568. Bengtsson, L., Hodges, K. I., Esch, M., Keenlyside, N., Kornblueh, L., LUO, J. J., & Yamagata, T. (2007). How may tropical cyclones change in a warmer climate?. Tellus À, 59(4), 539-561. Brown, J. M., & Wolf, J. (2009). Coupled wave and surge modelling for the eastern Irish Sea and implications for model wind-stress. Continental Shelf Research, 29(10), 1329-1342. Boldingh Debernard, J., & Petter Rsed, L. (2008). Future wind, wave and storm surge climate in the Northern Seas: a revisit. Tellus A, 60(3), 427-438. Campbell, J.D., Taylor, MA. Stephenson, T.S., Watson, R.A., Whyte, F.S. (2010). Future climate of the Caribbean from a regional climate model, International Journal of Climatology 31, 1866-1878, doi:10.1002/joc.2200. Chowdhury, J. U. (1994). Determination of shelter height in a storm surge flood risk area of Bangladesh Coast. Water Resources Journal, 182, 93-99. Colberg, F., & Mcinnes, K. L. (2012). The impact of storminess changes on extreme sea levels over southern Australia. JGR-Oceans, doi, 10. Coles, S. G., & Tawn, J. A. (1990). Statistics of coastal flood prevention. Philosophical Transactions of the Royal Society of London. Series A: Physical and Engineering Sciences, 332(1627), 457-476. Cooper, J. A. G., & Pilkey, O. H. (2004). Sea-level rise and shoreline retreat: time to abandon the Bruun Rule. Global and planetary change, 43(3), 157-171. Cordero, E. C., & Forster, P. D. F. (2006). Stratospheric variability and trends in models used for the IPCC AR4. Atmospheric Chemistry and Physics, 6(12), 5369-5380. Domingues, C. M., Church, J. A., White, N. J., Gleckler, P. J., Wijffels, S. E., Barker, P. M., & Dunn, J.R. (2008). Improved estimates of upper-ocean warming and multi-decadal sea-level rise. Nature, 453(7198), 1090-1093. Elsner, J. B., Kossin, J. P., & Jagger, T. H. (2008). The increasing intensity of the strongest tropical cyclones. Nature, 455(7209), 92-95. Emanuel, K.A. (1998). The power of a hurricane: An example of reckless driving on the information superhighway. Weather 54, 107-108. 23 [page 172] Emanuel, K.A. (2005). /ncreasing destructiveness of tropical cyclones over the past 30 years. Nature 436, 686-688, doi:10.1038/nature03906. Fish, MR., Côté, I.M., Gill, J.A., Jones, A. P., Renshoff, S. and Watkinson, A. R (2005). Predicting the impact of sea level rise on Caribbean sea turtle nesting habitat, Conservation Biology, 19, 482- 491. Goldenberg, S.B., Landsea, C.W., Mestas-Nuñez, A.M., Gray, W.M. (2001). The recent increase in Atlantic Hurricane Activity: Causes and Implications, Science 293, 474-478., Gualdi, S., Scoccimarro, E., & Navarra, A. (2008). Changes in tropical cyclone activity due to global warming: Results from a high-resolution coupled general circulation model. Journal of climate, 21(20), 5204-5228. Hunter, J. (2012). À simple technique for estimating an allowance for uncertain sea-level rise. Climatic change, 113(2), 239-252. IPCC, (2012). Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation. À Special Report of Working Groups 1! and I! of the Intergovernmental Panel on Climate Change [Field, C.B., V. Barros, T.F. Stocker, D. Qin, D.J. Dokken, K.L. Ebi, M.D. Mastrandrea, K.J. Mach, G.-K. Plattner, S.K. Allen, M. Tignor, and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, UK, and New York, NY, USA, 582 pp. IPCC, (2013). Climate Change 2013: The Physical Science Basis. Contribution of Working Group | to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T. F., D. Qin, G.-K. Plattner, M. Tignor, S. K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P. M. Midgley (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, in press Irish, J.L,, Resio, D.T., Ratcliff, J.J. (2008). The influence of storm size on hurricane surge, Journal of Physical Oceanography 38, 2003-2013, doi: 10.1175/2008JP03727.1. Irish, J. L, & Resio, D. T. (2010). A hydrodynamics-based surge scale for hurricanes. Ocean Engineering, 37(1), 69-81. Karim, M. F., & Mimura, N. (2008). Impacts of climate change and sea-level rise on cyclonic storm surge floods in Bangladesh. Global Environmental Change, 18(3), 490-500. Knutson, T. R., Sirutis, J. J., Garner, S. T., Vecchi, G. A., & Held, I. M. (2008). Simulated reduction in Atlantic hurricane frequency under twenty-first-century warming conditions. Nature Geoscience, 1(6), 359-364. Knutson, T.R., et al., (2010). Tropical cyclones and climate change. Nature Geoscience, 3, 157-163. Knutson, T. R., et al., (2013). Dynamical downscaling projections of 21st century Atlantic hurricane activity: CMIP3 and CMIP5 model-based scenarios. Journal of Climate, in press. 24 [page 173] Kossin, J. P., Knapp, K. R., Vimont, D. J., Murnane, R. J., & Harper, B. A. (2007). A globally consistent reanalysis of hurricane variability and trends.Geophysical Research Letters, 34(4). Landsea, C. W., Feuer, S., Hagen, A., Glenn, D. A. Sims, J., Perez, R., … & Anderson, N. (2012). A Reanalysis of the 1921-30 Atlantic Hurricane Database*. Journal of Climate, 25(3), 865-885. Lowe, J., Howard, T., Pardaens, A. Tinker, J., Holt, J., Wakelin, S., … & Bradley, S. (2009). UK Climate Projections science report: Marine and coastal projections. Mann, M.E., Emanuel, K.A. (2006). Atlantic Hurricane trends Linked to Climate Change, EOS 87 (24), 233- 244. Maue, R. N. (2009). Northern Hemisphere tropical cyclone activity. Geophysical Research Letters, 36(5). Mcinnes, K. L., Macadam, l., Hubbert, G., & O'Grady, J. (2013). An assessment of current and future vulnerability to coastal inundation due to sea-level extremes in Victoria, southeast Australia. International Journal of Climatology, 33(1), 33-47. McKenzie, A. (2012). Beach Responses to Hurricane Impacts: À Case Study of Long Bay Beach, Negril, Jamaica. Caribbean Journal of Earth Science, 43, 51-58. Meehl, G.A. T.F. Stocker, W.D. Collins, P. Friedlingstein, A.T. Gaye, J.M. Gregory, A. Kitoh, R. Knutti, J.M. Murphy, A. Noda, S.C.B. Raper, 1.G. Watterson, A.J. Weaver and Z.-C. Zhao, (2007). Global Climate Projections. In: Climate Change 2007: The Physical Science Basis. Contribution of Working Group | to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Menéndez, M., & Woodworth, P. L. (2010). Changes in extreme high water levels based on a quasi-global tide-gauge data set.Journal of Geophysical Research: Oceans (1978- 2012), 115(C10). Mitchell, J. F., Lowe, J., Wood, R. A., & Vellinga, M. (2006). Extreme events due to human-induced climate change. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 364(1845), 2117-2133. Mori, N., Yasuda, T., Mase, H., Tom, T., & Oku, Y. (2010). Projection of extreme wave climate change under global warming. Hydrological Research Letters, 4(0), 15-19. Mousavi, M. E., Irish, J. L., Frey, A. E., Olivera, F., & Edge, B. L. (2011). Global warming and hurricanes: the potential impact of hurricane intensification and sea level rise on coastal flooding. Climatic Change, 104(3-4), 575-597. Nakicenovic, N., Alcamo, J., Davis, G., De Vries, B., Fenhann, J., Gaffin, S., … & Dadi, Z. (2000). Emissions scenarios. Nicholls, R. J. (1998). Assessing erosion of sandy beaches due to sea-level rise. Geological Society, London, Engineering Geology Special Publications,15(1), 71-76. Nicholls, R. J., & Cazenave, A. (2010). Sea-level rise and its impact on coastal zones. science, 328(5985), 1517-1520. 25 [page 174] Oouchi, K., Yoshimura, J., Yoshimura, H., Mizuta, R., Kusunoki, S., & Noda, A. (2006). Tropical cyclone climatology in a global-warming climate as simulated in a 20 km-mesh global atmospheric model: Frequency and wind intensity analyses. J.Meteorol.Soc.Jpn, 84(2), 259-276. Rahmstorf, S. (2007). A Semi-Empirical approach to projecting future sea-level rise, Science 315, 368- 370, doi:10.1126/science.1135456. Ramsay, H. À. & Sobel, À. H. (2011). Effects of relative and absolute sea surface temperature on tropical cyclone potential intensity using a single-column model. Journal of Climate, 24(1), 183-193. Simpson, M.C., Scott, D. New, M., Sim, R., Smith, D.,Harrison, M., Eakin, C.M., Warrick, R. Strong, A.E.,Kouwenhoven, P., Harrison, S., Wilson, M., Nelson, G.C., Donner, S., Kay, R.,Geldhill, D.K., Liu, G. Morgan, J.A., Kleypas, J.A., Mumby, P.J., Christensen, T.R.L., Baskett, M.L,, Skirving, W.J., Elrick, C., Taylor, M. Bell, J., Rutty, M. Burnett, J.B., Overmas, M., Robertson, R. and Stager, H., (2009) An Overview of Modeling Climate Change Impacts in the Caribbean Region with contribution from the Pacific Islands, United Nations Development Programme (UNDP), Barbados, West Indies Simpson, M. C., Clarke, J. F., Scott, D. J., New, M., Karmalkar, A., Day, O. J., Taylor, M. Gossling, S., Wilson, M. Chadee, D. Stager, H., Waithe, R., Stewart, AÀ., Georges, J., Hutchinson, N., Fields, N., Sim, R., Rutty, M., Matthews, L., and Charles, S. (2012). CARIBSAVE Climate Change Risk Atlas (CCCRA) - Jamaica. DFID, AusAID and The CARIBSAVE Partnership, Barbados, West Indies Smith, J. M., Cialone, M. A., Wamsley, T. V., & McAlpin, T. O. (2010). Potential impact of sea level rise on coastal surges in southeast Louisiana. Ocean Engineering, 37(1), 37-47. Sugi, M., Murakami, H., & Yoshimura, J. (2009). A reduction in global tropical cyclone frequency due to global warming. Sola, 5(0), 164-167. University at Buffalo, (2008, February 12). Global Warming: Sea Level Rise Could be Twice As High as Current Projections, Greenland Ice Sheet Study Suggests. ScienceDaily. Retrieved from November 30,2013, from http://www.sciencedaily.com/release/2008/02/080211172517.htm Unnikrishnan, A. S., RameshKumar, M.R., & Sindhu, B. (2011). Tropical cyclones in the Bay of Bengal and extreme sea-level projections along the east coast of India in a future climate scenario. Current Science, 101(3), 327-331. Van Vuuren, D. P., Edmondés, J., Kainuma, M., Riahi, K., Thomson, A., Hibbard, K., … & Rose, S. K. (2011). The representative concentration pathways: an overview. Climatic Change, 109(1-2), 5-31. Vecchi, G. A. B. J. Soden, A. T. Wittenberg, I. M. Held, A. Leetmaa, and M. J. Harrison, (2006). Weakening of tropical Pacific atmospheric circulation due to anthropogenic forcing. Nature, 441, 73-76. Vecchi, G.A., Swanson, K.L., and Soden, B.J. (2008). Whither Hurricane Activity?, 322 (October). Villarini, G., Vecchi, G.A. (2012b). Twenty-first-century projections of North Atlantic tropical storms from CMIP5 models, Nature Climate Change (online publication), doi: 10.1038/NCLIMATE1530. Villarini, G., Vecchi, G.A. (2013). Projected increases in North Atlantic tropical cyclone intensities from CMIP5 models, Journal of Climate 26, 3231-3240. 26 [page 175] Wang, S., McGrath, R., Hanafin, J., Lynch, P., Semmiler, T., & Nolan, P. (2008). The impact of climate change on storm surges over Irish waters. Ocean Modelling, 25(1), 83-94. Webster, P.J., Holland, G.J, Curry, J.A., Chang, H.-R. (2005). Changes in Tropical Cyclone Number, Duration, and intensity in a Warming Environment, Science 309, 1844-1846, doi:10.1126/science.1116448 Woth, K., Weisse, R., & von Storch, H. (2006). Climate change and North Sea storm surge extremes: an ensemble study of storm surge extremes expected in a changed climate projected by four different regional climate models. Ocean Dynamics, 56(1), 3-15. Yu, J.Y., C.Chou, and P.G. Chiu. (2009) A revised accumulated cyclone energy index. Geophys.Res.Lett. 36, 114710, doi:10.1029/2009GL039254. Yu, J.Y. Chiu, P.G. (2012). Contrasting various metrics for measuring tropical cyclone activity, Terr.Atmos.Ocean.Sci 23 (3), 303-316, doi:10.3319/TA0.2011.11.23.01(A). 27 [page 176] APPENDIX 4: Hazard Profiles EMERGING — c<# GIDB À NORTHERN DEVELOPMENT CORRIDOR, HAITI Appendix Mtts ERM [page 177] A4 HAZARD PROFILES A4.1 Introduction The hazard profile in the pursuant sections characterizes each of the hazards in the Area of Study. A brief outline of each hazard and its potential manifestation within the area of study is presented. A description of the main elements utilized to determine or characterize each hazard is provided as is a map that delineates the spatial extent of the hazard to identify hazard prone areas within the study area. The incorporation of long- term meteorological data from regional climate change models is used to better understand the potential impact of climate variability on natural hazards. The distinction of natural hazards must be made between those hazards that are potentially affected by climate change and those that are not. In general, all hazards that are of hydro-meteorological origin are potentially affected by climate change, while geo-hazards are generally not influenced by climate variability. Table A4.1 provides a characterization of hazards identified for this study effort. Table A4.1 Categorization of Natural Hazards Natural Hazards Affected by Climate Change È [sm | 2 on £ S S © à © In-land + NN È £ Flooding £ Ÿ Coastal Flooding Source: Revisado de Schmidt-Thomé 2006 It is necessary to note that regional models provide generalized understanding of changes to precipitation, temperature and sea level rise, which are critical inputs to applicable hazard models (coastal flood, in-land flood, hurricane wind, and drought). All hazards have been determined or mapped using the best available data. Hazard maps, where applicable, are developed to identify the areas of general susceptibility. The hazard mapping utilizes a qualitative classification scheme that identifies hazard prone areas as very low, low, moderate, high and very high. A4.2 Seismic Hazard Context An earthquake is caused by a sudden motion or trembling of the earth due to an abrupt release of stored energy in the rocks beneath the earth’s surface. When stresses due to underground tectonic forces exceed the strength of the rocks, they will abruptly break apart or shift along existing faults. The energy released from this process results in vibrations known as seismic waves that are responsible for the trembling and shaking of the ground during an earthquake. Earthquakes are also caused by tremendous rock slides that occur along the ocean floor. 3 EMERGING Lo CHE GIDB \ Initiative ERM 1 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 178] The seismic hazard in Haïti has its origin in the interaction of the North American and Caribbean plates (Figure A4.1), which have a relative eastward movement of 2 cm/year (20 mm/yr). un] cusall || TL s Bahama Platform : S Cri à CRETE …. Oceanic | Les Transpression | Oblique collision subduction 1 “5 LE ee DEN | . Fm SHISP Eu PR 5 2] ‘ Transtengon OO So To A D) jo ë > ; à ù Ne 5 ) : + Strike-slip Predicted: 20 Predicted: 20 e Pi É S — e ce Observed: 21 Observed: 19 Ci . " 2 AVES. LA SE à Predicted: 21 à He | à DR Observed: 20.5 # + ss : OTsANA: TA Muus # NS Predicted 21 5 Lg « x : cd < mA, Observer 23 ff Ve ST 10" : . « u | ns L x 4 How is Caribbean plate motion Dominately margin- me, pe oned'n the Hispaniola perpendicular thrust motion? Dé H |; 4 Pré Dominately margin-parallel Predicted GPS vector V . strike-slip motion? Caribbean plate vector Observed GPS vector from DeMets et al. (2000) Es 0° 90° 60° +70 0° Figure A4.1 Relative vectorial displacement of the Caribbean tectonic plate Source: (Calais; 2001) The island of Hispaniola is considered a complex area of deformation which presents both subduction zones off the northern and southeastern coast and strike-slip fault zones that transect the northern and southern portion (Figure A4.2). It also has thrust faults within the island (Frankel et. al., 2010). The strike-slip component of the motion is due to the eastward movement of the Caribbean Plate relative to the North American Plate. On Hispaniola, the majority of the strike-slip plate motion is accommodated across two major features: the Septentrional fault zone, which runs across the northern boundary of the island, and the Enriquillo-Plantain Garden fault zone, which extends from southern-central Hispaniola to Jamaica. The location and characteristics of the significant seismic event that occurred in Haiti in January 2010 indicate that it occurred on a segment of the Enriquillo-Plantain Garden fault zone (RMS FAQ, 2010). The Matheaux Neiba Fault is a thrust fault that underlies the mountain ranges of Haiti. = EMERGING = Lo CR SUSTAINABLE “IDB EU ERM 2 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 179] 75 74 73 72 DA 70 69 -68° ar — 21° N 9hem Hispa D. Fe INiOIQ SUbdur+: 20° __ segment on” à 2OUndany One 20° e — sl Mrionai Foür—… m£ Haiti ‘ — 19 EL Eur 19" segment boundary ton, Dominican Republic RER Enriqui au N LS \ 18 « Muer —ertos — Tough 17 —_— | 7 segment boundary 16° 16 +75 -74° 73 72 LA -70° 69" 68 Figure A4.2 Crustal faults and subduction zones used in the hazard maps. Source: USGS, 2010; note: Red lines denote the portions of the Septentional and Enriquillo Faults. Green lines denote the western portions of the Septentrional and Enriquillo Faults that are treated separately. Blue lines denote the traces of the uppermost portion of the subduction zone faults considered. Locations of the inferred segment boundaries used in the hazard maps are marked by arrows. Active subduction zones are located off the northern and southern coasts of Hispaniola. The Northern Hispaniola subduction zone has produced a series of powerful earthquakes from 1946 and 1953 (USGS, 2010). The focal mechanisms for these earthquakes indicate a southwestward subduction of the North American Plate. The Muertos Trough subduction zone is located south of Hispaniola and extends eastward to south of Puerto Rico and there is evidence that this zone ruptured in 1751 in a large earthquake and produced a tsunami (McCann, 2006). There is a verifiable record of earthquake occurrences dating back more than 500 years in the Caribbean (Table A4.2). In general, the occurrence of seismic events in Haiti has been poorly recorded. A review of the information available has indicated that since 1750 the following major events have occurred: Table A4.2 The most important seismic events in the island of Hispaniola [pate | Magnitude Cities affected |__| 1564 Conception de la Vega, Santo Domingo [| 1684 Santo Domingo, Azua 15 Sept 1751 Port-au-Prince, Santo Domingo, Âzua 1770 Port-au-Prince, Léogane, 250 killed z- EMERGING Lo «CF GIDB Initiative ERM 3 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 180] Joste [magnitude [omesateæes [rMay fige? [80 [Porauprne [&may[ita2 [69 |capaien, 25008 Source: After McCann 2001; Calais 2001 Note: The impacts of these seismic events can be wide-ranging and therefore this table includes key events across Haiti, not just those for the north. Hazard Determination Seismic hazard maps were developed by USGS in response to the urgent need for seismic hazard information as a result of from the catastrophic earthquake of January 2010. The probabilistic maps were assembled based on the current available information on historical and instrumental seismicity and followed the general methodology developed for the 1996 U.S. national seismic hazard maps (Frankel and others, 2000). The methodology provided in the USGS report authored by Frankel et. al, 2010 is concisely summarized in the sections below, and consisted of adding seismic hazard calculated from crustal faults, subduction zones and spatially smoothed seismicity for shallow earthquakes and Wadati-Benioff- zone earthquakes (Frankel et al., 2010). Faults The fault zones included in the seismic hazard model are the Septentrional fault, Enriquillo-Plantain Garden fault zone, and the Matheux Neiba fault (see Figure A4.2). The Septentrional Fault and the Enriquillo-Plantain Garden faults are crustal faults, while the Matheaux Neiba fault is a thrust fault. For each fault, a frequency- magnitude distribution was applied to account for the random uncertainty in accounting for the magnitude of future earthquakes. Seismic moment rates were estimated for each fault from its estimated slip rate, segment length and width. Maximum magnitude of rupture were determined from the segment lengths and empirical relation between surface rupture length and moment magnitude were utilized following Wells and Coppersmith (1994). Subduction Zones The Northern Hispaniola and the Muertos Trough subduction zones were considered in the hazard model. The Northern Hispaniola subduction zone is thought to continue along the entire coast of Hispaniola. The Muertos Trough subduction zone is located south of Hispaniola and extends eastward to Puerto Rico. It appears that the Enriquillo fault merges into this trough. Spatially Smoothed Seismicity Model The spatial smoothed seismicity model, which was developed by Frankel (1995), was utilized to determine the earthquake hazard for Hispaniola. This assumes that future moderate and large earthquakes will occur near areas that have had significant historic seismic activity in the past. The analysis of historic seismicity is based on a review of background source zones and a review of historical and observed data. The model also utilizes attenuation relations (ground motion prediction equations) for each of the source zones. == EMERGING + Lo CR SUSTAINABLE “IDB Initiative ERM 4 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 181] Slip rates and recurrence rates were determined for each fault segment and subduction zone. A historical catalog, derived from U.S. Geological Survey PDE catalog, the Engdahl and Villasensor Catalog (2002) and the International Seismic Cennter (ISC) catalog, was reviewed and divided into thresholds based on the completeness of data! The seismicity was then further divided into depth ranges. Three depth ranges were utilized to develop a seismicity rate grid: 0-40km, 41-100km, and 101km and deeper). The data was combined and interpolated using Gaussian distribution methods and integrated into a mapping grid to calculate the hazard. The results were then combined with attenuation relations (ground motion prediction equations) for each of the crustal faults and subduction zones to determine the seismic hazard for Hispaniola for firm rock site conditions and with site amplification. Hazard Maps The published USGS hazard maps based on fault slip rates and historical and instrumental seismicity have been used to generate the contours having Peak Horizontal Ground Acceleration (PGA) Rock (firm-rock conditions) values for Haiti (Frankel and others, 2010). These hazard maps are for PGA (percent g) for 10 percent and 2 percent probabilities of exceedance (PE) in 50 years, respectively (Figure A4.3 and Figure A4.4). 75 74 73 72 71 70" -69 -68" 21 = — #4 À ” £a e F7 9 20° ET gr 20° 20 22) re LY es 80 - 60 #4 40 19° 19° 30 x > 25 = CE 10 20 15 £iL É 10 3 p 9 18° + . ; t 1e Li ®, 7 Y Ë 4 3 17 0 L:0} -17 2 1 | -75° 74 73 72 7 -70° -69" 68" Figure A4.3 PGA (% g) with 10% probability of exceedance in 50 years (475 years return periods) 1 Complete 1960 > M4.0 and 1915 for larger events. 3 EMERGING Lo À initistie ERM 5 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 182] 75° 74 73 72 LA 70° -69" 68 21 ar [0] ol Lo %9 20° s 20° ra Ê 180 100 80 60 7. 40 19° 102 Le 19° 30 11 25 20 FT 15 LE ri s 10 2 k 9 18 *, 18 , Fe. { 6 À 5 o4 4 17 9 & FLA 2 1 | u am NN ” 75 74 75 72 1 -70° -69 8" Figure A4.4 PGA (% g) with 2% probability of exceedance in 50 years (2500 years return periods) In the data conversion process, these maps have been first geo-referenced using GIS tools and then the contours present in these maps have been captured as polyline features with necessary attribute values associated with them. The contours are then processed using suitable interpolation techniques and PGA Rock data are distributed in raster grid format for the study area. The following figures (Figure A4.5 and Figure A4.6) present the PGA at Rock level developed by USGS. “À “+ Ê F € mn 1 \ / À | { à A \ M Legend Q "4 CI Haiti à 4 PGA Rock 475 RP Vo, / Value 4 { Le 0.388074 N Low : 0.235468 Figure A4.5 Distribution of PGA (in g) with 10% probability of exceedance in 50 years (475 years return periods) for the study area eZ EMERGING Lo CE GIDB \ Initiative ERM 6 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 183] ( { BR. \ { L. d Ne. + | Ÿ { \ 7. Legend à. / CI Haïti à. 7 PGA Rock 2500 RP ES / Value =... \ L. js 0.775157 Ÿ Low : 0.416854 Figure A4.6 Distribution of PGA (in g) with 2% probability of exceedance in 50 years (2500 years return periods) To validate the results initially, ERM has performed a comparison of the USGS PGA values with data derived from the UN Global study for the study area. It has been observed that the distribution and range of PGA values have similarity for the study area, e.g. the range of PGA (in g) values found in USGS data varies between 0.235 and 0.389, whereas, the UN Global data values vary between 0.402 and 0.419 for 475 years return period. Similarly, in the same area, the PGA for 2,500 years return period varies from 0.417 to 0.775 (in g) in USGS hazard data while PGA values in the UN Global data vary from 0.744 and 0.758 (less variation due to coarser resolution of about 38 km grid). After required data validation and quality checks, the USGS PGA rock data has been carried out to generate PGA Soil values using appropriate site amplification factors for the study area. Soil Modifications Local soil conditions can significantly affect earthquake ground motion of an earthquake. The soil top layers act as filters that can modify the ground motion as a function of their dynamic characteristics. Soft, weak soils tend to amplify long-period seismic motions and thus generally impart large ground displacements to structures, while very stiff soil and rock tend to de-amplify the ground motion. For dynamic purposes, soils are classified in terms of their shear wave velocity. A majority of authors, including the European and NGA developers (Schott et al., 2004; Campbell et al., 2009; Boore et al., 2011; Sandikkaya et al., 2013) have used the average shear-wave velocity in the upper 30 meters of sediments, Vs, as the parameter for characterizing effects of sediment stiffness on ground motions. Use of this parameter is considered to be diagnostic in determining site amplification than the broad and ambiguous soil and rock categories used in the earlier studies [with the exception of the relation of Boore et al. (1997), who used Vs,,]. Therefore, the site amplifications of ground motions relative to a reference rock condition are continuous functions of Vs and have been used for the study area, due to the absence of Haiti-specific relationships z- EMERGING = ms ERM 7 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 184] between site classes and amplification effects, and coarse surficial geology at 1:250,000 scale. The widely used NEHRP'Ss site amplification procedure based on shear wave velocities (Wills et al, 2000, BSSC, 2001) has been applied in this study (Table A4.3). Table A4.3 Soil classification scheme based on shear wave velocities Soil Index value JCDMG Class Brief Description ner uw | » | QUE HORS igneous rocks 15 BC Firm sedimentary racks (mid Miocene age) and 760 eathered metamorphic edimentary Formation Mid-Lower Pleistocene age 550-760 25 eak rock to gravelly soils - Deeply weathered and L CD highly fractured bedrock 270-550 | 3o [| D Holocene Alluvial soils 180-270 oung alluvium / Water-saturated alluvial deposits 90-180 40 Non-engineered artificial fill, soft clays, peat and <90 swamp deposits Topographic Slope Based Seismic Site Conditions Wald et al. (2004), and Wald and Allen (2007), describe a general methodology for deriving maps of seismic site conditions using topographic slope as a proxy. Vs: measurements (the average shear-velocity down to 30 m depth) are correlated against topographic slope to develop two sets of coefficients for deriving Vs: at grids. The site-specific Vs: values have been recommended to be used at finer scales or at particular locations. The basic premise of the method is that the topographic slope can be used as a reliable proxy for Vs: as an alternative method in the absence of geologically and geo-technically based site-condition maps by correlating Vs3o measurements and topographic gradient. Based on the past seismicity in northern part of Haiti, the seismic sources and the potential events in the region, the stable coefficient was utilized for Haïti site amplification and spatially interpolated the Vs; data using GIS tools. Due to the size of the study area, the evaluation of seismic site conditions could not rely on the USGS published Vs: data, which is available at 30 m resolution, or the average shear velocity to 30 m depth (Vs:). Therefore, a high resolution (2m) digital elevation model (DEM) has been used to compute topographic slope based site conditions. In this process, the USGS published Vs: data has been plotted against the high resolution (2m) DEM and correlation has been established between these two parameters (Figure A4.7). == EMERGING + ÉUL ERM 8 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 185] 30 LA 25 & 20 L 09 L'] = a SA 81 . n a Le 9 > 10 F < 5 | # , LL __Ë s LU ° 0 150 200 250 300 350 400 450 500 550 600 Vs30 Figure A4.7 Average slope and Vs 30 relationship for the study area This correlation provides the basis for establishing site specific values for the higher resolution elevation data which facilitated the calculation of a soil index map using the NEHRP's classification (Figure A4.8). S Î | os E | DATE (os m7 | Figure A4.8 The slope map derived from high resolution DEM (left) and distribution of Vs 30 within the study area In the next step, ERM has performed sample review of the distribution of Vs: values with the topography. The values of Vs, at specific sites like foothills, riverine plains have been compared to validate their correlation. The soil classification scheme follows the NEHRP scheme of 7 soil classes and their associated Vs;, distribution. The results of the analysis is presented in the below figure (Figure A4.9). _ EMERGING < et GIDB © ELA ERM 9 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 186] j # nc _—. ; [study Area ee Soil Index 21101 -1.50 EM 1.51 -200 Ÿ 4 201-250 ü ” EM 251-301 301-350 mm :5:-400 Figure A4.9 Topographic slope based soil classification for the study area, Haiti Site Conditions Validation from Geological Maps The soil index developed for the study area as per NEHRP classes (Table A4.2) has been validated against the available geological maps for the study area in Haïti. Since, the units in the geological map are not so detailed and the extent of the study area considerably small in size, the classified soil index map developed shows broad relationship with the geological classes, though the overall trend of the boundaries shows similarity. Site Amplifications To derive a more detailed understanding of the seismic hazard in the study area, specifically an understanding of amplification, site amplification factors were applied to the soil index values adopted from NEHRP. The site— dependent amplification factors have followed the non-linear two-dimensional soil amplification factors modified from Choi and Stewart (2005); and Walling, M, Walter Silva, and Norman Abrahamson (2008), which relate non-linear multipliers based on the level of ground motion (PGA) and averaged soil index assigned for a given location. The plot of amplification factors for different soil index classes (corresponding to respective Vs3 values) normalized by the amplification for reference BC soil Vs:,=760 m/s (soil index 1.5), used in the study is shown in Figure A4.6. 3 EMERGING & Initiative ERM 10 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 187] 10.0 — Soil Index 1.0 —— Soil Index 2.0 —— Soil Index 3.0 6 —— Soil Index 4.0 = © (1 LL ë ins 0 = Ê 5 ao 0.1 0.10 1.00 PGA (g) Figure A4.10 Site amplification factors for different soil index values (= VS30 values) The PGA values derived from USGS for the study area were multiplied with site amplification factors that were derived from the Vs:çbased soil index map. The site amplification factors that were derived from the Vs: data were correlated with values from the high resolution topography to derive appropriate site amplification factors for the soils in the study area (Figure A4.10). The outcomes are earthquake hazard maps, expressed in terms of PGA Soil values at 10 m horizontal resolution?. The final hazard maps developed for the study area is shown in the following figures (Figure A4.11 and Figure A4.12). 2 Note that the average slope derived from 2m DTM has been used to correlate with corresponding Vs 30 values. The result was at satisfactory level and was expressed in a 10m mapping resolution. = EMERGING = EU ERM 11 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 188] 4 4 1] 4 ñ À à K X À à. | 16 S Ÿ ; Q À Ÿ 1 Ÿ AR Legend 4 LL study Area N 7 PGA (g) 475 YrRP| VE 4 Value 4 | High : 0.44840 “ M Low : 024419 Figure A4.11 PGA probabilistic seismic hazard map for 10% probability in 50 years, i.e. 475-year return period F À \ / Ÿ si \ 4 dm } Ne { N \ =. \ — Legend à. d LT study Area & À PGA(g)2500 r RP Le / Value 4 4 # High : 0.85304 à» s Î Low : 040597 Figure A4.12 PGA probabilistic seismic hazard map for 2% probability in 50 years, i.e. 2,500 year return period eZ EMERGING Ë CHE GIDB | Initiative ERM 12 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 189] Hazard Frequency and Magnitude The basis of the understanding of the frequency and magnitude of a seismic hazard lies in the probabilistic data using modeled information concerning stochastic events. It has been observed that the peak ground acceleration (PGA in g) with site amplification varies from 0.244 to 0.448 within the study area for 475 years return period. In the same area, the PGA for 2,500 years return period varies from 0.406 to 0.853 (in g). Values for the cited return periods are expressed in tabular form in Table A4A4. Table A4.4 Return Period and Peak Ground Acceleration PGA (ranges within data set) Annual Probability 2500 0.406 to 0.853 1/2500% The frequency and magnitude of a seismic event is interpreted by rating the level of PGA, i.e. to a return period of 2500 years, with the annual probability of occurrence. The annual probability of experiencing an event with a range of acceleration of a 2500-year event, then, is 0.04 percent. A4.3 Hurricane Hazard Context Hurricanes and tropical storms are large-scale systems of severe thunderstorms that develop over tropical or subtropical waters and have a defined, organized circulation. Hurricanes have a maximum sustained (meaning 1-minute average) surface wind speed of at least 74 mph; tropical storms have wind speeds of 39 mph to 74 mph. Hurricanes get their energy from warm waters and lose strength as the system moves inland. Hurricanes and tropical storms can bring severe winds, inland riverine flooding, storm surges, coastal erosion, extreme rainfall, thunderstorms, lightning, and tornadoes. Hurricanes and tropical storms typically have enough moisture to cause extensive flooding throughout a large geographical area, or in the case of Haïti, the entire country. Hurricane magnitude is measured on the Saffir-Simpson hurricane scale, shown in Table A4.5, which categorizes hurricane magnitude by wind speeds and storm surge above normal sea levels. Table A4.5 Saffir-Simpson Hurricane Scale Wind Speed Expected Damage Minimal: Damage primarily to shrubbery and trees; unanchored mobile 1 74-95 mph : homes damaged; some damaged signs; no real damage to structures. 2 96-110 mph Moderate: Some trees toppled; some roof coverings damaged; major damage to mobile homes. 3 111-130 Extensive: Large trees toppled; some structural damage to roofs; mobile mph homes destroyed; structural damage to small homes and utility buildings. 131-155 Extreme: Extensive damage to roofs, windows, and doors; roof systems on mph small buildings completely fail; some curtain walls fail. Catastrophic: Considerable and widespread roof damage; severe window and 5 >155 mph : : A LL : door damage; extensive glass failures; entire buildings may fail. Haiti is among the most hurricane-prone locations in the world. In 2004, the Food and Agriculture Organization (FAO) reported that during a period from 1909 - 2004, forty-seven (47) tropical storms and hurricanes hit Haiti, of which nineteen (19) hurricanes or major climatic events (FAO, 2004). From 2004 to 2012, twelve (12) wind storms have made landfall in Haïti (See Table A4.6 below). == EMERGING = «#4 GIDB © LL ERM 13 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 190] Data from the Prevention Webi, which provides information on human and economic losses from disasters, indicates that between 1980 and 2010, over four million (4,171,407) persons have been affected by hurricanes. For this same period, Prevention Web furthers that there were 4,990 deaths caused by hurricanes and that the estimated economic impacts for the same period reached over USDS 822 Million in Haïti. Dr. Jeffrey Masters, in a summary of Hurricanes in Haiti entitled “Hurricanes and Haiti: A Tragic History“ indicates that the 2008 hurricane season was the cruelest for Haiti, with four (4) named storms making landfall and dumping heavy rains. He estimates that in 2008 alone, economic damages exceeded $1 billion US dollars. Table A4.6 Hurricane History in Northern Haiti E_vesr [Eve] Dern . 11-12 October: Hurricane Hazel affected every part of 1954 H Hazel tricane "aze Haiti. Grand Anse, Ouest, Arbonite, Nord-Ouest 1979 August; Location: limited impact on Nord-Ouest : 23 September: Hurricane Georges; Location: Sud-Est 1598 Hurricane Georges and Nord-Ouest departments. Hurricane/ 16 August: Tropical Storm Fay crossed the entire 2008 s Tropical Storm Fay country. 2008 Hurricane Hanna 1 September: Hurricane Hanna; Location: Artibonite and Nord-Est 2008 Hurricane Ike 6 September: Hurricane Ike; Location: Nord, Ouest and Nord-Ouest Source: NATHAT, 2012, National Meteorological Center of Haïti; One of the most serious components of hurricanes is high winds. Because of the extensive size of a catastrophic hurricane, a storm need not pass directly over Haïti to cause severe damage. A hurricane passing within close proximity to the island of Hispaniola can also cause major damage to property and even loss of life. Essentially there are no areas of Haiti that are free from hurricane force winds. The coastal and low lying areas, such as those of the study area, experience the first effects of damaging winds. This subsection is focused on Hurricane Winds and effects (inland and coastal) flooding are covered in Sections 4.4 and 4.5 respectively. The rains that accompany hurricanes are intense and last for several days. Intense and prolonged rainfall can cause flooding by which water overflows river banks and puts at risk all low-lying areas along with structures and critical facilities and infrastructure. Coastal flooding or storm surges are also prevalent during hurricanes and have the potential to severely impact low-lying coastal villages and overwhelm homes and other buildings near the ocean. Hazard Determination The methodology developed for the identification of wind hazards for this study was based on numerical modeling of hurricane motion and procedures developed by Vickery, 2008 in an assessment entitled “Development of Design Wind Speed Maps for the Caribbean for Application with Wind Load Provisions of ASCE 7. The reference study, which was performed for the Pan American Health Organization (PAHO) under a special grant from the Office of Foreign Development Assistance of the United States Agency for International Development (OFDA/USAID), extended a hurricane simulation model that was originally developed and tested for the Gulf and Atlantic coasts in the United States, to the Caribbean. The models have been calibrated to capture variations of storm characteristics throughout the Caribbean Basin, with specific attention being placed 3 (http://www.preventionweb.net/english/countries/statistics/?cid=74) 4 http://www.wunderground.com/resources/education/haiti.asp?MR=1 == EMERGING = LL ERM 14 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 191] on the Greater Antilles. The technical methodology of the report is concisely summarized below, which forms the basis for the hazard maps utilized in this study. Track and intensity Modeling The hazard model incorporated the associated wind field for historical cyclones in the Atlantic basin. The initial step was to understand the wind speeds for various return periods for locations distributed over the Atlantic and Hispaniola as described in (Vickery et al., 2000, 2008). Central pressure was calculated for each storm and a one dimension model was used to calculate ocean feedback as described in Emanuel et al. (2006). The relative intensity of each storm was then calculated and intensity values were incorporated into a statistical model to understand vertical wind shear. Storm Filling Unlike the United States, there is insufficient data to calculate the effects of roughness on reducing the central pressure of hurricanes in the Caribbean. A filling model is usually used to compute the variation of central pressure of a storm during landfall and to take into consideration the time that the storm is over land and the variation in the intensity of the storm. The limited data that is available is associated with the HURDATS data, which only has six (6) hour temporal resolution and that lack of landfall and exit pressures when storms make landfall and cross islands in the Caribbean (Vickery, 2012). Instead, a filling model that was developed by Vickery (2005) for the New England coast was utilized to model storm weakening as it provided the best data comparisons of storm central pressure statistics when compared to other models utilized in the region (Vickery, 2005). For the purposes of this study, it was assumed that the wind speed will equal the basic values from the hurricane model due to small size of study area and sparse development (ï.e. limited roughness). Model Validation Vickery, as part of the hazard model developed for PAHO/OFDA/USAID implemented a validation procedure that compared the statistics of storm heading, translation speeds, and distance of closest approach, central pressure and annual occurrence rates of modeled and historical storms passing within 250 kilometers of a grid point (Vickery 2008). The tropical cyclones from the period of 1900 to 2007 from HURDAT were used in model validation process. To verify the ability of the model to reproduce the historical storms, statistical tests were performed. The statistical tests include t-tests for equivalence of means, f-test for equivalence of variance and the Kolmogorov- Smirnov (K-S) tests for equivalence of cumulative distribution functions (CDF). The results indicate that overall model reproduces the observed heading data very well and the variance of the observed data is strongly dependent on a few outliers. In most cases, these outliers were associated with one/two storms heading in easterly direction in the southern part of the Caribbean. The modeled values of central pressure represent the minimum pressures anywhere within 250 kms of an established model grid point that is likely to be exceeded, on average, once in 50 years. Quantitative comparison of central pressure shows that the model reproduces wind speeds expected from an intense hurricane passing to the south of the greater Antilles and up through the Yucatan channel. The magnitude of modeled 50 year return period pressures are similar to the observed values, but reflects the smoothing expected for predicted mean values rather than single point observations from 50 year record. In terms of the peak gust, the model and observed wind speeds are in good agreement, however there are relatively few measured gusts with wind speeds greater than 100 mph. 5 The North Atlantic hurricane database, or HURDAT, is the database for all tropical cyclones in the Atlantic Ocean, Gulf of Mexico and Caribbean Sea, since 1851. = EMERGING = LL ERM 15 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 192] While Vickery provides a summary of how the actual hurricane simulation model was validated, ERM also compared the results to a wind hazard data provided in a recent United Nations Strategy for Disaster Reduction (UNISDR) and Global Assessment Report on Disaster Risk Reduction (GAR) report entitled “Probabilistic Modeling of Natural Risks at the Global Level: Global Risk Model, 2013”. This global assessment models the cyclonic winds based on previous trajectories recorded for the main oceanic basins of the world. The hazard model used in the UN model forecasts the maximum intensities associated with the potential occurrence and track of a tropical cyclone in the Atlantic region. The calculations are performed for each of the selected historical tracks and for a set of “children” tracks obtained through the use of a statistical procedure known as disturbance, which allows generating random tracks that conserve the main characteristics of the historical ones so as to produce probabilistic wind speeds for a series of return periods (i.e. 50-, 100-, 250-, 500- and 1,000- year return periods). The data is available at a resolution of 30 km for wind speed. A comparison of the PAHO/OFDA/USAID data with that of the UN model indicates that the UN global wind hazard maps show only minor variation in wind speed over Haiti (See Table A4.7). For instance, for a 50 year return period the variation between minimum and maximum wind speed was only 3 km/h. For 100 year return period variation between minimum and maximum wind speed was only 3.2 km/h. The data from the PAHO/OFDA/USAID wind hazard maps show a much better variation and distribution of wind speeds over Haiti. Since these maps are derived from wider area maps of Hispaniola region, they provide a good distribution and variation in the study area. For 50 and 100 year return period variation between minimum and maximum wind speed is about 70 km/h. The PAHO/OFDA/USAID wind speed data and hazard maps provide a more refined model for Hispaniola, the spatial resolution of this data is quite good. Table A4.7 Comparison of PAHO/OFDA/USAID and UN Wind Hazard Map over Entire Haïti | sou | Minimum Wind (Km/PH) Maximum (Km/PH) Range (Km/PH) USAID 50 Year 114.2 (71) 185 (115) USAID 100 Year 130.3 (81) 201.1 (125) **Values in Parenthesis are Wind speed in MPH Climate Variability, Hazard Frequency and Magnitude Climate Change Variability The structure and areal extent of the wind field in tropical cyclones is largely independent of intensity storms and play an important role on potential impacts. With the use of satellite imagery and other instruments, intensity measurements have become more accurate, and as a result, the recorded intensities of wind storms in the Atlantic have been increasing (Kossin et al., 2007; Elsner et al. 2008). Time series of cyclone indices such as power dissipation show upward trends in the North Atlantic since the late 1970s (Emanuel, 2007) but interpretation of longer-term trends is constrained by data quality concerns (Landsea et al., 2012). The IPCC Fourth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC AR4) concluded that a range of modeling studies project a likely increase in peak wind intensity and near storm precipitation in future tropical cyclones. Simulations with high resolution dynamical models (e.g. Oouchi et al., 2006; Bengtsson et al., 2007; Gualdi et al., 2008; Knutson et al., 2008; Sugi et al., 2009; Bender et al., 2010) and statistical- dynamical models (Emanuel, 2007) consistently find that greenhouse warming causes tropical cyclone intensity to shift towards stronger storms by the end of the 21st century, with an expected 2 to 11% increase in mean maximum wind globally. Frequency and Magnitude = EMERGING = Lo «CF GIDB LL ERM 16 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 193] The IPCC Fifth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC AR5) indicates that the frequency of the most intense storms is more likely than not to increase by more than +10% (IPCC 2013, AR5), while the annual frequency of tropical cyclones are projected to decrease or remain relatively unchanged for the North Atlantic. This suggests no major change in the frequency of hurricanes in North Atlantic region comprising Haïti. The SRES scenario B2 for study area of Haïti suggests that the wind speeds are projected to decrease by very small magnitude of 0.25 m/s (0.559 mph) over the projected for the 20405 relative to the 1960-1990 baseline. These projected changes have applied to model wind speed over the return period to develop wind hazard maps for Haïti that reflect projected climate change scenarios. The resultant maximum wind speed with projected climate change scenario are compared to modeled wind speeds for Haiti and are outlined in Table A4.8. Table A4.8 Hazard Wind Speeds with Climate Change Wind Speed (mph)Without Climate Change Wind Speed (mph) With Climate Change 1700 170.000 169.441 With negligible change in the wind speed (intensity) and no major change in frequency of hurricanes, there should be little effect in terms of climate change on the wind storm hazard that will be used for the risk assessment. Results Hurricanes in their nature are difficult to model for their all associated parameters. It is also very difficult to quantify the impact of climate change on frequency and intensity due to complex nature of phenomenon and interrelations with other variables such as sea surface temperature and changes in land use and climate in inland areas. The summary of the impact of the climate variability is best explained by looking at the changes in wind speed levels found throughout the study region. Presented below are a series of figures which provide an overview of hazard maps in terms of wind speed for three second gusts for a height of 10 meters for a flat terrain for return periods of 50, 100, 700 and 1700 years. == EMERGING = ÉUL ERM 17 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 194] 50 Year Wind Hazard Map without climate change 50 Year Wind Hazard Map with climate change 50 Year Wind Hazard with Climate Ch: fear Wind Hazard with Climate Change <= 60 Year Wind Hazard Wind Speed (mph) Value High : 114.257 Low :71 1 1 100 Year Wind Hazard Map without climate 100 Year Wind Hazard Map with climate change change 100 Year Wind Hazard with Climate Change ns 400 Year Wind Hazard CIE) on ion: 124296 High: 125 Low : 801845 = Se 5 EMERGING +8 GIDB | EC Initiative sit d ERM 18 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 195] 700 Year Wind Hazard Map without climate 700 Year Wind Hazard Map with climate change change > 700 Year Wind Hazard vith Climate Change 700 Year Wind Hazard Pre Spot nee) ss Hi : 154002 Li 17 Low: 10844 Low: 110 se se 1700 Year Wind Hazard Map without climate 1700 Year Wind Hazard Map with climate change change 1700 Year Wind Hazard with Climate Change ss 1700 Year Wind Hazard Wind Speed (mph} Value High : 169.441 Low 119441 ET = A4.4 Inland Flood Hazard Floods can arise from a variety of causes. The most commonly understood floods occur when water levels in rivers rise and the waters overtops their banks, and adjacent floodplains and lowlands are subject to recurring floods. This type of flooding usually occurs after intense or prolonged rainfall. There also occurs in Haiti land flooding due to heavy rains where infiltration of rainfall is impeded (through either impermeable soils or development impacts). This form of localized flooding has not been assessed in this study. 3 EMERGING = Lo CHE BIDB EC Initiative . ERM 19 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 196] Floods in Haiti, as in other Caribbean islands, follow tropical weather patterns. Haïti has two distinct rainy seasons, one from April to June and another from October to November. There have been a number of large- scale devastating flooding events in Haïti through time (see Table A4.9 below). Historically, most of the flooding events have been tied to large-scale climatic events (i.e. tropical cyclones) and have historically had the greatest impact. Recently, however, smaller low pressure systems have impacted Haiti on a yearly basis. Table A4.9 History of Floods in Northern Haiti A A 7 1996 Flooding Very heavy rains have caused floods in several parts of the country, in particular the departments of the North, the Northwest, the Grande Anse and the Gonave island. 2003 Flooding Following rains between Saturday, 20 and Monday, 22 December, severe El 2006 Flooding On the 22 and 23 November, heavy rains caused flooding in Grand'Anse Department and the Nippes and Nord-Ouest departments; Damage to roadways including the collapse of a bridge across Ravine Sable at Trou-Bonbon. 2007 Flooding On 17 March, 50 people had to be evacuated from areas at risk in Grand Anse. New floods were recorded in Cap-Haitien, in other parts of North and Grand Anse. 2012 Flooding Tropical Storm Isaac hit Haiti on 25 Aug 2012, killing at least 19 people. 15,000 people had to be evacuated and 335 homes were destroyed. 2012 Flooding Flooding caused by Hurricane Sandy killed 60 people and significantly damaged critical infrastructure such as roads, schools and hospitals. 1.8 million people have been affected, and more than 18,000 homes have been flooded, damaged or destroyed. (UN News, 2 Nov 2012) 2012 Flooding Heavy rains during the night of 8-9 Nov 2012 in the Nord, Nord-est, Nord- ouest and Nippes departments of Haïti resulted in flooding, damage to homes and 10 deaths in Cap Haitien. More than 1,500 people were housed in 14 shelters. 2013 Flooding Heavy rains on 14, 15 and 28 Jun 2013 caused flooding in Haiti's Artibonite, Nord-ouest and Centre departments. Six people were killed and over 6,600 families affected. Extensive damage was reported in the agriculture and livestock sectors. Source: Relief web, accessed on December 10, 2013, http://reliefweb.int/disasters?f#5B%5D=field country%3A1138f%5B%5D=field_disaster type%3A4611 Haitis rugged and mountainous terrain coupled with environmental degradation and poor watershed management has created optimal conditions for over bank flooding problems. Haiti’s surface waters are concentrated in a restricted number of important rivers that account for about 60 percent of the flow regime (World Bank, 1991). At present it is not unusual for Haiti’s rivers to reach or exceed the high water line twice each year (USAID, 2007). The two principal watersheds in the Study Area are the Trou du Nord and the Grande Rivière du Nord. The principal river of the Trou du Nord Watershed is the Rivière Trou Du Nord. The Trou du Nord watershed measures 110 Km? and the average annual flow rate is estimated to be 0.98 m°/sec (UniQ, 2010). It has a permanent source of water available year round. River flow tests, which were conducted as part of a hydrological assessment of the watershed in 2011, estimated river flows to be 0.45 m/s (February, 2011) and 70 m°/s in (July, 2011). The Rivière Franiche, Rivière Pilette and Rivière Cabaret are the main tributaries to this river and are intermittent streams and are dry part of the year. The Petite Rivière, an intermittent river, is also located in this basin and drains into the Trou du Nord plain. The Grande Rivière du Nord watershed measures 680 Km? and the average discharge (mean daily flow) 5.44 is m/sec. (USAID, 2007). The Rivière Caracol and = EMERGING Lo sé * SUSTAINABLE EX ÉUL *IDB ERM 20 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 197] Rivière Cartache are main tributaries to this river. The Rivière Caracol is a permanent river with a constant source of water, while the Rivière Cartache is an intermittent stream and is dry part of the year. In a 2007 study to compare and rank Haitis watersheds quantitatively, USAID along with the Haitian Government, established review criteria for prioritizing watershed interventions. The USAIDS study explored the relative ranking of watersheds based on their vulnerability to loss of human life, productive infrastructure, soil potential, or erosion risk. Out of Haïiti’s 54 watersheds, the study found that the relative vulnerability ranking of the Trou de Nord and Grand Rivière du Nord watersheds to be significant. à Lt L ë . ou/lu Nord € PRE à 7 x Pic , # éd * à Figure A4.13 Key watersheds in the Study area Source: ERM, CNGIS The general susceptibility of these watersheds in the Northern Development Corridor, has not declined in recent years. In fact, the widespread deforestation, clearing of land for agriculture and increased urbanization has served to exacerbate flooding problems in the region. Urban expansion and unplanned urban development, does not allow aquifers to function as storage and floodplain to work as filters during intense rainfall events (USAID, 2007). The commune of Quarter Morin, which is situated in a moist, low lying alluvial plan and bordered on the east by the Grand Riviere du Nord, is prone to flooding. Several factors have worked to increase the susceptibility of flooding, including more intense climatic events, increased run-off, and the accumulation of debris 5 USAID 2007, Environmental Vulnerability in Haiti: Findings & Recommendations > EMERGING Ë 4 GIDB L Initiative ERM 21 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 198] downstream. Limonade is bordered by the Grand Riviere du Nord on the west. While the Barrage de Tannerie previously helped to contain flood waters and provide irrigation during the dry season, the dam failed in the 19605 and has not been repaired. Intense rainfall causes flooding and the accumulation of water in low-lying areas which are slow to drain following flooding events due to limited or inadequate storm water drainage infrastructure. Limonade receives an average of 1200-1400 mm rainfall annually. In Trou du Nord, the terrain, rainfall and soil types, in both the mountains and plains, give rise to a fairly dense network of rivers. The Trou-du-Nord river is the most important river system. In low-lying portions of commune, there are a series of smaller tributaries and torrential gullies. Large areas of the plain are subject to frequent flooding caused by torrential downpours. The heavy clay content of the soil causes erosion and results in frequent sediment build up in streams. Urban areas are adjacent to the main river with development occurring in riparian zones. Historically, the city has been flooded severely. Flood risks are also present in the northeast portion of the Terrier Rouge commune, sometimes impacting the city on its northern edge. To the south, settlements experience higher annual rainfall amounts, and as a result, experience flash floods. The urban development of the city is constrained by low lying topography, which is prone to flooding. The annual rainfall averages 900 mm on the coast to 1200 mm south of the RN6. Determination of Flood Hazard A detailed flood hazard assessment methodology was pursued to include a meteorological analysis that includes a probabilistic simulation of rainfall which has considered climate change. The hydrological modeling has been conducted for the Basin de la Grande Riviere du Nord and Basin Trou de Nord. The Hydrologic Modeling System (HEC-HMS) is designed to simulate the precipitation-runoff processes of dendritic watershed systems so as to take into consideration total drainage to account for geometric profiles of watersheds. We have used high resolution DTM to conduct the hydrological modeling to generate the precipitation-runoff processes (i.e. flows). Finally, hydraulic modeling of the main rivers identified above have been undertaken to develop probabilistic flood forecast maps for six return periods (ï.e. 2-, 5-,10-, 25-, 50-, 100-return periods) for the portions of the basin that intersect the study area. Challenges in Flood Hazard Model The main challenges presented focused attention on filling in gaps of daily precipitation data and resolving issues associated integrating DTMs at different resolutions of DTMs for the study area. Precipitation Modeling The lack of reliable flooding data, as well as the lack of instrumental rainfall and discharge or flow data within the region, has predicated the need to depend on older national level historical rainfall data or regional proxies in order to understand the hydrology of watersheds within the region. National and regional precipitation data from pluviometric stations that was collected revealed that there was extensive data collected at monthly time intervals and very poor or limited data for 24 hr intervals. The paucity of 24 hr data, which is needed for flood modeling, predicated the need to fill gaps by developing a 24 hr precipitation baseline by comparing monthly rainfall data and global models (i.e. Santa Clara University). This was done in order to ensure that an appropriate approximation was developed to understand flows in the hydrological model. The development of a 24 hr intensity baseline was essential for the execution of a hydrological model given the small size of the study area. It is important to note that various global rainfall data models were compared, evaluated against each other and available monthly rainfall data. The various data compared have been given in Table A4.10. = EMERGING = Lo s4 * SUSTAINABLE EX EU *IDB ERM 22 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 199] Table A4.10 Various Global Data Compared for Deriving Daily Rainfall Distribution FE see [De | TempoaiReouten | Avebiiy | Pen | 1 CPC .25x.25 Daily US Unified Daily USA 1948 to Precipitation 2006 3 CMAP global gridded precipitation Monthly Global 1979 to near RE 4 Global Precipitation Climatology Monthly Global 1901- RE EE 5 GPCP V2.2 Precipitation Monthly Global 1979- present U. of Delaware Precipitation and Air Monthly Global 1901-2010 RE LE LE EEE 7 TRMM- Tropical Rainfall Measuring Radar Based Rainfall Sub | near global 2000 (ee ER Es | Santa Clara University - Gridded Daily Global 1950-1999 RE EE For probabilistic return period analysis of rainfall, at 30 years of continuous records are desirable so were data at a daily time interval. Therefore, the Santa Clara University data was utilized for the time period of 1950-1999, which matches with the longest and consistent data recorded in Haiti. Figure A4.14 provides a depiction of a comparison of the monthly precipitation data from Santa Clara University and that of the Cap Haitien pluviometric station. The 24 hour rainfall data from nearest grid point of Santa Clara University has been processed and a time distribution has been applied over the monthly rainfall data that has been collected for pluviometric stations in and around the study area. Snapshot of estimated daily time step data is given below (Figure A4.15). 3000 :- Ë 2500 d ë 2000 & 8 1500 £ 1000 $ 500 LS e LA 0 0 500 1000 1500 2000 2500 3000 Observed Data Cap Haïitian Station Annual Rainfall, mm Figure A4.14 Comparison of Santa Clara Data with observed monthly precipitation data “#8 GIDB ms ERM 23 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 200] [vear [Month [Day ]sc Daily Rainfall [UD TSanta Ciara Monthiy [Distribution % [Observed Monthly [Estimated Daily | [1950[ 1] 1] 7 3.23/950 1 | 51.58] 7 6:26%] 7 621] 7 3.5] [1950[ 1] 2) 7 ofa9501 | 51.58] 0.00%] 621] 0.0 [1950f 71] 7 3] 7243/1950 1 | 51.58) 2776] 7 621] 172] [1950f 1) 4 ofa950 1 | 51.58) 0.00%] 621] 0.0 [1950 7175] 13h50 1 | 5158) 258%] 7 621] 7 1.60) [1950f 1] 7 6] 7 ofa9501 | 51.58] 0.00%] 7 621] 0.00 [1950f 1) 7] ofa9501 | 51.58] 0.00%] 7 621] 0.0 [1950[ 1] 8] 7 aséfoso à | 51.58) 7 8.84%] 7 621] 7 5.4] [1950[ 1] 7 9 7 of1950 1 | 51.58] 0.00%] 621] 0.0 [1950f 1] 10)" of1950 1 | 51.58] 0.00%] 621] 0.0 [1950f 1) 21)" ofi950 1 | 51.58] 0.00%] 621] 0.0 [1950 7 177 12] 7523/0501 | 5158) 10.53%] 7 621] 7 6:54 [1950 7 1] 13] 7 ofsso1 | 5158) 000%] 7 621] 7 0.00 [1999[ 12] 16] 7 o15/1998 12] 7 66.79] 0.22%] 7 531] 012 [1999[ 12] 17] 7 of1998 12] 7 66.79] 0.00%] 531] 0.0 [1999[ 12] 18] 7 of998 12] 7 66.79] 0.00%] 53.1] 7 0.0 [1999[ 12] 19] 77 ouafi9se 12] 777 66.79] 016%] 7 53.1] 70.0 [1999/1220] 0.25/1998 12] 7 66.79] 0.37%] 53.1] 0.2 [1999 12] 21] 7123/1980 12) 7 66.79) 184%] 7 531] 7 08] [1999[ 7 12] 7 22] 7022/1990 12] 7 66.79] 0.33%] 77 531] 7 017 [1999[ 12] 23] 7215/1998 12] 66.79] 3.22%] 7 531] 171] [1999[ 12] 24] 7 oa5f1998 12] 7 66.79] 0.67%] 7 531] 7 036] [1999[ 12] 25] 7137/1998 12] 7 66.79] 2.05%] 7 531] 10 [1999[ 12] 26] 7347/1998 12] 7 66.79] 5.20%] 7 531] 7 2%] [1999[ 12] 27] 7285/1998 12] 7 66.79] 4.276] 531] 277 [1989[ 12] 28] 7 ofisso 12] 7 66.7) 0.00%] 7 531] 7 0.00 [1999[ 7 12] 7 29] 7 of1999 12] 7 66.79] 7 0.00%] 7 531] 7 0.00 [1999[ 12] 30] 7542/1990 12] 7 66.79] 7 su] 7 531] 431] [1999[ 7 12] 7 31] 7402/1998 12] 7 66.79] 7 6.02%] 7 531] 7 3.20 Figure A4.15 Snapshot of estimated daily rainfall using observed monthly rainfall and Santa Clara data The simulated daily rainfall data for the pluvimetric stations have been used to estimate probabilities or return period rainfall. Using simulated data series, annual maximum 24hr was derived for 49 years to fit the probability distribution. Using Log Pearson Type Ill distribution return period rainfall for 2, 5, 10, 25, 50, and 100 years has been estimated (Figure A4.16). = EMERGING = È Initiative ERM 24 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 201] —— Without Climate Change —— With Climate Change 300 250 À} € 200 + T € 5 150 —" )û0 | £ & 100 + t 50 +—— CE ES 1 10 100 Return Period, Years —— Without CC — Engg Firm Study —— With CC 300 250 + E 200 +—} + + + — E 2 150 | £ & 100 + = ———— 50 +—— — o 1 10 100 Return Period, Years Figure A4.16 Estimated Return Period Rainfalls The estimated return period rainfall has been given as input to HEC-HMS model for simulation of flows to corresponding return periods. Digital Elevation Model Development Initially only a 10 m DTM was available for the study area. The hydrological modeling was initiated to determine if this resolution was sufficient for modeling purposes. It was found that the 10 m DTM was quite flat in coastal areas, with very negligible variation (i.e. the coastal areas showed mostly 0 m elevation). Instead, a 2 m DTM was made available for a more confined study area and required that the 2 m DTM be merged with the 10m DTM in order to create an elevation model with sufficient resolution for flood modeling. The merging of the two elevation models presented challenges as there was a high variation in the elevation in overlapping areas particularly at the southern edges (Figure A4.17). = EMERGING et GIDB © L Initiative. ERM 25 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 202] ( ( um 1066 ie Je CP \ EE cer as .. & \ L de SN \ N ù < HAN À «ee Cûe fire L \ : $ { Ds \ < 2mDTM 10 m DTM Figure A4.17 Two different DTMs available for study area and beyond As discussed above, two DTMs (of 2m and 10m resolutions) were evaluated separately. It was found that overlapping areas of two DTMs showed elevation difference of + 10 m at the edge of 2 m DTM predicating a more complex merger procedure that involved the overlaying of contours to understand elevation differences and the use of generalized and point specific data to refine data to desired the 2 m resolution. This merged 2 m DTM was cross checked with 9 Geodetic Control Points made available by IDB and the comparison revealed only a 0.12m average differential in elevation (Table A4.11). L . EMERGING 4 GIDB ELA ERM 26 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 203] Table A4.11 Various Global Data Compared for Deriving Daily Rainfall Distribution | GcPPoint | GCP Elevation Merged DTM Elevation, m 5 sm ss je | RS PE CT Hydrological Model A Hydrological model for key watersheds/drainage basins of the study area have been developed using HEC- HMS model. The hydrologic model was established in a GIS framework using Environmental Systems Research Institute (ESRI) ArcGIS and HEC-GeoHMS 4.2 software. HEC-GeoHMS 4.2 is an extension application that supports identification of the river network and division of the catchment (basin area) into a number of sub areas. This procedure requires a digital elevation model (DEM). As mentioned earlier, the 2 m merged DTM was used to delineate basin and river networks. Using HEC Geo-HMS, the river network and sub basins have been delineated using a systematic approach. The approach creates raster grids for catchment delineation. Activities to complete the model include filling sinks, creating flow direction and flow accumulation grids, processing catchment grid, and processing drainage line. The physical representation of the basin incorporates various hydrologic elements (sub basins, river reaches, and junctions), which are connected in a dendritic network to simulate the rainfall-run-off process. Based on the DTM, soil and land use information, various parameters such as abstractions, infiltration, routing have been estimated for each sub basin and are given as input to the model. Grand Rivere du Nord: The hydrological basin area of Grand Rivere du Nord is estimated to be approximately 611 square kilometers. Through basin delineation and hydrological model development process, fifty-one (51) sub basins were developed for this basin, which also included 25 stream reaches. The HMS model schematic for this basin is shown in Figure A4.18. == EMERGING = ÉUL ERM 27 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 204] Fée Gét View Componerts Parumeters Cempute Ress Teck Help COLPEMELETI ELLE ELLLE) 8 - 5 RTE fs her À «0 sg BE ‘wrome paires BR Ye ue A chien Drenves pr NS 8e 2 CE Tarn _ men g 3 re DAS Cap = CONS AAC CET “ Er: 4 | a Sedment: de e Ÿ Pas che Figure A4.18 HMS set up for River Grand du Nord Trou du Nord: The basin area of main stream of Trou du Nord is estimated to be approximately 106 square kilometers. Through basin delineation and hydrological model development process in all 11 sub basins were developed for this basin, which also included 5 stream reaches. HMS model schematic for this basin is shown in Figure A4.19. The HMS model schematic for another stream in Trou du Nord basin is shown in Figure A4.20. Fe Et Ven Comperents Premims Comp Mets Ven PP DO: +sLnFrTs +55 dE erracen y cu Te D | _ ï 12 | ; ARR ve ocre. = que Figure A4.19 HMS set up for River Trou du Nord L EMERGING 4 GIDB ELA ERM 28 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 205] anne sorte Se NE OROR OCER Le Eee Dm Compomts Emmes Compte Em Lente Ep Dégah+ascwp er “mmme de siemm + re | { \ pere d | Figure A4.20 HMS set up for River Trou du Nord (Stream 2) The key selected parameters three watersheds are given in Table A4.12. Table A4.12 Various Global Data Compared for Deriving Daily Rainfall Distribution? River Basin length of principal Slope, Concentration river, m m/m time (min) Grand River Du Nord 107,949 0.0044 | 1,186 In absence of the observed historical flow or discharge information, detailed hydrological model calibration and validation was not possible. In order achieve the reasonable confidence in developed models, runoff coefficients (ratio of runoff to rainfall) from reported studies (MARNDR, MPCE, MICT, MDE, 2000) were compared with simulated runoff coefficient from the model developed in this study. The study MARNDR, MPCE, MICT, MDE, 2000 reported runoff coefficient for River Grand Du Nord as 20.5%. Various simulations of the model used for this study gave an average runoff coefficient of 18.1%. The difference between observed and simulated runoff coefficient is 2.4 % and was determined to be within acceptable range, and as a result, the average run-off coefficient (i.e. 18.1%) was used to simulate the return period flows for various streams. 7 In absence of recorded daily rainfall, analysis is based on mix of observed monthly and modeled rainfall records and only with very extensive and reliable flow information would we be able to accurately model hydrological regime. = EMERGING EU ERM 29 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 206] Hydraulic Model In many applications of river flood modeling, a one-dimensional full hydrodynamic modeling system is used. The one-dimensional hydrodynamic model HEC-RAS, developed by the United States Army Corps of Engineer’s Hydrologic Engineering Centre, was used for performing hydraulic calculations for the river stretches. HEC-RAS is an integrated system of software. It can calculate water surface profiles for both steady and unsteady, gradually varied flows for a full network of natural and constructed channels. Steady flow simulation was adopted for this study. The model comprises channel and floodplain geometry, which is defined by a series of cross-sections or transects together with hydraulic structures such as bridges, weirs, and levees. The basic computational procedure is based on the solution of the one-dimensional energy equation. Energy losses are evaluated by friction, expansion, and contraction losses. The momentum equation is utilized in situations where the water surface profile is varying rapidly. The situations include a mixed flow regime. Cross Sections The HEC Geo-RAS software application was used in a Geographic Information Systems (GIS) (ESRI Arc GIS 10) to develop and verify geometry of principal river systems by developing cross section of the channels and floodplains. HEC- GeoRAS is a set of procedures, tools and utilities that were used to process georeferenced data in a GIS environment to facilitate and complete the job with HEC- RAS . For this study, the cross sections of the channel and the geometry of the terrain in HEC- GeoRAS were digitized and then the file is exported to HEC- RAS to calculate flow rates. Cross-sections have been extracted from DEM at a spacing of about 100 m. The HEC- RAS tool interpolates channel cross sections, as shown in the figures below. After performing hydraulic modeling with HEC- RAS , HEC- with a post-process generated the final GeoRAS results that are flooding surfaces for each return period was performed. Figure A4.21 and Figure A4.22 show the cross sections for river Grand Du Nord and River Trou Du Nord respectively. ee 0 ER k dt j = Figure A4.21 Cross section for River Grand du Nord == EMERGING = EU ERM 30 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 207] gene EE. © CR OU. —- Te) GT pr neue nou F 3 route MTS aun Î A crées Figure A4.22 Cross section for River Trou du Nord Roughness Coefficients The Manning roughness coefficients were determined based on land use information available. Manning coefficients for the main channel and floodplains are shown in Table A4.13 below. Table A4.13 Roughness Coefficients mer | Foodptain | mainchame | River Grand Du Nord 0.035 0.025 River Trou Du Nord 0.035 0.025 = EMERGING Lo CHE GIDB L Initiative. ERM 31 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 208] Peak flow Rates Table A4.14 presents the flow rates (cumec) for return periods of 2, 5, 10, 25, 50 and 100 years. Table A4.14 Peak flow rates for major rivers . Grand River Du Nord Trou Du Nord Return Period, 130 123.0 238.5 263.5 320 3590 340 2360 5280 59.0 Lasr las |s00 350 n30 108.0 The summary of flow rates represents appropriate scenarios for modeling the flood hazard in the study area. Having said this, the analysis is based on modeled rainfall records and only with very extensive and reliable flow information would we be able to accurately model hydrological regime. Flood Hazard Maps The above data was used to understand flood hazard, particularly the overflow channels in the Northern Development Corridor of Haiti by defining associated flood events with different probabilities. For the calculation of these variables the HEC-RAS hydraulic model was used to characterize inland flooding in the study area. Flood hazard maps for 50 and 100 year return periods are shown in Figure A4.23 and Figure A4.24. re u = & 1 Ô F EX oen Legend —— Ruers Flood 50 Year RP Without Climate Change Flood Depth,m UT (= 934005 een T Low:0 Fe TZ s 0 kaometers Country Beuntary Figure A4.23 Flood Hazard Map without Climate Change 50 year return period == EMERGING = EU ERM 32 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 209] a —— = = € C \ Legend Rues Flood 100 Year RP Without Climate Change Flood Depth.m LS 101874 men t Low: 0 FE B 5 m) County Beundary [EL nov Proposeu sus area — Figure A4.24 Flood Hazard Map without Climate Change 100 year return period Climate Variability, Hazard Frequency and Magnitude Climate Change Variability including Climate Change The main step in delineating the extent of flooding under future climatic conditions involves the use of climate projections (taken from the University of West Indies studies in Appendix 3). Rainfall data, particularly the potential increase in extreme rainfall events, in return flow values at 2, 5, 10, 25, 50, and 100 year return periods to understand changes in susceptibility to flooding due to climate change were applied. The resulting impacts are manifested in changes in the extent of flooding. The projected impact on climate change is presented in Table A4.15, which shows the variance in maximum flow rates for events of 2, 5, 10, 25, 50, and 100 year return periods, as well as shown in Figure A4.25 for main stream of river Trou Du Nord for all return periods. Table A4.15 Peak flow rates for major rivers with and without climate change 528.0 632.0 635.0 737.0 593.0 707.0 713.0 829.0 == EMERGING = Lo «CF GIDB LL ERM 33 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 210] — Simulated Flow Without CC —=Simulated Flow With CC 140.0 120.0 1 100.0 Î 9 Ë £ 800 ++ << +1 (e] 3 60.0 —— + —— +" ———— 2 [= 40.0 ——— EE ———— — —— ———— ——— 20.0 Î 0.0 î 1 10 100 Retrun Period, Years Figure A4.25 Return Period Flows with and without climate change for Trou du Nord river Frequency and Magnitude Table A4.16 shows maximum flood depths simulated for the study area with and without climate change. Results shows that with climate change on an average flood depth will increase by about 0.23 m (23 cm) across all return periods. With climate change, the flood depth for a 100 year return period is expected to be 10.17m. Flood hazard maps for 50 and 100 year return periods with climate change are shown in Figure A4.26 and Figure A4.27. Table A4.16 Various Global Data Compared for Deriving Daily Rainfall Distribution Maximum Flood Depth, m [50 |oga lion 1017 1036 z- EMERGING = gs ERM 34 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 211] É LE À. NS F l re FDA : Pa . F A on sen. JF 3 A St Legend LE Ruers Flood 50 Year RP With Climate Change Flood Depth,m EX | High : 10.1609 son. Ü Low : 0 FT [_] Country Boundary [ET New Proposed Study Area Figure A4.26 Flood Hazard Map with Climate Change 50 year return period sun : { | ER } JE À FA . es É CN ER RE Lun son. E È BR Legend Rues Flood 100 Year RP With Climate Change Flood Depth,m EX | High : 10.355 son. Ü Low : 0 FT LL] New Proposed Study Area Figure A4.27 Flood Hazard Map with Climate Change 100 year return period EMERGING Lo 85 GIDB ELA ERM 35 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 212] Through climate change impact assessment on floods it is observed that slight increase in frequency and magnitude will have minor impact on vulnerability of the structures in the study area. Overall impact in terms of climate change induced flood risk shall have minor implications in planning stage. Comparing the areas of the floodplain without climate change reveals that the flood zone for the event 50 years would be 12.21 km? and the floodplain for the event 100 years would be 13.09 km?. In other words, there is only a slight change (7.2 % increases) in the extent of the floodplain for the 100 year event. The comparison between flood zones for events 50 and 100 years with climate change reveals bit smaller difference in the flooded area (about 5.2 % increase as shown in Table A4.17) Table A4.17 Comparison of Inundation area with and without climate change Return Period, Inundation Area, sq km 5 9e las 10.59 11.37 11.53 12.41 12.21 12.97 13.09 13.64 A comparison between the return period of 50 years without and with climate change reveals just a 6.2 % increase, while the difference in the floodplain for the event 100 years with and without climate change was 4.2 %. A4.5 Coastal Flood Hazard Context High waves associated with tropical cyclones are potentially very dangerous and damaging to the coastal settlements. The United States National Oceanic and Atmospheric Administration (NOAA)5 identify this phenomenon as storm surge, which is defined as an abnormal rise of water generated by a storm, over and above the predicted astronomical tides. NOAA furthers that storm surge should not be confused with storm tide, which is defined as the water level rise due to the combination of storm surge and the astronomical tide. This rise in water level can cause extreme flooding in coastal areas particularly when storm surge coincides with normal high tide. The storm surge is produced by water being pushed toward the shore by the force of the winds moving cyclonically around the storm. The impact on surge of the low pressure associated with intense storms is minimal in comparison to the water being forced toward the shore by the wind. In the Caribbean, hurricane categories can be used to approximate expected storm surges. The resulting potential inundation areas are grouped by categories which refers to the Saffir-Simpson Hurricane intensity Scale described in Table A4.18. 8 http://www.nhc.noaa.gov/surge/ = EMERGING = EU ERM 36 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 213] Table A4.18 Saffir-Simpson Hurricane Scale and expected storm surge Storm Surge (feet above normal sea level) put The intensity of the storm surge is affected by the width and slope of the continental shelf. A shallow slope will potentially produce a greater storm surge than a steep shelf. In the north, continental shelf drops off very quickly, which works to lessen the impact of storm surge. The interaction of these various processes is illustrated in Figure A4.28. The total design water level can be computed by addition of three components. The wave setup heights are the heights of wave crests above the storm surge level in open water. The coastal provinces of Northern Haiti experience coastal flooding due to the destructive effects of total design water levels. LS ALES ANR Wind Waves Re LS EP ve Storm Surge LÉ LPS LS LR LE Highest Te LAIIRLN PL D HR LE, Dr Level LRO LR CAS 02 iowest Tide Figure A4.28 Systematic diagram illustrating the contributions to coastal sea level from tides, storm surge and wave processes Source: http://www.cmar.csiro.au/sealevel/s!_drives_short.html Zahibo° (2012), however, postulates that the position of the island of Hispaniola may give it special protection from extreme waves. The presence of several banks (Figure A4.29) , such as the Turks and Caicos, Mouchoir and Silver banks, protect coastal areas against the long swells of generated in the North Atlantic. Zahibo furthers that these extensive shallow areas (50-150 meters in depth) dissipate the energy of longer swells and thus reduce the potential impact from waves. ° unpublished paper as part of NATHAT z EMERGING = Lo CHE GIDB L Initiative. ERM 37 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 214] Bathymetry (m) of the Hispaniola Island 24N 23N -1000 22N -2000 21 LL 4 -3000 20N -4000 19N S si -5000 18N -6000 17N -7000 16N -8000 15 N 76W 74W 72W 70W 68 W 66 W Figure A4.29 Bathymetry of the Island of Hispaniola Source: Adopted from Zahibo, 2012, General Bathymetric Chart of the Oceans, 2008 In Northern Haïti, towns such as Bor de Mer de Limonade, Caracol and Phaeton are susceptible to coastal flooding caused by storm surge. The American Association of Architects, indicate that these settlements are in a precarious location to shoreline (American Institute of Architects, 2012). Similarly, populated areas that are peripheral to the study area, such as Cap Haitian and Petite Anse to the west and Fort Liberte in the east, are also susceptible to coastal flooding. There is, however, limited documented history concerning storm surges in Haiti, let alone well documented instances of coastal flooding within the Northern Development Corridor. Hazard Determination The hurricane induced probable storm surge heights can be derived in a scientific manner based on hydrodynamic modeling. This approach consists of two major components. The first component comprises a hurricane wind field model that provides estimates of the wind speed and direction based on key hurricane parameters at an arbitrary position. The second portion of the model is the surge model uses numerical methods to predict storm surges by solving equations to determine the response of the sea and associated inland extent of flooding that could be generated by a hurricane (cyclonic surface wind field) crossing any coastal stretch. Such models are data intensive and require extensive modeling of a number of parameters such as oceanographic and meteorological parameters, hydrological input, basin characteristics, coastal geometry, wind stress and seabed friction, and information about astronomical tides. In addition, the accuracy of the surge height and associated flood depth and extent of horizontal inundation depends heavily on the accuracy and resolution of the bathymetry data and elevation data that is available for coastal areas. Coastal Hazard Flood Model Since the data requirements for the development of a detailed determination of the coastal flooding hazard were unavailable for this study effort, a regional model was utilized and adopted to understand wave and surge heights in the study area. The information utilized for this study effort was derived from the Atlas of Probable Storm Effects in the Caribbean Sea, which was developed under the Caribbean Disaster Mitigation Project == EMERGING = Lo «CT # GIDB LL ERM 38 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 215] (CDMP), a joint effort of the Organization of American States (OAS) and the US Agency for International Development (USAID). The Arbiter of Storms (TAOS) hazard modeling system was used to create the storm hazard data sets that were used for this study. Technical details of the TAOS model system are documented in various published papers noted in the bibliography, in particular Watson and Johnson (1999). The TAOS combined three wind, hydrodynamic and wave models using the ensemble assimilation methodology used to analyze the results of multiple models may be found in (Watson, 1995; Watson and Johnson, 1999) TOAS wind modeling Winds represent sustained 1-minute winds at 10 m above the surface, and include both surface friction and topographic effects at a resolution of 30 arc-seconds. Friction factors were used for land-cover classification, with water, forest and open land predominating. TAOS Wave Modeling The wave model configuration used for the Caribbean simulation consisted of the NOAA Wavewatch III model, modified within the TAOS storm centered grid system. Modeling was carried on regional basis on a 30 meter grid cell, allowing for an adequate treatment of the near-shore wave environments. Waves are the heights of wave crests above the storm surge level in open water. TAOS Storm Hazard modeling The TAOS storm hazard modeling system contains three primary storm surge modeling modules that include the effects of astronomical tides, which take into consideration relative to mean sea level. Wave setup (but not wave run-up) is included in the storm surge values. Surges over land are shown as elevation above sea level, not water depth. Hazard Maps Based on the results, TAOS provided wave and storm surge heights for four return periods (10, 20, 50, and 100 years), which were reported for specific points along the Haïitian coast. For this study, wave height and surge heights that were reported for Cap Haïitian were adopted for the entire study area. The total surface water levels were computed as a linear addition of storm surge amplitudes, tidal amplitudes, and wave setup. These water levels were then projected onto the coastal land using GIS techniques and intersected with a 2M Digital Terrain Model (DTM) to demarcate the horizontal extent of inundation. As the surges considered tidal information and are assumed to reflect flooding at mean sea level, the flood depths are deduced by subtracting local topography from the total water levels. Based on this information given in the Table A4.19, probabilistic coastal inundation maps of the following the 10-, 25-, 50- and 100-year return periods were developed for the entire study area and are depicted in Figure A4.30, Figure A4.31, Figure A4.32 and Figure A4.33. = EMERGING = CR SUSTAINABLE “IDB EU ERM 39 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 216] Table A4.19 Newly Available high resolution DTM for study area (DTM source used) Return Period Wave Height (m) Surge Height (m) Total Surface Water Levels (m) ES EE EE | sovear [2 Jos 50 100 Vear À mieu" à: d FEU ons À # | sr T ur uen —— Et 7" 4 | E Wei deg, 8 A7: nf ? À fe es é x à 1 Legend [CL] New Proposed Study area }Hasen FE { New DTM Study Area Elevation, m | High : 186.379 PS] Low : 0 Coastal Surge 10 Year RP Flood Depth, m ù mn High : 32 Figure A4.30 10 Year Return Period _ EMERGING Lo 2H GIDB À Initiative ERM 40 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 217] RE RE A ue * ; AS LT Se ( Lin D ai” Re 8 lé M P* | es V4 d'hiie ALP | à + N [CL] new Proposed Study area Fbssen jus { New DTM Study Area Elevation, m me High : 186.379 [I Low: 0 Coastal Surge 25 Year RP Flood Depth, m ° 25 5 10 Kiomele RX | High : 43 | Figure A4.31 25 Year Return Period WF | $ Sd) Nu x2 k À ” ; mé nn pe L= Pda 7.17% .248 1: _ N (CL new Proposea stuay Area seen FT [ New DTM Study Area Elevation, m | High : 186.379 Es. Coastal Surge 50 Year RP Flood Depth, m ° 25 5 10 Kiomete XX | High: 5 Figure A4.32 50 Year Return Period _ EMERGING È “#5 GIDB | Initiative ERM 41 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 218] A 4 : Naf 7%, à | Legend | (CL nov Proposed Study Area uen te | New DTM Study Area Elevation, m Eu 186.379 CCE | Coastal Surge 100 Year RP Flood Depth, m 0 25! 5 " = |. 1 53 Fe a ‘ (| Low: 0 Figure A4.33 100 Year Return Period Climate Variability, Hazard Frequency and Magnitude Climate Change Variability The Caribbean region has been characterized as among the most vulnerable to climate change and climate extremes. One of the major challenges facing island states is that posed by tropical cyclone events and sea level rise. Sea-level rise greatly impacts human activity near the coastal zone (IPCC, 2007) since in many cases the majority of human settlements, economic activity, infrastructure and services are located at or near the coast and local economies are often reliant on just a few sectors such as tourism and agriculture (Nicholls, 1998). Sea level rise therefore exacerbates the vulnerability of coastal regions to other physical processes (e.g. storm surges, storm waves). Evaluations of global sea level change suggest that the current average of rise is approximately 1.5mm per year meanwhile the global mean surface temperature risen around 0.5° C has been widely accepted. Frequency and Magnitude The IPCC Fifth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC AR5) indicates that the frequency of the most intense storms and associated storm surges or coastal floods is more likely than not to increase by more than +10% (IPCC 2013, AR5), while the annual frequency of tropical cyclones and associated storm surges or coastal floods are projected to decrease or remain relatively unchanged for the North Atlantic. This suggests no major change in the frequency of hurricanes and associated storm surges or coastal floods in North Atlantic region comprising Haiti. The SRES scenario A1B for study area of Haiti suggests that the sea level rise are projected to increase by small magnitude of 0.35 m over the projected for the 20405 relative to the 1960-1990 baseline. These projected changes have applied to estimated water surface elevations over the return period to develop coastal flood hazard maps for Haïti that reflect projected climate change scenarios. 3 EMERGING < Initiative ERM 42 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 219] The ‘estimated maximum depth of flooding, including the projected climate change scenario, are compared to estimated maximum flood depth without climate change for Haïti and are outlined in Table A4.20. Table A4.20 Coastal Flood Depth with Climate Change . Estimated Maximum Depth of Estimated Maximum Depth of Flooding with Return Period : n Flooding, m Climate Change, m Results Minor changes in sea level rise are represented on the extent of flooding. The difference between the total area of the surface of the water (inundation) for all normal return period events (from 10 year to 100 year) and the water surface with the projection scenario of climate change (i.e. Projection with Climate Change for all return period events) are presented in Table A4.21 below. Table A4.21 Extent of Coastal Flooding with and without Climate Variability, Km? Flooding Area without CC Flooding Area with CC 100 — Year 134.21 km? 140.79 km” Comparing the areas of the coastal flood inundation area without climate change reveals that the flood area for the 50-year event would be 123.4 km? and the floodplain for the event 100 years would be 134.21 km?. In other words, there is only a slight change (ï.e. an 8.8% increase) in the extent of the floodplain for the 100 year event. The comparison between flood zones for events 50 and 100 years with climate change reveals almost the same difference in the flooded area (about 10.2 % increase). À comparison between the return period of 50 years Without and with climate change reveals just a 3.6 % increase in the inundation area, while the difference in the floodplain for the event 100 years with and without climate change was 4.9 %. A4.6 Drought Hazard Context Drought manifests itself in many forms. The drought hazard in Haïti results from a combination of erratic rainfall patterns during the two distinct rainy seasons: April-June and October-November. According to the World Bank, El Niño/ ENSO episodes have tended to delay the arrival of the rainy season(s) and create drought conditions !° in the country. NATHAT (2012), in a national level hazard assessment, indicated that farmers are reporting longer dry seasons and wetter and shorter rainy seasons. The NATHAT study has also categorized most of the northern coast as being susceptible to drought hazard. The cumulative effects of longer dry periods are crop losses and that more families are becoming reliant on food assistance during the “hunger season”, which is the three-month period between rainy seasons, in which there is little harvesting and employment opportunities. 1 http://www.gfdrr.org/sites/gfdrr.org/files/Haiti-2010.pdf == EMERGING = «# GIDB © LL ERM 43 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 220] ! j Î i ; Ë Ë 3 ô Légende RS, à ml. Locaité MNT (m.s.n.m.) Si "ARE | me Me | Eses F ro L__] Gountor du pays [FE] 211 - 489 rs) LA Le De 1 = _ En « Set SRE —|SBID EM 50-1555 = Mia. D A. : = : D LR Perte At ec _ ] 1366-2080 en E 7 AT 20h …| RÉPUBLIQUE D'HAÏTI CRE E A | vo Pa + ie LA ù Bu) Do |, | Susceptibilité à la ” pe ou L"" 700 LS H se sécheresse dé Universal Transverse Mercator - 18N DE AO nt ee "7 | T bee Ka Romane PREDICT rar j pe re Tom || an eco tte mn AS se a A ne LL <a scénarse des ons cimatques i É É f i i | É i È i f Date Mean Figure A4.34 Areas Most Likely To Drought And Land Degradation on the island of Haiti Source: NATHAT, 2012 EM-DAT, the OFDA/CRED International Disaster Database, reported that during a period of 1960 to 2003, there were seven (7) droughts in Haïti and have indicated where they have located and number of persons affected (See Table A4.22 below). Table A4.22 Droughts in Haiti from 1968-2000 1968 PE A A A A ES PS ET 1984 A RE A PE A ES A ET EE DS A PE AS AE PE ES EE [rotal [7 [6 [6 15 15 Te To Te [71 | Source: Cartes et etude de risqué, de la vulnerabile et des capacities de response en Haïti; EM-DAT: The OFDA/CRED International Disaster Database — www.emdat.be — Université catholique de Louvain — Brussels — Belgium. The Famine Early Warning System Network (FEWS NET) reported in August 2011 that the north and northeast were affected by drought and estimated that major crop yields would be diminished by 20 percent. The FEWS EMERGING & «HE GIDB L Initiative ERM 44 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 221] Net furthered detailed what is a recurring phenomenon in the northern coastal plains and reported that “rainfall in wet mountain areas [of the Northeast] has helped spur crop growth and development, while crops in Ferrié, Fort Liberté (except on the Maribaroux plains), Terrier Rouge, Caracol, and Trou-du-Nord have failed due to the drought conditions in these areas, prolonging the lean season, which generally “ends in June... [and that]. virtually the entire northern region has been affected by the drought which delayed the start of the spring planting season, which eventually got underway in June with the onset of the rains..” (FEWS, 2012). Of greater concern to stakeholders is the impact that these short term fluctuations in precipitation will have on the surface and subsurface water supply or the hydrological regime of watersheds that intersect the Northern Development Corridor. It takes longer to recognize the affects of hydrological drought on soil moisture levels, stream flows, as well as in groundwater and reservoir levels. The frequency of hydrological drought is typically measured over the longer term and predicates a need for understanding of both the supply and demand for water. Hydrological drought concerned with the problems associated with deficiencies in precipitation (the supply) and that of competing interests for water access and utilization (the demand). Water availability and access is exacerbated by limited water management infrastructure"? and in an IDB commissioned assessment of environmental resources of the Northern corridor, the American Association of Architects (AIA) maintained that future urban expansion will place increasing pressure on water resources. Future development, the study maintained, will continue to consume agricultural soils. The study furthered that to maintain current levels of food security, promote economic development and improve climate resilience, investments in irrigation and improved drainage would be required (AIA, 2012). It should be noted that the access to water, represents one of the most significant challenges that Haïti faces. The water access issue is influenced by socio-economic, political, developmental and environmental factors and consequently is an issue that is beyond the scope and intent of this study. Determination of Hydrological Drought The assessment of drought in this study will therefore be focused on the influence of precipitation, and how this is coupled with the anticipated impacts of development and climate variability, will impact the current and future water supply. The current and future water balance will be estimated for the two main watersheds that intersect the study region. This study does not attempt to address and factor in the broader environmental, political and socio-economical factors that also play a role in exasperating the problem of water access. The hydrological drought assessment has been done by estimating components of the classical hydrological cycle. The movement of water through the hydrological cycle varies significantly in both time and space. The hydrological cycle emphasizes the four factors of interest to hydrologists: precipitation, evapotranspiration, surface runoff and groundwater. For this analysis, the hydrological models, that developed as part of this study for the flood hazard assessment, along with other conventional methods of hydrological assessment, have been used to assess potential water availability for the watersheds of Grand River Du Nord and Trou Du Nord. Grand Rivere du Nord: The hydrological basin area of Grand Rivere du Nord is estimated to be approximately 611 square kilometers. Through basin delineation and hydrological model development process, fifty-one (51) sub basins were developed for this basin, which also included 25 stream reaches. Trou du Nord: The basin area of main stream of Trou du Nord is estimated to be approximately 106 square kilometers. Through basin delineation and hydrological model development process in all 11 sub basins were developed for this basin, which also included 5 stream reaches. 1 http:// sdwebx.worldbank.org/climateportalb/doc/GFDRRCountryProfiles/ wb_gfdrr_climate_change_country_profile_for_HTI.pdf = EMERGING = Lo s4 * SUSTAINABLE EX EU *IDB ERM a5 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 222] A water balance model can be considered as a system of equations designed to represent various aspects of the hydrological cycle. À bucket model has been used in the assessment of water balance that considers a unit area (such as watershed) as a bucket, which is filled by the rainfall and emptied by evapotraspiration. When bucket is full, extra water is assumed as deep drainage (usually some part goes to groundwater). This model requires a recorded rainfall and evapotranspiration. A simplified concept of water balance is shown in Figure A4.35. ET P RE me + Q Figure A4.35 Simplified Concept of hydrological water balance Where, S = water storage (as ground water recharge); ET = evapotranspiration; P = rainfall; Q = Surface runoff and base flow. In mathematical terms, the bucket model can be represented as follows. P-ET-Losses = Q+S The losses in the above equation include the interception, infiltration and percolation. The water balance estimates have performed at a daily level. Water availability potentials have been summarized on monthly and annual basis using the outcomes of the daily level calculations. The details of the estimation of various components of the cycle are given below. ° Rainfall: The rainfall is often the largest term in the water balance equation, which forms as major source of input to the system. The historically recorded monthly rainfalls have used in the analysis. As described in the flood hazard assessment and methodology, monthly rainfall data has been distributed to the daily time step. Overall the study area has average annual rainfall of about 1400 mm for current conditions and 1600 mm for climate change. e Evapotranspiration: Evapotranspiration combines the two terms evaporation (generally from the ground surface) and transpiration (generally from the plants). This term forms one of the largest sources of losses from the system. Evapotranspiration can be estimated from meteorological and soil moisture data or measured directly. In this analysis available monthly estimates have been used to derive an average annual evapotranspiration rate. In the study area, it is estimated to be approximately 1600 mm for current climate conditions and approximately 2000 mm considering climate change projections. ° Surface Runoff: Surface runoff parameters have been estimated by using the hydrological models to assess surface runoff. Using rainfall and evapotranspiration data, effective rainfall values have been estimated. These effective rainfall values have been used in the estimation of surface runoff and other abstraction losses. The losses include the interception, infiltration and percolation. The annual average surface runoff for the study area has been estimated at 165 mm for current conditions and 141 mm == EMERGING = CR SUSTAINABLE IDB LL ERM 46 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 223] taking in account climate change projections. ° Ground Water Recharge: The losses pertaining to infiltration and percolation are often termed as the abstraction losses. These abstraction losses in the soil strata are divided into two parts 1) baseflow and 2) groundwater recharge. Base flow is the lateral sub-soil flow, which usually joins the surface runoff and is generally a small portion in the hydrological cycle. The vertical component of sub-soil flow, which moves down, is known as ground water recharge. The estimated value of annual average ground water recharge for the study area is 195 mm for current climate conditions and 288 mm considering climate change projections. Figure A4.36 and A4.37 provide a monthly snapshot the parameters of hydrological cycle in study area for current conditions and climate change respectively. The projected climate change estimates also considers the change in land use over years. Ë 150 — € n $ JT D = s I] N > _ — 3 50 \ L RTE o ESS 7 Jan Feb Mar April May June July Aug Sep Oct Nov Dec Month —— Rainfallkmm —— Runoff, mm > Recharge, mm => £Evapotranspiration, mm Figure A4.36 Monthly parameters of hydrological cycle for current conditions = EMERGING = Lo L Initiative. ERM 47 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 224] : _e £ ALT E 150 A = D ET NA É 100 IN L / 50 TE NV = MED Jan Feb Mar April May June July Aug Sep Oct Nov Dec Month —— Rainfallkmm —— Runoff, mm —% Recharge, mm —— £vapotranspiration, mm Figure A4.37 Monthly parameters of hydrological cycle for climate change projections Water Availability Based on the hydrological estimates as summarized above, the water availability potential has been estimated on volumetric terms (million cubic meters). These numbers have been estimated over the entire study area for current climate conditions and projected to consider climate change. The climate projections suggest the increase in the rainfall by 13% and temperature increase by 2°C. The scenario of climate change also consider changes in land use based on parameters derived from CE3, which introduced sustainable land use planning recommendations as a risk reduction measure. Due to combined impact of climate change and changes in land use, the surface runoff will decrease, ground water recharge will increase. Monthly values of water availability potential are summarized in Table A4.23. In both the cases the ground potential is more than the surface water potential. These numbers are estimates of potential, which needs to be harnessed by developing them. Table A4.23 Monthly variations in water availability potential = EMERGING = Lo L Initiative. ERM 48 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 225] Surface Water Groundwater Total Surface Water | Groundwater a : . : : : Total Potential, Potential, Potential, Potential, Potential, Potential, Mm° Mm° Mm° Mm° Mm° Mm° Laor | 08 | ma | m2 | 83 | 9 | 22 | Lay | 60 | wa | ina | 47 | 1584 | 22 | un | 08 | 16 | 24 | 05 | 43 | us | au | 01 | 03 | 04 | 01 | 05 | 0 | au | 06 | 08 | 14 | 08 | 08 | 13 | sep | 08 | 14 | 22 | 06 | 16 | 22 | 103.7 1913 1529 228.0 Water Demands Considering the developmental scenarios, two projections have been used to assess water demand in the study area (Table A4.24). Table A4.24 Developmental scenarios for water demands Data Element Current (2013) Climate Change and CE3 Land Use Projections, 2040 Urban Demand (Epatal Fe Agriculture Land Use (spatial), ha 12630 17,000 Industrial Demand, sq m 67,000 229,000 The water demands include domestic demands from rural and urban areas for domestic uses, industrial demands and agricultural or irrigation requirements. The domestic use includes all of the expected per-capita use by person for activities such as drinking, showers, faucets, and toilets. A demand of 170 liters per day per head and demand of 80 liters per day per head have been estimated for urban and rural areas respectively. Individual demand estimates are based on earlier water assessments conducted for the Haiti (USACE, 1999), and for the industrial park (IDB, 2011). These studies are consistent with projections cited in global studies. Industrial demand is estimated on per square meter basis based on current utilization and future projections for the PIC and other targeted development projects for the study area. This estimates industrial utilization to approximately 100 liters per square meter of the industrial area. Other water use and demands have considered as 250 liter per day per head. The agricultural or irrigation requirements have also been estimated using the crop area, crop coefficients (kc) and evapotranspiration. 3 EMERGING + Lo HS GIDB Initiative ERM 49 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 226] For the future development projections, climate change projections have been incorporated. All the demands have estimated on the annual time step on volumetric basis. Table A2.25 shows the water demands for the current and future development scenario as contemplated by CE3. Table A4.25- Summary of Water Demands Climate Change and CE3 Land Use Water Demand, Mm3 : 8 : Projections, 2040 Urban Population, M3 Rural Population, M3 124 | 60 industrial Demand, Mm3 124 | 84 Water Use and Demand ,Mm3 Agricultural Demand, Mm3 101.0 163.2 Total Demend,Mm3 112.6 188.8 Water Balance Based on the estimates of water availability potential and demands, overall summary of water balance is shown in Table A4.26. As represented in the table, the current water availability potential is considerably more than the demands. But in future projections, the water availability potential is merely sufficient to meet the projected demands. Table A4.26 Summary of water balance of study area Climate Change and CE3 Land Use Projections, Current 2040 Water Demand, Mm° Urban Population, M? Rural Population, Mn? ES EC Industrial Demand, Mm* Water Use and Demand, M? Agricultural Demand, Mm° 101.0 163.2 Total Demand, Mm° 112.6 188.8 Water Availability Potential, Mm Surface Water Potentia (Runoff), M! Ground Water Potential (Recharge), Mm° 103.7 152.9 Total Water Potential, Mm° 191.3 228.0 Surplus/Deficit, Mm° Water Surplus, M! The monthly variations in the availability and demand potentials are shown in Figure A4.38 for present day, while Figure A4.39 shows the monthly variation for future growth scenario that takes into consideration climate change. Monthly variation shows that during the dry season of the year, the gap between demand and availability increases as compared to the wet season. 3 EMERGING + Lo Initiative ERM 50 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 227] Do A AE RE A AE A A A EE EE EE 40.0 & so Lt tt | ll | | | £ 35.0 N 2 00 tt | | | | | [JR PNG MEE RE £ 25.0 ù ; oo et tt tt pl € 20.0 ne. 2 15.0 8 00 LE EN | | | | /) | | £ 10. ——_—_—__—— | ——— DO A A À A A PE EE oo Lt tt | LT | Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Month —— Available Potential, Mm3 —— Demand, Mm3 Figure A4.38 Monthly variation in demand and availability for current conditions 45.0 po AE RE AE AE AE ES ES ES EE UE pepe A AE A EE RE ES ES ES ES ET D CE 3 300 NT OO [= PNA EEE DIE 8 00 Lt TNT OT OT | | | A 20. DO D AR A A A À S a ———— ————_—_—_—_— ——— E 7 \ 8 00 Lt À | | | 7 ol tt LT NT TT TT mot tt tt | Ni | | | Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Month — Available Potential, Mm3 —— Demand,Mm3 Figure A4.39 Monthly variation in demand and availability with climate change projection From the hydrological drought assessment conducted for study area, it can be concluded that on annual basis, water availability potential is sufficient to meet the current as well as future demands. However, as shown in the Figures A4.38 and A4.39 above, on monthly basis, there is a significant gap between the demand and availability potential particularly during the dry season (June to October) as compared to the wet season. As seen from the figures, during the current and future scenarios, there is a considerable water deficit during the dry period indicating prolonged period of hydrological drought. Climate change and projected growth will reduce the available water stock and make the impacts of prolonged periods of water deficit which is likely to impact agricultural production, further exacerbate food storages, reduce water quality, and increase land degradation. Such impacts will become more pronounced during years with below average rainfall. 4 GIDB ELA ERM 51 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 228] A4.7 References American Institute of Architects, Inter-American Development Bank, USAID / OTI, 2012, Cumulative Impact Assessment For Regional Development In The Cap-Haïtien To Ouanaminthe Urban Corridor accessed at http://www.ute.gouv.ht/caracol/images/stories/docs/Al_CIA_Final_26-08-12.pdf) Bender, M.A., Knutson, T.R., Tuleya, RE. Sirutis, J.J., Vecchi, G.A., Garner, S.T., Held, I.M. (2010). Modeled Impact of Anthropogenic Warming on the Frequency of Intense Atlantic Hurricanes, Science 327, 454-458, doi:10.1126/science.1180568. Bengtsson, L., Hodges, K.I., Esch, M., Keenlyside, N., Kornblueh, L., LUO, J. J., & Yamagata, T. (2007). How may tropical cyclones change in a warmer climate?. Tellus A, 59(4), 539-561. Boore, D. M. and W. B. Joyner (1997).Site amplifications for generic rock sites, Bull. Seismol. Soc. Am. 87, 327-341. Boore D. M., Thompson E. M. and cadet H., (2011), Regional Correlations of VS30 and Velocities Averaged Over Depths Less Than and Greater Than 30 Meters, Bulletin of the Seismological Society of America, Vol. 101, No. 6, pp. 3046-3059, December 2011, doi: 10.1785/0120110071 Building Seismic Safety Council (BSSC), 2001 Nehrp Recommended Provisions For Seismic Regulations For New Buildings And Other Structures 2000 Edition, Part 1: Provisions (Fema 368), Washington, D.C., 2001. Calais, E. 2001. Vers un projet "Aléa sismique en Haïti". CNRS ; Géosciences Azur. Sophia Antipolis, France. Unpublished report; 11pp. Campbell K., Abrahamson N., Power M. Chiou B., Bozorgnia Y., ShantzT., and Roblee C. (2009), Next Generation Attenuation (Nga) Project: Empirical Ground Motion Prediction Equations For Active Tectonic Regions, Sixth International Conference on Urban Earthquake Engineering, March 3-4, 2009, Tokyo Institute of Technology, Tokyo, Japan Atlas of Probable Storm Effects in the Caribbean Sea Choi, Y., and J. P. Stewart (2005), “Nonlinear site amplification as function of 30 m shear wave velocity”, Earthquake Spectra, vol.21, no. 1, pp.1-30 Donald L. Wells and Kevin J. Coppersmith (1994): New Empirical Relationships among Magnitude, Rupture Length, Rupture Width, Rupture Area, and Surface Displacement Bulletin of the Seismological Society of America, Vol. 84, No. 4, pp. 974-1002, August 1994. Elsner, J. B., Kossin, J. P., & Jagger, T. H. (2008). The increasing intensity of the strongest tropical cyclones. Nature, 455(7209), 92-95. Emanuel, K. A. S. Ravela, E. Vivant and C. Risi (2006), A Statistical determinstic approach to hurricane risk assessment,Bull Amer Meteor Soc., 19, 299-314. Emanuel, K.A. (1998). The power of a hurricane: An example of reckless driving on the information superhighway. Weather 54, 107-108. Engdahl, E.R., and A. Villaseñor, Global Seismicity: 1900-1999, in W.H.K. Lee, H. Kanamori, P.C. Jennings, and C. Kisslinger (editors), International Handbook of Earthquake and Engineering Seismology, Part À, Chapter 41, pp. 665-690, Academic Press, 2002. Famine Early Warning Systems Network, HAITI Food Security Outlook Update, June 2012 = EMERGING = EU ERM 52 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 229] Food and Agriculture Organization of the United Nations, 2004. Technical Cooperation Programme. Project Title: Assistance to improve Local Agricultural Emergency Preparedness in Caribbean countries highly prone to hurricane related disasters. Frankel A., Harmsen S., Mueller C., Calais E. and Haase J., 2011, Seismic Hazard Maps for Haïti, Earthquake Spectra, Volume 27, No. S1, pages S23-S41 Frankel, A., Mueller, C., Barnhard, T., Leyendecker, E., Wesson, R., Harmsen, S., Klein, F.,Perkins, D., Dickman, N., Hanson,S., and Hopper, M., 2000, USGS national seismic hazard maps: Earthquake Spectra, v. 16, p. 1-19. Frankel, Arthur, Harmsen, Stephen, Mueller, Charles, Calais, Eric, and Haase, Jennifer, 2010, Documentation for initial seismic hazard maps for Haiti: U.S. Geological Survey Open-File Report 2010-1067, 12 p. Frankel, A.D., 1995, Mapping seismic hazard in the central and eastern United States: Seismological Research Letters, v. 66, p. 8-21. Gualdi, S., Scoccimarro, E., & Navarra, A. (2008). Changes in tropical cyclone activity due to global warming: Results from a high-resolution coupled general circulation model. Journal of climate, 21(20), 5204-5228. IDB, 2011, Preliminary Hydrological Assessment for the Development of an Industrial Park in Haiti, Prepared for: Inter- American Development Bank Washington, D.C. Prepared by: ENVIRON International Corporation, Washington, D.C. IPCC AR4, 2007, The IPCC Fourth Assessment Report of the Intergovernmental Panel on Climate Change IPCC AR5, 2014, IPCC Fifth Assessment Report of the Intergovernmental Panel on Climate Change Knutson, T.R., Sirutis, J. J., Garner, S. T., Vecchi, G. A., & Held, 1. M. (2008). Simulated reduction in Atlantic hurricane frequency under twenty-first-century warming conditions. Nature Geoscience, 1(6), 359-364. Kossin, J. P., Knapp, K.R., Vimont, D. J., Murnane, R. J., & Harper, B. A. (2007). A globally consistent reanalysis of hurricane variability and trends.Geophysical Research Letters, 34(4). Landsea, C. W., Feuer, S., Hagen, A. Glenn, D. A. Sims, J., Perez, R., … & Anderson, N. (2012). À Reanalysis of the 1921-30 Atlantic Hurricane Database*. Journal of Climate, 25(3), 865-885. MARNDR, MPCE, MICT, MDE. 2000. Actes de l’atelier de concertation interministérielle pour la gestion des bassins versants. 120 pp. McCann, W.R., 2006, Estimating the threat of tsunamagenic earthquakes and earthquake induced landslide tsunamis in the Caribbean, in Aurelio, M., and Philip, L., eds., Caribbean tsunami hazard: Singapore, World Scientific Publishing, p. 43-65. McCann, W. 2001, Amenaza de Terremoto en La Hispaniola, Conferencia Internacional Sobre Reduccion de Riesgo Sismico en la Region del Caribe, Segundo Seminario Dominicano de Ingenieria Sismica, July, 2001, Santiago, Dominican Republic. NATHAT 2010, Analysis of Multiple Natural Hazards in Haïti. (NATHAT). Port-au-Prince, Haïti. March 26, 2010. Government of Haiti. Nicholls, R. J. (1998). Assessing erosion of sandy beaches due to sea-level rise. Geological Society, London, Engineering Geology Special Publications,15(1), 71-76. = EMERGING = EU ERM 53 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 230] Oouchi, K., Yoshimura, J., Yoshimura, H., Mizuta, R., Kusunoki, S., & Noda, A. (2006). Tropical cyclone climatology in a global-warming climate as simulated in a 20 km-mesh global atmospheric model: Frequency and wind intensity analyses. J.Meteorol.Soc.lpn, 84(2), 259-276. RMS FAQ: 2010 Haïti Earthquake and Caribbean Earthquake Risk, Accessed at https://support.rms.com/publications/Haïiti_Earthquake_FAQ.pdf Sandikkaya M.A., Akkar S., and Bard P-Y., (2013), À Nonlinear Site-Amplification Model for the Next Pan-European Ground-Motion Prediction Equations, Bulletin of the Seismological Society of America, Vol. 103, No. 1, pp. 19-32, February 2013, doi: 10.1785/0120120008 Schott C., and Schwarz J., (2004), Reliability Of Eurocode 8 Spectra And The Problems Of Their Application To Central European Earthquake Regions, 13th World Conference on Earthquake Engineering, Vancouver, B.C., Canada, August 1-6, 2004, Paper No. 3403 Sugi, M., Murakami, H., & Yoshimura, J. (2009). A reduction in global tropical cyclone frequency due to global warming. Sola, 5(0), 164-167. The World Bank, 1991, Haiti Agricultural Sector Review Report No. 9357-HA UniQ, 2010. Projet de création d’un Parc Industriel dans la région Nord-Est d'Haïti. Analyse géologique & hydrogéologique des sites sélectionnés, Rapport final. Université Quisqueya, Port-au-Prince. UN Global Study - UNISDR (2013) From Shared Risk to Shared Value -The Business Case for Disaster Risk Reduction. Global Assessment Report on Disaster Risk Reduction. Geneva, Switzerland: United Nations Office for Disaster Risk Reduction (UNISDR) USACE, 1999, Water Resources Assessment of Haiti, US Army Corps of Engineers, Mobile District and Topographic Engineering Center. United Nations Strategy for Disaster Reduction (UNISDR) and Global Assessment Report on Disaster Risk Reduction (GAR) report = > UNISDR (2013) From Shared Risk to Shared Value -The Business Case for Disaster Risk Reduction. Global Assessment Report on Disaster Risk Reduction. Geneva, Switzerland: United Nations Office for Disaster Risk Reduction (UNISDR) USGS 2010. Frankel, Arthur, Harmsen, Stephen, Mueller, Charles, Calais, Eric, and Haase, Jennifer, 2010, Documentation for initial seismic hazard maps for Haiti: U.S. Geological Survey Open-File Report 2010-1067, 12 p. USAID, 2007, Environmental Vulnerability In Haïti Findings & Recommendations, http://pdf.usaid.gov/pdf_docs/PNADN816.pdf Vickery, P.J. P.F., Skerij, A.C. Stekley and L.A. Twisdale Jr. 2000, Hurricane wind field model for use in Hurricane Simulations. J Struct Eng 126. 1303-1221. Vickery, P. (2008), Development of Design Wind Speed Maps for the Caribbean for Application with Wind Load Provisions of ASCE 7 ARA Rep. No. 18108-1, Pan American Health Organization, Regional Office for the Americas World Health Organization, Disaster management Programme, 525 23rd Street, NW Washington, DC. Vickery, P. (2012) Design Wind Speeds in the Caribbean. Advances in Hurricane Engineering: pp. 1136-1147. doi: 10.1061/9780784412626.099 Vickery, P.J., 2005. Simple empirical models for estimating the increase in the central pressure of tropical cyclones after landfall along the coastline of the United States. J. Appl. Meteorol. 44, 1807-1826. = EMERGING = Lo CR SUSTAINABLE “IDB EU ERM 54 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 231] Wald, D. J. and T. I. Allen (2007). Topographic slope as a proxy for seismic site conditions and amplification, Bull. Seismol. Soc. Am. 97,1379-1395. Wald, D.]J. B. Worden, V. Quitoriano, and K. Pankow, (2004). Shake Map Manual: Technical Manual, Users Guide, And Software Guide, U.S. Geological Survey Open-File Report, in preparation Walling, M, Walter Silva, and Norman Abrahamson (2008). Nonlinear site amplification factors for constraining the NGA Models, Earthquake Spectra 24:1, 243-256 Watson, C. C. and M. E. Johnson. (1999). “Design, Implementation, andOperation of a Modular Integrated Tropical Cyclone Hazard Model,” AMS 23rd Conference on Hurricanes and Tropical Meteorology, Dallas, TX. Watson, C. C., Jr. (1995), "The Arbiter of Storms: À High Resolution, GIS-based System for Integrated Storm Hazard Modeling," National Weather Digest, 20, 2-9. Wills, C. J., Petersen, M. Bryant, W. A. Reichle, M. Saucedo, G. J., Tan, S., Taylor, G., and Treiman, J., 2000. A site- conditions map for California based on geology and shear-wave velocity, Bull. Seismol. Soc. Am. 90 (6B), S187-S208. == EMERGING = EU ERM 55 ESCI HAITI — APPENDIX 4: HAZARD PROFILES [page 232] APPENDIX 5: Characteristics of Assets Exposed EMERGING = es GIDB © NORTHERN DEVELOPMENT CORRIDOR, HAITI Appendix ste ERM [page 233] A5 CHARACTERISTICS OF ASSETS EXPOSED A5.1 Context The inventory of exposed assets involves understanding the distribution of people, buildings and infrastructure that may be affected by natural phenomena. In the Northern Development Corridor of Haiti, there was not enough detailed information available to carry out a site specific assessment of all buildings, and it was therefore necessary to carry out a rapid field assessment method to estimate the number and distribution of assets in the study area. A review of various reports and geospatial databases was conducted in order to ascertain useful information for understanding land uses and building practices in Haïti, and in the study region. Initially it was perceived that reports and databases would provide information that would be useful for ascertain the distribution buildings and infrastructure and their characteristics. However, much of the information focused on damages that occurred as a result of the 2010 Earthquake and was confined to dense urban areas of the capital. Damage assessment reports, nevertheless, provided useful information to understand building practices. The review of geospatial data helped us to understand land uses, population distribution and density, and the location of critical facilities and infrastructure. Remote sensing data as well as information gathered from a rapid field assessment were utilized to capture priority building information. Within the study area, satellite images are interpreted to understand building density and types for each land use category and administrative boundary. Administrative boundaries (i.e. section communal) were then subdivided based on density of building footprints so as to allow for the definition of an appropriate scale from which to capture inventory elements. This mapping process provided the basis for classifying buildings and for using a suitable classification hierarchy for the capture of a wide range of structures and densities. A rapid field assessment that was conducted in August, 2013 informed the interpretation of remote sensing data and facilitated the classification of buildings into general occupancy classes. Photographic survey was utilized to inform the development of an exposure model so as to incorporate a number of statistical assumptions to calculate the building type (house or apartment), age, structure (wall and roof type, number of stories), construction and to estimate replacement costs. Such an approach is consistent with standard methodologies used to develop exposure models and supports the required inputs for undertaking a probabilistic risk assessment by providing an approximate spatial location of exposed elements for each block subdivision within the section communal. A5.2 Building Base Map for Inventory Assessment Haiti is divided into ten (10) departments. The departments are subdivided into arrondissements, which are further subdivided into communes, which in turn are divided into sections communales. Because the section communales represent too crude of an administrative boundary from which to categorize the built environment to facilitate a mapping of exposed elements and risk, the section communales have been further subdivided into blocks to facilitate analysis at finer level. The block boundaries have been created by overlaying commune boundaries on the remote sensing data in a GIS so as to facilitate an understanding of the density buildings and distribution of different building types throughout the study area. == EMERGING = «#% GIDB À LL ERM 1 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 234] NORTH ATLANTIC OCEAN A” Had où AD 2 hs LE EE DE: ES gi Ke — RS & CR æ: TILRIT | RE R LA) Pi | ne DIN ES — - Lo () COR V7 co: We x mr mou / \ CT meme cd _ 1] pérnone une —— 29 cit cs Car maéren Figure A5.1 Distribution of block boundaries in the study area Each section communale was divided into multiple blocks by taking into consideration building density and road network. Smaller blocks have been created in the cities where there are higher building densities, while in less dense rural areas larger blocks have been delineated. In most of the cases road lines have been followed for creating blocks boundaries to facilitate a consistent mapping across the study area. The total number of blocks in each section communale. Table A5.1 Number of blocks for each section communale Section Communales No of Blocks RS EE #7 <s EMERGING” di «HE GIDB L L Initiative ERM 2 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 235] A5.3 Building Occupancy Mapping and Distribution The field survey conducted in August 2013 facilitated an understanding of general building distribution in the study area. A building distribution schema was developed to facilitate an understanding of the distribution of buildings and uses. Using remote sensing data, attention was given to densities and texture information to each sub-divided administrative unit (i.e. Block) to identify homogenous land uses where the distribution of buildings would be similar. By using remote sensing data, each block was visually checked to determine the distribution of building types and to identify any anomalies so as to aggregate building information to each block. The development of a building mapping schema, which was informed by field work, allows for a more precise definition of the expected distribution of different types of buildings and infrastructure in the study area. It also informed the distribution of building uses or occupancy. The schema, coupled with building footprints data, facilitated a general understanding of the square area of buildings, which is important for determining the generalized replacement value of each structure. This will facilitate that application of standard construction costs per building type and allow for exposure values to be aggregated by administrative unit. Statistical tools were used to help determine that the final results are in line with the sample data. In addition, the building distribution schema was reviewed graphically in a mapping interface to assure that the results were reasonable. This was an involved process, where generalized distribution was checked against homogenous zones that were investigated in the field, to make sure that the number and distribution of buildings was appropriate. Structure Classification Structural information is an important factor in determining the vulnerability or how likely structures are to fail when they are subjected to hazards, such as wind pressure that exceeds their design. In order to conduct basic analyses and gather information useful to determine general loss estimates, structural engineers and planners categorized buildings into the eight (8) different structure types, which are similar to those identified by Institut Haïtien de Statistique et d'Informatique (IHSI) of the Ministère de l'Economie et des Finances. This was done to capture the general features of the structures of the buildings according to local current construction practices. The models of the different types of infrastructure were determined based on experience with the typical construction of Haiti. The basic structural systems were grouped according to the following general construction: reinforced concrete, masonry structure, unreinforced masonry, and earthen. The field analysis, however, indicated that there was a need to slightly modify building classifications outlined by the IHSI. À new structure distribution schema (shown in Table A5.2 below) was developed for the Northern Development Corridor, which were reviewed and verified by architects and civil engineers familiar with the study area. == EMERGING = Lo s4 * SUSTAINABLE EX EU % IDB ERM 3 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 236] Table A5.2 Structure Classification Matrix in Northern Development Corridor of Haiti L'mauerne [cie | au [ee [re] us | met | raniair | Low Rise ST-1 Poor 5-10 yrs 1 Reinforced Wood Mass Reinforced Columns, block Frame, Masonry/ Concrete in-fill Corrugated Rubble Metal Low Rise ST-2 Moderate | 5-10 yrs 1 Reinforced Concrete Concrete Slab Reinforced Columns, block Slab Concrete in-fill Mid Rise ST-3 Poor 5-10 yrs 2 Reinforced Concrete Mass Reinforced Columns, block Slab Masonry/ Concrete in-fill Rubble Mid Rise ST-4 Moderate | 5-10 yrs 2/3 Reinforced Concrete Concrete Slab Reinforced Columns, block Slab Concrete in-fill Low Rise Steel ST-5 Good 1-5 yrs 1 Slab on Grade Concrete Concrete Slab Frame Slab Mid Rise Steel ST-6 Good 1-5 yrs 2/3 Slab on Grade Concrete Concrete Slab Frame Slab Low Rise ST-7 Moderate | 20 Yrs+ 1 Mass Masonry Wood Mass Masonry Frame, Masonry/ Corrugated Rubble Metal Mid Rise ST-8 Moderate | 20 Yrs+ 2/3 Mass Masonry/ Wood Mass Masonry Concrete Frame, Masonry/ Corrugated Rubble Metal Wattle and ST-9 Poor 20 Yrs + 1 Earth, covered Wood Earth/Rock Daub/Adobe with light Frame, concrete plaster | Corrugated Metal Each building type has a unique and distinct behavior, due to a number of factors including the behavior of the material, height, pre and post disaster incorporation of new building requirements, and quality of prevalent construction practices. Building Replacement Values To relate the number of building and occupancy classes to specific building types, a two-dimensional matrix was developed. It allowed project team members to understand the distribution of identified building types for a specific occupancy class for each block. To estimate replacement values for buildings in each block, local construction parameters were developed by a review of local construction practitioners (i.e. architects, engineers) to understand the construction cost (USD/m2) for the following occupancy classes and are provided in Table A5.3. = EMERGING + Lo CR SUSTAINABLE “IDB Initiative ERM 4 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 237] Table A5.3 Construction Costs, Occupancy Type Average Construction Cost Occupancy Class . Estimates (USD/M°) Residential (formal) $800 to $1100 Housing that goes through the formalities of architecture plans, permits, etc. Residential $600 to $800 Housing that would be self-built or with the help of skilled personal (informal 1) Residential $450 to $500 Housing that would be self-built without help of skilled personal (informal 2) Residential (international $800 to $1,200 Housing that has been built or USAID or other international donors donor) This may include construction for commercial enterprises which Commercial $1000 follows formal construction process of developing architectural plans and permits and is constructed by professional contractors. Only construction costs required. This may include construction for industrial enterprises (i.e. : ; agricultural processing facilities), which will follows formal Industrial (outside : . : A ark) $1000 construction process of developing architectural plans and permits p and is constructed by professional contractors. Only construction costs required Industrial (PIC) $1200 Construction costs for the development of industrial production facilities inside the park. The identification and understanding of construction parameters as well as field work assisted in the determination of values for urbanized or built-up areas by occupancy class. Image analysis informed a determination of building counts for certain building type clusters found in rural areas. Inventory Aggregation and Valuation This methodology facilitates the development of occupancy-building type mapping schemes which help determine the distribution and aggregate exposure for general building stock. Specific values estimates and site specific geospatial data were collected to understand specific exposure values for infrastructure and critical facilities. General Building Stock The composition of the general building stock (i.e. residential, commercial and industrial building stock) was therefore aggregated to the given block boundary. As previously described, its distribution was determined by field work and careful review of imagery so as to provide the most accurate distribution of building types for block boundary. = EMERGING Lo “## GIDB LL ERM 5 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 238] raw raw rw rvse mens svun vou ASEN “ Legend Residential Exposure Value in USD { _lo-228312 DIN 2.283123 - 6.372.369 Fes 1vsen] | 6:372,370 - 13,757,970 MM 3757971 -23400005 M 2:20 - 4713467 ss, C2] Haït Boundary raw Er EI pe mr Figure A5.2 Distribution and Exposure Values of Residential Buildings in the Area of Study ru mew ru ren mere ou 1940N- TT en Legend Commercial Exposure Value in USD | l0-433073 NN 433,974 - 1238,622 7m sos] | 1,258.623 - 2,544,736 MM 254737 - 5,606,387 Er mn rm Tin men me Figure A5.3 Distribution and Exposure Values of Commercial Buildings in the Area of Study L EMERGING 4 GIDB Initiative ERM 6 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 239] new rw rw nsew rvsaw . avan 3 Esoson va A73EN ie Legend Industrial Exposure Value in USD [| 0-480454 UN 480.455 - 2,055,989 ASTM send | 2,055,900 - 6,612,628 M 5612.62 - 23.756.844 me PR L onn Tw En an new nm Figure A5.4 Distribution and Exposure Values of Industrial Buildings in the Area of Study This analysis, which is consistent with HAZUS-MH Level 2 Analysis’, was conducted using a Geographic Information System, and Table A5.4 presents the results which were aggregated for both the Commune (Administrative Level 3) and Section Commune (Administrative Level 4) administrative boundaries in the Northern Development Corridor. L'Atthis point, some basic background information about methodology utilized to develop the inventory. The methodology undertaken is undertaken is consistent with methodologies used for HAZUS-MH. There can be three levels dependent on the amount of data that is available: Level 1 analyses use default data that has been assembled for HAZUS-MH from US national databases, which is not applicable outside the United States. This data provides a basic estimate of exposure that is useful for broad scale planning efforts. Level 2 analyses involve the input of more detailed data about local conditions, building stocks etc. and yields correspondingly more detailed and useful results. Likewise, Level 3 analyses provide the most accurate estimates of losses but require detailed engineering and physical conditions data to customize results at a level that is not clearly warranted by this type of study. _ EMERGING NE # Le MABLE SZ initiative ERM 7 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 240] Table A5.4 Economic value (Million USD) for General Building Stock, by Occupancy types Commune/Section Residential Commercial Industrial Commune (Million USD) (Million USD) (Million USD) RE PS ES PE RE RE PES RE ES RE PES AS RE ES Facilities and Infrastructure For purposes of this study, the following three part definition of critical facilities and infrastructure has been applied: Critical Facilities - are those facilities that provide services to the community and should be functional after a hazard event. They include: e Hospitals o Hospitals (Level 1 - Major Medical Facility) o Hospitals (Level 2 - Medical Facility/Clinic) o Hospital (Level 3 — Rural Clinic) + Education o Kindergarten o Primary o College o University Transportation Infrastructure - enables the movement of goods, particularly goods and emergency relief supplies. They include: + Roads, km (Highway) + Roads, km (Secondary) + Roads, km (Tertiary) + Bridges Utilities and Infrastructure - are facilities that, if damaged, could have far-reaching consequences for the environment. They include: e Electric Infrastructure o Electric Power Plant == EMERGING + ÉUL ERM 8 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 241] o Electric Lines, km e Water Infrastructure o Water Lines, o Water Pumping Stations, o Reservoirs/Catchment, and o Wells e Waste Water Infrastructure o Culverts, o Waste Water Treatment Plants Facilities and infrastructure have been categorized facilities and infrastructure by their structural characteristics relevant to vulnerability to the prominent hazards identified in the study. Replacement and content values for facilities were evaluated based on field inspections which allowed for approximation building area and construction cost (i.e. exposure) by facility type or infrastructure class. Facility and infrastructure costs were informed by civil engineers familiar with construction practices in Haiti. Table A5.5 Estimated Value of Critical Facilities and Infrastructure Class indergamen | 5 | am | unpersy | 5 | mm | —s. EMERGING = Lo TES GIDB Initiative ERM 9 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 242] APPENDIX 6: Impacts and Losses EMERGING — c<# GIDB À NORTHERN DEVELOPMENT CORRIDOR, HAITI Appendix Mtts ERM [page 243] A6 IMPACTS AND LOSSES A6.1 Methodology This section presents an estimate of the losses attributable to each hazard. The findings can be used to support local and regional planners understanding of the potential impacts of each hazard and allow a comparison of hazards by quantifying potential impacts. The application of a risk assessment methodology has resulted in an approximation of risk. These estimates should be used to understand relative risk from hazards and potential losses; however it is important to understand that uncertainties are inherent in any loss estimation methodology, arising in part from incomplete scientific knowledge concerning natural hazards and their effects on the built environment. Uncertainties also result from approximations and simplifications that are necessary for a comprehensive analysis (such as abbreviated inventories, and model parameters such as precipitation data or economic parameters). Risk Metrics The economic loss results are presented here using three risk indicators: e Probable Maximum Loss (PML), which provides an estimate of losses that are likely to occur, considering existing mitigation features, due to a single hazard event; e _Loss Exceedance Curve which plots the consequences (losses) against the probability for different events with different return periods; and e Average Annualized Loss (AAL), which is the estimated long-term value of losses to assets in any single year Within the study area. By annualizing estimated losses, we understand historic patterns of events so as to provide a balanced assessment of risk. The AAL is the summation of products of event losses and event occurrence probabilities for all stochastic events in a loss model and is expressed as: Average Annual Losses= };P;L; The use of the annualized losses approach has two primary benefits, including: the ability to assess potential losses from all future disasters; and provides an objective means to evaluate mitigation alternatives. Risk Mapping The risk metrics described above, particularly, the AAL, can be used to provide an understanding of the spatial extent of losses and help to identify and prioritize the urban areas or localities that are under risk. The street light indicator methodology, where the colors on the map coincide with the level of risk, has been used to map risk: By mapping risk at the block level, stakeholders have a better understanding of where potential losses will be the highest and where monies should be allocated for risk reduction. All the areas and exposure categories that have a high and very high risk are automatic choices for risk reduction measures. == EMERGING TRS SUSTAINABLE IDB Initiative ERM 1 ESCI HAITI — APPENDIX 6: LOSS ESTIMATION [page 244] A6.2 Earthquake Hazard Earthquakes have the potential to cause damages due to waves that are produced by a release of energy through the earth. The way the earth responds to the energy of an earthquake depends on the geology of the area. Shaking may not cause damage to the ground itself, but may cause significant damage to structures including infrastructure and lifelines. The expected intensity of an earthquake in the area is significant with an expected peak ground acceleration (PGA in g) with site amplification varying from 0.244 to 0.448 for 475 years return period and 0.406 to 0.853 (in g) for 2500 year return period. The Northern Development Corridor of Haïti is almost entirely comprised of alluvial deposits (i.e. areas of wet sand or silt), which makes it susceptible to liquefaction ( i.e. where the ground could be shaken enough that the pressure of the water in the soil builds up to make it liquefy). In addition to physical geography, the potential impacts of the earthquakes in northern Haiti are tied to the construction practices and materials. The pattern of construction that is found throughout the areas is the prevalent factor influencing the physical vulnerability of the area (ï.e. structural damages). The majority of the buildings are low rise structures (i.e. one story) constructed out of one of four materials: concrete/blocks, earthen materials, ‘clisse’ (translated “woven wood mats”), or bricks/rocks. The construction of walls and flooring is with little or no reinforcement, which are the key factors that influenced the collapse of numerous buildings and loss of life during the catastrophic seismic event in 2010. Estimated Probable Losses The estimates for the earthquake hazard are provided in Table A6.1 below, and represent the maximum probable losses for general occupancy classes. Losses are presented for two return periods. Table A6.1 Probable Maximum Losses (PML) for Earthquake Hazard Loss (106 USD) Return Period Years | Residential | Commercial | Industrial | 2500 1,071.5 194.7 161.0 == EMERGING TRS SUSTAINABLE BIDB Lo EU ERM 2 ESCI HAITI — APPENDIX 6: LOSS ESTIMATION [page 245] Loss Exceedence and Average Annual Losses Figure A6.1 depicts the Loss Exceedance Curve and the AAL for the earthquake hazard in the study region, and Figures A6.2-A6.4 show the corresponding AAL maps for Earthquake Hazard. LEC, Residential LEC, Commercial 0.003 0.003 € 0002 È 0002 ê 4 $ 0.002 $ 0.002 Es & Ë 0001 Ë 0001 è è $ 8 È 0.001 Ë 0.001 0.000 0.000 0 500 1,000 1,500 o 50 100 150 200 250 Millions Millions Loss (USD) Loss (USD) LEC, Industrial 0.003 È 0002 4 El © 0.002 & £ 0.001 H Ë 0.001 0.000 o 50 100 150 200 Millions Loss (USD) Average Annual Losses (AAL) USD$ x 1016 | %deExposure | USDS x 1076 % de Exposure USDS x 106 | Figure A6.1 Loss Exceedance Curve and the AAL for Earthquake Hazard = EMERGING #4 GIDB © À initiative ERM 3 ESCI HAITI — APPENDIX 6: LOSS ESTIMATION [page 246] PA en Legend Annual Average Losses (USD) Earthquake - Residential M 0-2:55 D 2:55 -6,989 MN 6089 -18,043 mure 4 I 5525-7745 een D PS Re Pre EE Figure A6.2 Risk Map: AAL for Earthquake Hazard, Residential M AE AE EE . Fe, nr, | uen see ve a Legend ‘Annual Average Losses (USD) Earthquake - Commercial M 0-55 I 555-1901 EN 1,901 -4,609 poses ln. 4 M 5: 254 sen Legs TT | Figure A6.3 Risk Map: AAL for Earthquake Hazard, Commercial ce 4% GIDB NS LE ERM 4 ESCI HAITI - APPENDIX 6: LOSS ESTIMATION [page 247] x ren . re. AREAS | un imeon ST 4 Legend Annual Average Losses (USD) Earthquake - Industrial M 2-50 IN 550-2274 DIN 2274-7,662 I 2225-1175 een eee 1 |] Figure A6.4 Risk Map: AAL for Earthquake Hazard, Industrial Losses to Critical Facilities and Infrastructure Table A6.2 depicts critical facilities and infrastructure losses for the earthquake hazard. Table A6.2 Losses to Critical Facilities and Infrastructure for Earthquake Hazard Number of Facility/Infrastructure Type Facilities, 475 Year 2500 Year Class Cri Fcites wi | « uommun] * | Hospitals (Level 1- Major Medical Facility) | 4 2,092,581 31.7% 3,390,100 51.4% Hospitals (Level 2 - Medical Facility/Clinic) 879,788 35.3% 1,339,951 53.7% Hospital ( Level 3 — Rural Clinic) 69,121 16.3% 112,487 26.5% College | 4 931,367 40.8% 1,322,628 57.9% Roads, km (Secondary) 8,683 65,710 4 Ss EMERGING” di Lo Es GIDB à: ERM 5 ESCI HAITI — APPENDIX 6: LOSS ESTIMATION [page 248] Electric Power Plant 141,774,845 35.4% 234,249,808 58.6% Electric Lines, km 96,399 356,803 13.7% Water Pumping Stations 85,578 175,047 143% Reseviors/Catchment 153,617 10.2% 356,672 23.8% Waste Water Treatment Plants 591,690 16.4% 1,024,518 28.5% A6.3 Hurricane Hazard The interplay of factors such as wind speed, wind direction, storm duration makes it quite difficult to predict the impact of a hurricane, nevertheless, the expected intensity of a hurricane in the area is significant with an expected 124 mph wind speed for three second gusts for a height of 10 meters for a flat terrain for 100 year return period and 156 mph for a 700 year return period. With this level of expected intensity, the expected damages to housing, institutional buildings and infrastructure is expected to extensive. The key variable for defining the vulnerability to the wind hazard in Northern Development Corridor of Haïti is the quality of construction, of which the majority of construction takes place without proper architectural and engineering oversight. Residential construction, in particular, is especially vulnerable with the roofs being the most susceptible component to damages. Residential housing in both rural and urban areas either had roof constructed of inadequate materials or had poor connections to walls making a majority of residential structures vulnerable to heavy wind gusts associated with hurricanes. Institutional buildings such as local government offices and industrial structures were found to be engineered a bit better and will most probably respond better, but may still experience extensive damages due to heavy wind loads or gusts associated with hurricanes. Estimated Probable Losses, Hurricane Hazard The estimates for the hurricane hazard are provided in Table 6.3 below, and represent the maximum probable losses for general occupancy classes. Losses are presented for four return periods. Table 6.3 Probable Maximum Losses (PML) for Hurricane Hazard Loss ( 106 USD) Return Period Years Residential [Commercial | Mindustrials| 2392 1700 2860 | 600 | 22 | 5 EMERGING ES GIDB Initiative ERM 6 ESCI HAITI — APPENDIX 6: LOSS ESTIMATION [page 249] Loss Exceedance Curve and the AAL Figure A6.5 depicts the Loss Exceedance Curve and the AAL for the hurricane hazard in the study region, and Figures A6.6-A6.8 show the corresponding AAL maps for Hurricane Hazard. Figure A6.5 Loss Exceedance Curve and the AAL for Hurricane Hazard LEC, Residential LEC, Commercial 0.025 0.025 4 0.020 È 0.020 3 3 8 0015 £ 0015 £ 0010 £ 0010 ë 0.005 ë 0.005 0.000 0.000 0 100 200 300 400 500 0 20 40 60 80 Millions Millions Loss (USD) Loss (USD) LEC, Industrial 0.025 # 0020 3 5 $ 0015 8 £ 0010 E ë 0.005 0.000 0 5 10 15 20 25 30 Millions Loss (USD) Average Annual Losses (AAL) USDS x 1016 % de Exposure USDS x 1016 % de Exposure USDS x 1016 % de Exposure = EMERGING = Lo es GIDB À initiative ERM 7 ESCI HAITI — APPENDIX 6: LOSS ESTIMATION [page 250] M PEER un Legend Annual Average Losses (USD) Hurricane - Res idential M 0-277 I 2737-7714 ON 7714-17,377 I 001-2254 en EE A A Pr EE Figure A6.6 Risk Map: AAL for Hurricane Hazard, Residential : RE an = Fe ea AN 4 raex Legend Annual Average Losses (USD) Hurricane - Commerical M 000000 - 37359917 M 37359918 - 1090.23348 DM 1090.23349- 2493293181 I 245353182- 7148093433 ln. 4 Figure A6.7 Risk Map: AAL for Hurricane Hazard, Commercial = EMERGING »« «#5 GIDB © L Initiative ERM 8 ESCI HAITI — APPENDIX 6: LOSS ESTIMATION [page 251] A EE A D PS En = AA ARE M son SIT EE Legend Annual Average Losses (USD) Hurricane - Industrial EM 0-10 DM :00- 200 IN 00-2252 IN 2252-7559 M 75-1522 soon Figure A6.8 Risk Map: AAL for Hurricane Hazard, Industrial Losses to Critical Facilities and Infrastructure Table A6.4 depicts critical facilities and infrastructure losses for the hurricane hazard. Table A6.4 Losses to Critical Facilities and Infrastructure for the Hurricane Hazard Facility/Infrastructure Type Number of 100 Year 700 Year 1700 Year Facilities/ Class CrtialFacties | | unions | % | uspane | % | unions | « | Hospitals (Level 1 - Major la 238,272 882,309 13.4% | 1,533,785 | 23.2% Hospitals (Level 2 — Medical 89,594 333,592 13.4% | 579,908 23.2% Hospital ( Level 3 — Rural Clinic) 15,495 58,414 13.8% 98,553 23.2% 3 EMERGING e<E GIDB L Initiative ERM 9 ESCI HAITI — APPENDIX 6: LOSS ESTIMATION [page 252] 55 (747m) 4,862,519 | 29.6% | 10,555,678 | 64.2% | 13,100,210 | 79.7% Electric Power Plant 118,632,228 | 29.7% | 256,769,64 | 64.2% | 318,666,06 | 79.7% Electric Lines, km 765,519 29.3% | 1,502,118 | 57.6% | 1,710,287 | 65.5% RS SE water times tm | à | | 00% | | 006 | | 0x | Water Pumping Stations 89,294 327,630 26.7% 569,544 46.5% Reseviors/Catchment 42,009 150,442 10.0% 261,525 17.4% nes 8 00 | | 006 | | Waste Water Treatment Plants 89,294 327,630 569,544 15.8% A6.4 Flood Hazard There are several factors that must be emphasized in regards to flooding in the Northern Development Corridor of Haiti. They relate to the precipitation data used to model the flood hazard, the surrounding environmental degradation that increases overland flows and the increasing population which has put pressure on the existing storm water drainage infrastructure. As outlined in the section above, the lack of reliable 24 hour precipitation data in flood hazard model predicated the need to rely on a limited 24-hour intensity data and monthly rainfall data to global models (i.e. Santa Clara University). This facilitated an approximation to understand flows in the hydrological model, but also introduces a high degree of uncertainty in the analysis of risk. The results show that the flooding hazard is quite significant in the study area and that with climate change the average flood depth will increase by about 0.23 m (23 cm) across all return periods. The expected flood depth for a 100 year flood event is expected to be 10.17m, which is significant given the flat terrain of the majority of the study area. Flooding must also be viewed in the context of environmental degradation occurring throughout Haïti which has resulted from deforestation. This has resulted in an increase in flood event frequency due to increasing volumes of overland flows. The combination of hillside degradation and urban development has reduced infiltration. The interplay of these factors has placed strains on existing drainage infrastructure, which are inadequate for population densities in urban areas and/or not present in rural areas. Urban drainage infrastructure is susceptible to blockage from an inflow of both earthen and man-made debris (garbage). Such debris inflows have resulted in blockages and/or increased and prolonged flooding in urban areas. In rural areas, the lack of drainage infrastructure or riverine stabilizers has resulted in erosion of river banks during high magnitude rainfall events and furthers siltation, blockage or overflow where infrastructure is in place. Therefore, it is likely given the interplay of these factors that flooding will become a bigger problem unless storm water drainage infrastructure is improved to increase infiltration and alleviate flash flooding problems. 3 EMERGING + Lo HS GIDB Initiative ERM 10 ESCI HAITI — APPENDIX 6: LOSS ESTIMATION [page 253] Estimated Probable Losses, Flood Hazard The estimates for the flood hazard are provided in Table 6.5 below, and represent the maximum probable losses for general occupancy classes. Losses are presented for six return periods. Table A6.5 Probable Maximum Losses (PML) for the Flood Hazard Loss ( 1016 USD) 5 | 47 | 08 | 04 | Lao | 55 | 09 | os | Las | 67 | 11 | 06 | so | 72 | 12 | os | [ao | 80 | 13 | 10 | Loss Exceedance Curve and the AAL Figure A6.9 depicts the Loss Exceedance Curve and the AAL for the flood hazard in the study region, and Figures A6.10-A6.12 show the corresponding AAL maps for the Flood Hazard. LEC, Residential LEC, Commercial 0.600 0.600 È 0.500 È 0.500 Ë 0400 Ë 0400 £ £ % 0300 % 0300 ÈË ÈË Ÿ 0200 Ÿ 0200 ë 0.100 ä 0.100 0.000 0.000 0 2 4 6 8 10 00 02 04 06 08 10 12 14 Millions Millions Loss (USD) Loss (USD) LEC, Industrial 0.600 Æ 0.500 É £ 0400 2 $ 0300 $ 0200 El ë 0.100 0.000 00 02 04 06 08 10 12 Millions Loss (USD) Average Annual Losses (AAL) [ Residential | Commeral Jindwmidl USDS x 1016 % de Exposure USDS x 1016 % de Exposure USDS x 1016 % de Exposure 0.092% 0.095% 0.058% = EMERGING = Lo #5 GIDB LL ERM 11 ESCI HAITI — APPENDIX 6: LOSS ESTIMATION [page 254] Figure A6.9 Loss Exceedance Curve and the AAL for the Flood Hazard M PE AE EE - re. A AN | een seen EL a Legend Annual Average Losses (USD) Hlood- Residential M 0-::2 M 5529 23978 EM 2,978 -66,182 M 11954-24555 - omeon M PS EE Figure A6.10 Risk Map: AAL for Flooding Hazard, Residential : A .. F =: Fe ea AN | ce k L Legend Annual Average Losses (USD) Flood-Commercial CT EN 50-2574 DIN 2575-8660 moe ln. 4 M 14550-55677 een A PP EE Æ EMERGING + #5 GIDB Initiative ERM 12 ESCI HAITI — APPENDIX 6: LOSS ESTIMATION [page 255] Figure A6.11 Risk Map: AAL for Flooding Hazard, Commercial ram él Legend Annual Average Losses (USD) Flood -industrial D 0-72 IN 72-2371 DIN 2371-5469 I 540-1115 RE 15-5421 con 217] 7 country Boundary Figure A6.12 Risk Map: AAL for Flooding Hazard, Industrial Losses to Critical Facilities and Infrastructure Table A6.6 depicts critical facilities and infrastructure losses for the flood hazard. Table A6.6 Losses to Critical Facilities and Infrastructure for the Flood Hazard Facility/Infrastructure Type Number of 25 Year 50 Year 100 Year Facilities/ Class Critical Faces [uso | x l'usouons | x | usaors ximdergamen | 0 | | 00 | | 00 | | om cotege À | om | | 00 | | om ones 0 À | où | | 00 | | 0% 3 EMERGING e<E GIDB L Initiative ERM 13 ESCI HAITI — APPENDIX 6: LOSS ESTIMATION [page 256] Roads, km (Highway) 1,153 0.029% 1,224 0.030% 1,307 0.032% Roads, km (Secondary) 0.11% 0.11% 1,063 0.13% Electric Lines, km 0.0027% 00030% | # | o0034% water master] no | | œuvens |] | 006 | | 006 | | 0 Waste Water Treatment Plants 236,033 298,540 349,981 A6.5 Coastal Flood Hazard The Northern Development Corridor is most likely to be affected by a combined effect of storm surges and wave action associated with hurricanes and strong winds as indicated by the Total Arbiter of Storms (TAOS) hazard modeling system. The TAOS results combined with climate change suggests that the expected estimated water surface elevations (i.e. maximum depth of flooding) for a 50 year return period is expected to be 5.35 m, while the projected estimated water surface elevations for a 100 year return period is expected to be 6.15 m. These results clearly show that there will be significant impact to coastal developments and that the flooding could occur up to +1000 m inland impacting - population, agriculture, industry and infrastructure. Coastal flooding will cause disruption of settlements and may inhibit the movement of persons. In coastal settlements such as Limonade du Mer and Caracol the situation will become dire, especially as drainage systems, which are already under stress, are overwhelmed by a combination of water from coastal surges and waves and accumulation of inland flows associated with storms. Therefore, it is likely that coastal flooding will become a bigger problem if settlements in high hazards areas continue to grow and drainage systems are not improved. Estimated Probable Losses, Coastal Flood Hazard The estimates for the flood hazard are provided in Table A6.7 below, and represent the maximum probable losses for general occupancy classes. Losses are presented for four return periods. Table A6.7 Probable Maximum Losses (PML) for the Coastal Flood Hazard Loss ( 106 USD) 3 EMERGING + Lo HS GIDB Initiative ERM 14 ESCI HAITI — APPENDIX 6: LOSS ESTIMATION [page 257] 10 34.6 5.1 1.4 Loss Exceedance Curve and the AAL Figure A6.13 depicts the Loss Exceedance Curve and the AAL for the Coastal Flood Hazard in the study region, and Figures A6.14-A6.16 show the corresponding AAL maps for the Coastal Flood Hazard. LEC, Residential LEC, Commercial 0.120 0.120 E 0.100 £ 0.100 £ 0.080 £ 0.080 ë 0.060 $ 0.060 Ë Ë Ÿ 0.040 Ÿ 0.040 è ë Ë 0.020 ä 0.020 0.000 0.000 o 20 40 60 80 0 2 4 6 8 10 12 Millions Millions Loss (USD) Loss (USD) LEC, Industrial 0.120 E 0.100 £ 0.080 ë 0.060 Ë Ÿ 0.040 è Ë 0.020 0.000 o 1 2 3 4 5 6 Millions Loss (USD) Average Annual Losses (AAL) [Residential | Commerl [induml USD$ x 1016 % de Exposure USDS x 1016 % de Exposure USD$ x 1016 D | 0.209% [0.6 | 010% 0.075% Figure A6.13 Loss Exceedance Curve and the AAL for Coastal Flood Hazard = EMERGING = Lo «HE GIDB Initiative ERM 15 ESCI HAITI — APPENDIX 6: LOSS ESTIMATION [page 258] RE PE AE EE PS - Fe RaAN seen seen SIT A Legend Annual Average Losses (USD) Coastal Flooding - Residential 0-::22 AN :3,212-39252 EM 2,252-98,131 M 557-5512 . ose Leger 7 Figure A6.14 Risk Map: AAL for Coastal Flooding Hazard, Residential RE PE EE …. - pe TN rem Legend Annual Average Losses (USD) Coastal Flooding - Commercial M 0-25 IN 2155 -7,589 DIN 7,589 -16,748 — fées ln. 4 I 222 -62.560 een RS PE Figure A6.15 Risk Map: AAL for Coastal Flooding Hazard, Commercial Æ EMERGING + #5 GIDB Initiative ERM 16 ESCI HAITI — APPENDIX 6: LOSS ESTIMATION [page 259] PE PE ES <a | > nl soon: … Fe 4 à Legend ‘Annual Average Losses (USD) Coastal Flooding - Industrial D 0-2: ER 23-150 DIN :,509-3,299 Figure A6.16 Risk Map: AAL for Coastal Flooding Hazard, Industrial Losses to Critical Facilities and Infrastructure Table A6.8 depicts critical facilities and infrastructure losses for the Coastal Flood Hazard. Table A6.8 Losses to Critical Facilities and Infrastructure for the Coastal Flood Hazard Facy/ntastruene nr Facilities/Class Critical Faites | Jus x | Hospitals (Level 1 - Major Medical Facility) la | low | Hospitals (Level 2 - Medical Facility/Clinic) 116,000 Hospital ( Level 3 — Rural Clinic) 2,931 RS PS Road kr (High) Us | om Road km Secondary) Du 1 om 2 %s EMERGING gi “4 GIDB Initiative ERM 17 ESCI HAITI — APPENDIX 6: LOSS ESTIMATION [page 260] cuves | Jo A6.6 Drought Hazard Drought has been estimated using a hydrological model to obtain water balance of two major watersheds, and has sought to delineate the watershed and obtain the drainage and precipitation characteristics, including evaporation and outflows so as to determine water balance in the study area, which is primarily focused on understanding current and future water utilization demands. An impact assessment of the drought hazard was not pursued as part of the study. The assessment of drought is concerned rather with the capacity of the current water system to store, accumulate and release ample water for current and future demands, incorporating the impact of climate change. All things considered, water shortage by hydrological droughts combined with growing water demand could lead to increasing water stock reduction and negatively impact agriculture, including cultivation of land and grazing. Food production will be one strain on water as will growth of urban settlements. Climate change plays into the equation as land use changes due to urbanization will increase evaporation and surface and subsurface water discharge, thus further decreasing water supply. It is also necessary to note that changes in water availability, both increases as well as decreases, can affect water quality. Increased water availability, especially through heavy rainfall, can lead to increased suspended matter in water bodies, as well as a washing out of contaminants (e.g. fertilizer, toxins) into water and groundwater. == EMERGING = CR SUSTAINABLE “IDB Lo LL ERM 18 ESCI HAITI — APPENDIX 6: LOSS ESTIMATION [page 261] APPENDIX 7: Restrictions Maps EMERGING — e<## GIDB | NORTHERN DEVELOPMENT CORRIDOR, HAITI Appendix DA ERM [page 262] Figure A7.1 - Weighted Impact of Natural Hazards Risk: Flooding and Seismic Sn = N ; fi 7 El x j Ë 4 2 } SR, D LS fl ñ ( E Q + ” + { f # 4 = L CL LS \ à À # s g > A } \ { { F Î FE. W A, \ { Vu Pré “ “ 4 \: ÿ port À » .. = Fi É à J CR à se % f À K A Fe %, » &, à ne 0 f 25 $ 10 À “ a H Ce com LL) :..3 Limite commune Recif corallien Less restricted Completely restricted MN Océan Atlantique Map source: From geographic information layers by AIA (2012), CIAT (2012), CNGIS (c.2012), IADB (c.2013), NATHAT (2010), USAID-OFDA (c.2012), OSM (2013), OCHA (c.2010), PDNA (2010) and DTM (2013). Georeferencing, digitizing and remote sensing analysis from satellite imagery 1986, 2010 and 2013. [page 263] Figure A7.2 Restriction factors: Topography and slopes See FA . LE LL? \ = n < | 1 F4 | 62 si Le, | = F j \ { À fs Q : ra 7 È : . A - \ EL . + s ri _d : ES Li « ; RER, CL ' FR + (R F u ”. ( | Re é à 1 "a c . . # . , y’ Li à à | ia s r [4 ‘ " * À ei LS L 5 CES. L z : (! & Li.‘ pr) à &" té --. \ CRPLÉNETT à a À ML A6, Auf À D. — A CD Et dy arms Late corerrste LL LL | |} ect coaber vent cost CNET sn Cetan htaetque … [page 264] Figure A7.3 Restriction factors: Hydric System -— Superficial Water 4 7 en D. ; LR VE LAN UR CS NT mnt) 74 Re ss [page 265] Figure A7.4 Restriction factors: Strategic Ecosystems and Protected Areas LE | ER 7 ef 1 | ni U LE AN Se A Pi AT JUN M FPT Ni dr D D susysre (TT Limité Ecrmemane LL LL | | [page 266] Figure 1 Restriction factors: Strategic Ecosystems and Protected Areas ne mr. 4 : f à > SE | l L: | Li 1 ; 1 À n : (A S, = bb E L ; * d | - œ S \es € e | è { = ” * . L L | re . L 2 ' } : 1 4 { 0 - | { . . \ | CE # À © sus are Fecil coton Less sosttizéd Congdotnl g route heal Mn Cher Atterrique [page 267] Figure A7.6 Restriction factors: Adequate Land Use D 2, ; _ Ne 17 4 “ ns ! Vis L ô Dé | Î L L 2e (bd “/ 7 » Li ra =. ù ) À m— FA Eure LL LL Fecil coraien Less sesttnéé Congéstnl g retire bent Mn Cohen Allwrique [page 268] Figure A7.7 Restriction factors: Mining sr _— mn % =, NT. f sd | ER À — } | } fr | EL 2 n.” À : 1 Î w \ \ À PP | T2 | Lo | “ j, l # ‘ » \ { . se p) " À © sut are Fecil cocaten Cunghotnl y veuiPetet au Ccéss Aterique Lens ltd [page 269] APPENDIX 8: Attractions Maps EMERGING — e<## GIDB | NORTHERN DEVELOPMENT CORRIDOR, HAITI Appendix dti ERM [page 270] Figure A8.1 Attraction factors: Agglomeration P PE ——, Ÿ ee 72. Ar? 18 L 6 | Ï > | à Q / j "ATX i x 2 Sr" fe". 1e . ! 4 LR | ef 1 er \ A “ % | / y de À à À à x | | si = » ” À Lrréte corremine _ Recil comen set misinre (COTES 222 01 Octar Alterèque [page 271] Figure A8.2 Attraction factors: Road System ai e — € > 5 FLE V Dares ans Lrréié ccrrenae _ Fecil comen Must mise Lots dhèe me un! Ccéss Alterique [page 272] Figure A8.3 Attraction factors: Public Utilities —- Water supply PE _ At ES 7 +. ® e. a - ; | : } l L ; ? f } : 2 { | FA ( k k ? } : + | ; | (1 \ { / x N | a" — [ \ | (.— a L : Lil * | L J 4 | fl | . \ Ÿ V } / 27 5 À Dares ns Lrréié corremae _ Fecil cocon Must miss Lente dits Vue MN Cotes Altrique [page 273] Figure A8.4 Attraction factors: Public Utilities — Electric System ms = _. ce LL? 1 y | | = - 1 / » — { - L | #” | , ] L « e ( { { \ | : » . 0 * \ À { “À Lrréié corremane _ Fecilccaen net mrsunre Lens abs De Mn Chan Alterrique [page 274] Figure A8.5 Attraction factors: Social Services sr as. Ÿ pe, | _— PE + CR ! ; ' : É | | Î ' ° | L É? ‘ | : } 4 Le à : : 1 - / } F js LEA [ ; ! . > . % LR L { L \ + oi [ \ | (. | L À | #” | * > . 1 f | ) A \ | . \ | “! = ‘ ï À pus ns a du 0 | bus Minute Lots dits ee MN Cohen Alierique [page 275] APPENDIX 9: Development Project Maps EMERGING = #4 GIDB © NORTHERN DEVELOPMENT CORRIDOR, HAITI Appendix haies ERM [page 276] Figure A9.1 Attraction factors: Urban Agglomeration ' AT... + ENS j . & Lors L NS « Le [ J à. PR \ S = i, H La S S j Î } À] / * ! f [ N° &, NS \ } 0! 25 5 10 !11 Limite commune Recif corallien Most attractive Less attractive bn Océan Atlantique [page 277] Figure A9.2 Attraction factors: Sea Port Expansion Vn Î ss us 4 LES \ bai j \ 2 4 à f \ ! m ui 1 2 # ! S } { Led se \ Ni \ . ? LA 4 0 25 5 10 À À ! Km Ras RE {171 Limite commune Recif corallien Most attractive Less attractive bn Océan Atlantique [page 278] Figure A9.3 Attraction factors: Infrastructure ' AT... Ï ENS F ra H | Tres \ © pi ; a j sé { $ \ F o } . LT \ { | 5 \ À. L | S | # S 7 / i à N 6: 4 S ( Là 1 & \ + € LR \ \ =. FX, Œ t d. L Î Ÿ |] } À à ÿ À 4 D Î “ l } | PA Le ! f [ N° e | \ x } 0 25 5 10 ©7713 Limite commune Recif corallien Most attractive Less attractive bn Océan Atlantique [page 279] Figure A9.4 Attraction factors: Employment | e ‘e É = ÿ 2 _ = _ Le] [=] Lu 2 [e] | à @ Ë F4 C2 @ x 0 25 5 10 C1 Study area Frontiere In Océan Atlantique Most attractive Less attractive [page 280] APPENDIX 10: Risk Reduction Assessment EMERGING — “#5 GIDB © NORTHERN DEVELOPMENT CORRIDOR, HAITI Appendix tie ERM [page 281] A10 RISK REDUCTION ASSESSMENT A10.1 Approach Specific recommendations and projects that are assessed in this section have been scrutinized using a Cost Benefit Analysis (CBA) model to assess the likely costs and benefits of identified mitigation measures. It is not surprising, given the intensity and frequency of flooding (both inland and coastal) that these hazards have been given priority for risk reduction assessment. This section of the report highlights the methodology followed for a risk reduction assessment and introduces and conducts a scenario analysis for various mitigation strategies in high risk areas so as provide an estimate of the costs and benefits. The risk reduction assessment comprises of all activities, including structural and non- structural measures, to avoid (prevent) or to limit (mitigate) adverse effects of hazards (ISDR 2004). Outlining the benefits of risk reduction in terms of reduced damages and other associated benefits can help in decision making for allocation of limited resources for investments in risk reduction. A CBA is used to assess the likely costs and benefits of risk reduction measures. The terms “mitigation alternatives” and “what if analysis” have been used interchangeably to indicate options that can be to be taken forward for more detailed engineering analysis and specific implementation. Table A10.1 provides a listing of the scenarios, identified by the project team, and analyzed in this study. Table A10.1 List of Mitigation Scenarios | projet | Mitigation Scenario Upgrade Urban Drainage Infrastructure Rural Drainage Infrastructure Implementation Revitalize Historical Canal System To Alleviate Flooding Upland Reforestation Mangrove Protection In The 3Bay Park The risk assessment that has been carried out for the area, provides the basis to assess the benefits and costs of possible mitigation measures. The risk assessment considers hazard, exposure, and vulnerability for buildings, infrastructure, and society. The overall approach followed for the CBA is shown in Figure A10.1 and considers the various benefits expected from strategies and the corresponding costs involved. The risk models developed for this study have been used to analyze the risk (losses) if a particular mitigation option is implemented. Next, the risk metrics [average annual loss (AAL), loss exceedence curve and loss cost] are used to identify and prioritize areas or localities that are under risk. The benefits of mitigation are then estimated by taking the difference between direct and indirect losses with and without mitigation. The benefits have been estimated as present value of future (recurring) benefits considering the life of the building/infrastructure that is being proposed. The costs of mitigation have also been estimated. These include the cost of structural interventions, setting up systems, recurring costs such as maintenance, etc. “+ #%% GIDB sn ÉLL 4 ERM 1 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 282] Risk Analysis: Potential Risk Analysis: Potential Li Reduced Hazarc Benefits of Mitigation | Costs of Mitigation Figure A10.1 Cost Benefit Analysis A10.2 Mitigation Option No. 1.: Upgrade Urban Drainage Infrastructure in Limonade Limonade has been selected to demonstrate how improvement to urban infrastructure can be used to effectively mitigate damages and reduce losses. The example has been identified to address the problem of overland flow when the amount of rainfall exceeds the capacity of existing infrastructure. The increasing urbanization and inadequate infrastructure results in urban flooding Often rainfall water exceeds the capacity of storm water drainage infrastructure or the system becomes blocked due to debris. An increase in artificial drainage infrastructure (open catchpit drainage and culverts) is proposed to address alleviate the flooding so as to discharge water into the receiving watercourses and eventually the ocean. The proposed project shall benefit seven blocks in the Limonade (Figure A10.2). The estimated length of open drains is about 18 km, of which 9 km is assumed to be in place (existing), and an additional 9 km is proposed as new construction. Along with the 9 km of new proposed construction drains, it is anticipated that approximately 75 culverts may be needed at the various road crossings to facilitate drainage. ; RGING = Le es GIDB : ERM 2 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 283] à b . Un SEL... seen Dést JJ … {1 HE l té de Tropicaf M | L_] D LE LT LT LT A NUIT uen — n— le EP d en v Legend se EL, ° Fes Rosds D / DU Food Extent 100 Year Return Period ü8,, 200 400fMetefs {sex vont L “\ | [| Beneficisry Aress in Limonsde Tin ra7en Figure A10.2 Potential Beneficiary Area in Limonade Approach The urban drainage system comprising of open ditches and culverts is proposed as mitigation scenario for inland flood protection. This system will be helpful in reducing the flood inundation by providing a passage to the flood waters. With reduced flood inundation, it is expected the risk of flooding in the area of interest (Limonade) will also be reduced. Probabilistic risk assessment results derived earlier in this study have used to estimate the reduction in risk and overall benefits (losses avoided). The base case risk metrics (AAL, LEC and Loss Cost) have been used to identify blocks within the Limonade which are likely to be benefited from the drainage infrastructure improvements. As shown in the above map, much of the urban area in Limonade benefits from this mitigation action. Since the town is vulnerable to both inland and urban floods, it is assumed that this system will bring down the overall risk in this area by 50%. Finally, the benefits to the area of interest are calculated by comparing the base case (AAL) and the AAL with the mitigation options in place. Cost benefit ratio measures the costs incurred and the benefits accrued from a policy or action - in this case the construction of urban drainage system. Costs: The costs of drainage system construction and upgrades are considered. The system comprises of construction of new open ditches or catch-pits, upgrading of existing ditches and construction of new culverts. Costing of construction includes all the components such as costs of materials, equipment, labor forces, support, benefits, and other miscellaneous costs. The project costs include: ° New Open catch-pit of 0.5x0.5 mis estimated at about USD $180,000 per km. … EMERGING Lo “<#"# SIDB PK \ Dove ERM 3 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 284] + Upgrade of the existing drainage system is assumed to be 70% of new construction, which amounts to USD $126, 000. + Cost of construction of the culvert is estimated to be USD 3,600 for 6 m length of the culvert. As mentioned above, in the seven beneficiary urban blocks of Limonade, the estimated length of open drains is about 18 km. About 9 km are assumed as existing ones and rest 9 km is proposed for new construction. Along the 9 km of new proposed construction drains, an about 75 culverts are estimated. Thus the total cost of proposed drainage system is estimated to be USD 3,024,000. The costs are assumed one- time investments with expected benefits spread over the life of the system. The life of the system has been taken as 20 years. Benefits: Different benefits due to mitigation are considered for the cost benefit analysis that include indirect and direct benefits (both tangible and intangible). Benefits of mitigation are estimated by taking the difference between losses with and without mitigation (Kunreuther, 2001). A portion of the direct tangible benefits of a flood damage reduction project is due to a reduction in the inundation area which will reduce damage due to flooding of structures and other properties. Mathematically, we compute this inundation-reduction benefit Bk as: Bir = Xwithout — Xwith in which: Xwithout — expected damages or economic impact without project; and Xwith - expected damage or economic impact if the project is implemented. The benefits include reduced losses to residential and commercial buildings due to the improvement of the drainage system. The benefits due to reduction in losses for various sectors (as shown in above equation) are estimated as the difference between present values of future flood AAL with the project in place and AAL without improvements to the drainage system. Since these benefits accrue over the life of the project (buildings/infrastructure/systems), it is important to discount them to a present value so that benefits accruing at different times can be made comparable. The benefits are estimated as present value of future (recurring) benefits considering the life of the system. The life of system is usually considered as the minimum time period in which system will be functional. In case of drainage systems (catch-pits, culverts, etc), it is taken to be 20 years. The present values of future benefits are estimated as: PV=C G@+d)-1 = Cox ——————— 9 dx(1+d)t Where, PV = Present value Co = Cost (In this case average annual loss) d = discount rates (assumed 3%) t= time, years (assumed = life of the system 20 years) The outcome of the above equation is the present value of future benefits over the time (considered as the life of the system). The total present values of benefits with the proposed mitigation action is USD $ 7,694,826. Cost Benefit Analysis As described above, various benefits and costs are considered in the cost benefit analysis to estimate the cost benefit ratio. The proposed urban drainage system is expected to benefit most of the urban area in Limonade and reduce flood losses to various exposure sectors. +. EMERGING 42 SSTAINABLE PR) Ses ERM 4 ESCI HAITI - APPENDIX 10: COST BENEFIT ANALYSIS [page 285] TableA10.2 evaluates the mitigation option of introducing urban drainage system in comparison to base case. The base case is status quo condition generally assumed to be the losses without a mitigation action. Table A10.2: Cost benefit analysis of urban drainage system in Limonade Other Information Life of the Drainage System, Years L Gromthraator | 19 | ms | Flood AAL Residential Building AAL, USD 838,692 419,346 [| Commercial Building AAL, USD 194,171 97,086 || industrial Buiäing aAL uso S Present value of future Flood losses Residential Building AAL, USD 12,496,504 6,248,252 [| Commercial Building AAL, USD 2,893,148 1,446,574 || industrial Buiäing AL usD D Cost of Drainage System Construction, USD l 3,024,000 |__| rotalcost,usn |__| 274000 Present Value of Benefits| Residential Building, USD l 6,248,252 |__| commercial Buiäing, USD | 1426578 Cost benefit Ratio Cost benefit Ratio | | 24 Base Case: The results from base case analysis of the probabilistic risk assessment have been used to establish this scenario. The results from the defined beneficiary urban area suggest that AAL for buildings due to floods is USD $1,032,863. The present values of future flood AAL to the buildings in the base case condition are estimated over a 20 year period to be USD $15,389,652. Cost benefit Ratio of Mitigation Option: Since the total cost of construction of drainage system is estimated to USD $3,024,000. The total present values of benefits with the proposed drainage options are estimated to be USD $7,694,826. The cost benefit ratio (BCR) is computed by taking a ratio of present value of all benefits due to mitigation and total costs of mitigation. BCR Ÿ Present Value of Benefits : Y Costs of Mitigation Considering the total benefits of USD $7,694,826 against the total costs of the mitigation action (USD 3,024,000), the BCR for this mitigation scenario is 2.54. A BCR, greater than one, suggests that benefits are higher than the cost of the project, indicating that the project is a good investment (i.e. the benefits outweigh the costs). Considering the satisfactory cost benefit ratio, this option may be taken forward for possible planning, pre- feasibility studies, etc. #8 GIDB nn? \ ERM 5 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 286] A10.3 Mitigation Option No 2. Rural Drainage Infrastructure Implementation A proto-typical rural infrastructure solution has been identified to address the problem of overland flows in rural areas. Where in urban areas problems arise due to the fact that drainage infrastructure is overloaded or outdated, in rural areas infrastructure to divert water flows is largely unavailable. To address flooding, especially along transportation corridors and settlement areas, a system of linear concrete drainage outlets such as culvert pipes and rolling cross drains (catchpits) are proposed to divert water away from settlements. Drain canals are to be reinforced with rock and masonry to prevent erosion. It is important that consideration to the impact the flow of water will have on settlement areas in the lower reaches of the watershed so as not to exacerbate flooding in other areas. As shown in Figure A10.3, the diversion drain (1.0 X 1.5 m) starting from area south of industrial park and draining in to the sea (west of Caracol town) may comprise a length of about 7.1 km and 5 culverts. It is assumed that at present there is no infrastructure available and proposed infrastructure will be newly constructed. FE ——. _ AT un 3 & fl AT | 4 f ’ | acquezil 1 Mab F — D 4 À Ï nina 7. sen ? $ Legend | Garde MimaigChabert — Rural Drains Petit-Co ne a IN Flood Extent 100 Vear Return Period) 365 | 720 1,460 LE [184 our | Lee Figure A10.3 Potential Beneficiary Area for Rural Infrastructure Approach To alleviate flooding, especially along transportation corridors and settlement areas, a system of linear concrete drainage outlets such as culvert pipes and rolling cross drains (catch pits) are proposed to divert water away from settlements. Drains canals are to be reinforced with rock and masonry to prevent erosion. This system will be helpful in reducing the flood inundation by providing a diversion to the flood waters. With reduced flooding, it is expected the risk of flooding in the area of interest (linear settlements occurring near and around the industrial park) will also be reduced. Probabilistic risk assessment results derived earlier in this study have used to estimate the reduction in risk and overall benefits (losses avoided). The base case risk metrics (AAL, LEC and Loss Cost) have been used to identify and blocks within the area of interest which are likely to be benefited. As shown in the above map, approximately 39 blocks will benefit from this mitigation action. Since the area is vulnerable to inland floods and proposed system dimensions shall carry more floodwaters, it is assumed that this system will bring down the overall risk in this area by 70%. Finally, the benefits to the area of Lg" HG Lo PK \ EL ERM 6 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 287] interest are calculated by comparing the base case (AAL) and the AAL with the mitigation options in place. Cost benefit ratio measures the costs incurred and the benefits accrued from a policy or action - in this case the construction of rural drainage system. Costs: The costs primarily include the costs of drainage system construction. The system comprises of construction of new open ditches or catch-pits and new culverts. Generally, the costs of drainage system construction and upgrades include all the components such as the costs of materials, equipment, labor forces, support, benefits, and other miscellaneous costs. The project costs include: ° New open catch-pit of 1.0X1.5 mis estimated at about USD $135,000 per km. + Cost of construction of the culvert is taken USD 3,600 for 6 m length of the culvert. As mentioned above, there are 39 beneficiary blocks and the estimated length of the proposed drains is about 7.1 km. Along the 7.1 km of new proposed construction drains, a about 5 culverts are estimated. Thus the total cost of proposed drainage system works out at about USD 980,550. The costs are assumed one-time investments with expected benefits spread over the life of the system. The life of the system has been taken as 20 years. Benefits: Different benefits due to mitigation are considered for the cost benefit analyses that include indirect and direct benefits (both tangible and intangible). These studies are generally taken up later in the implementation and engineering design stage of the project. Benefits of mitigation are estimated by taking the difference between losses with and without mitigation (Kunreuther, 2001). A portion of the direct tangible benefits of a flood damage reduction project is due to a reduction in the inundation area which will reduce damage due to flooding of structures and other properties. Mathematically, we compute this inundation- reduction benefit Bk as: Bir = Xwithout — Xwith in which Xwithout — expected damages or economic impact without project; and Xwith - expected damage or economic impact if the project is implemented. The benefits include reduced losses to residential and commercial buildings due to the improvement of the drainage system. The benefits due to reduction in losses for various sectors (as shown in above equation) are estimated as the difference between present values of future flood AAL with the project in place and AAL without improvements to the drainage system. Since these benefits accrue over the life of the project (buildings/infrastructure/systems), it is important to discount them to a present value so that benefits accruing at different times can be made comparable. The benefits are estimated as present value of future (recurring) benefits considering the life of the system. The life of system is usually considered as the minimum time period in which system will be functional. In case of drainage systems (catch-pits, culverts, etc), it is taken as 20 years. The present values of future benefits are estimated as: PV=C G@+d)-1 = CX ——— % dx(1+d) Where, PV = Present value Co = Cost (In this case average annual loss) d = discount rates (assumed 3%) t= time, years (assumed = life of the system 20 years) The outcome of the above equation is the present value of future benefits over the time (considered as the life of the system). 4x. EMERGING ) de = ERM 7 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 288] The outcome of the above equation is the present value of future benefits over the time (considered as the life of the system). The total present values of benefits with the proposed mitigation action is USD $1,049,532. Cost benefit Analysis As described above, various benefits and costs are considered in the cost benefit analysis to estimate the cost benefit ratio. The proposed rural drainage system is expected to benefit most of the area near industrial park and reduce flood losses to various exposure sectors. TableA10.3 evaluates the mitigation option of introducing rural drainage system in comparison to base case. The base case is status quo condition generally assumed to be the losses without a mitigation action. Table A10.3: Cost benefit analysis of rural drainage system Other Information Life of the Drainage System, Years U érowhra | us | us | Flood AAL Residential Building AAL, USD 7,238 2,171 |__| commercial Building AL, USD 1,402 | Industrial Building AAL, USD 91,986 27,596 Present value of future 107,848 32,355 Flood losses Residential Building AAL, USD [| Commercial Building AAL, USD 20,885 6,265 D | Industrial Building AAL, USD 1,370,599 411,180 Cost of Drainage System Construction, USD l 980,550 | | Total cost, USD | | 980,550 Present Value of 75,494 Residential Building, USD | mm |__| commercial Building, USD [| m6 [| industrial Building, USD [| ose Cost benefit Ratio Cost benefit Ratio | | 1o | Base Case: The results from base case analysis of the probabilistic risk assessment have been used to establish this scenario. The results from the defined beneficiary urban area suggest that AAL for buildings due to floods is USD $ 100,626. The present values of future flood AAL to the buildings in the base case condition are estimated over a 20 year period to be USD $1,499,332. Cost benefit Ratio of Mitigation Option: Since the total cost of construction of drainage system is estimated to USD $ 980,550 and the total present values of benefits with the proposed drainage options are estimated to be USD $ 1,049,532. The cost benefit ratio (BCR) is computed by taking a ratio of present value of all benefits due to mitigation and total costs of mitigation. Y Present Value of Benefits BR = — Y Costs of Mitigation The comparison of the total benefits of USD $1,049,532 against the total costs of mitigation action of USD $980,550, the BCR is 1.07. The cost benefit ratio, greater than one, suggests that benefits are higher than the cost of the project, indicating that the project is a conservative investment (i.e. the benefits slightly outweigh + Fu SIDB Ke À \ LL ERM 8 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 289] the costs). Considering the satisfactory cost benefit ratio, this option may be taken forward for possible planning, pre feasibility studies, etc. A10.4 Mitigation Option No. 3: Revitalize Historical Canal System to Alleviate Flooding The revitalization of the canals in northern Haiti is focused on a revitalizing existing drainage and irrigation systems that were put in place during the plantation era. In historical times, these canals helped serve as a regulatory mechanism for the conservation of fresh runoff water during times when there was a significant lag time between significant rainfalls. Today, the canal system lies in decay (silted) on defunct plantations, and this mitigation scenario, explores whether revitalizing certain areas can help play a role in mitigating flood problems by providing critical drainage in the lower reaches of the watersheds. Specifically, the revitalization of canals will include dredging to remove silt and increasing the cross-sectional area of the canals so as to divert flooding and overland flows to natural watercourses and/or ocean. Key considerations for this project will include providing an adequate site for discarded material and ensuring that natural barriers are in place to manage sediment and nutrients loads in coastal environs. Another key consideration is to ensure that there are mechanisms in place to ensure that there is maintenance or upkeep. As shown in Figure A10.4, the possibility of dredging a series of canals that total a length of 3.8 km is evaluated. Considering the dimensions 4m X 6m for a typical canal system, the dredging volume is estimated at 40,800 m°. Fragen rrsgeue x “& Hs rue LE | QC | 7 \ Ÿ =. Legend \) + Flac =— Canss | DU Food Extent 100 Year Return Period Éo—24 20 * L [1 Bloc Boundery _ [__] Beneficiary Aress ren rrsfon Figure A10.4 Potential Beneficiary Area for Canal System Approach Dredging provides a mitigating effect for flooding by increasing the cross-sectional area of the canals so as to divert flows to natural watercourse and/or ocean. A possibility of rehabilitation of canal system in the area of interest has been evaluated in this section. With reduced flooding, it is expected the risk of flooding in the area of interest (near Jacquezil) will also be reduced. Probabilistic risk assessment results derived earlier in this 3x. EMERGING nŸ \ Dove ERM 9 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 290] study have used to estimate the reduction in risk and overall benefits (losses avoided). The base case risk metrics (AAL, LEC and Loss Cost) have been used to identify and blocks within the area of interest which are likely to benefit. As shown in the above map, about four blocks in the area of interest will benefit from this mitigation action. Since the area is vulnerable to inland floods and proposed system dimensions shall carry more floodwaters, it is assumed that this system will bring down the overall risk in this area by 70%. Finally, the benefits to the area of interest are calculated by comparing the base case (AAL) and the AAL with the mitigation options in place. Cost benefit ratio measures the costs incurred and the benefits accrued from a policy or action - in this case the construction of rural drainage system. Costs: The costs include the costs of dredging. The project cost of dredging of canals of 4 X 6 m for the length of 3.8 km (40, 800 m°) is estimated at about USD $ 183,600 considering a unit cost of USD 4.5/m°. The costs are assumed one-time investments with expected benefits spread over the life of the system. The life of the system has been taken as 20 years. Benefits: Different benefits due to mitigation are considered for the cost benefit analysis. For this project, a portion of the direct tangible benefits of a flood damage reduction is expected as inundation decreases. Mathematically, we compute this inundation-reduction benefit B, as: Bir = Xwithout — Xwith in which Xwithout — expected damages or economic impact without project; and Xwith - expected damage or economic impact if the project is implemented. The benefits include reduced losses to residential and commercial buildings due to the improvement of the canal drainage system. The benefits due to reduction in losses for various sectors (as shown in above equation) are estimated as the difference between present values of future flood AAL with the project in place and AAL without improvements to the drainage system. Since these benefits accrue over the life of the project (buildings/infrastructure/systems), it is important to discount them to a present value so that benefits accruing at different times can be made comparable. The benefits are estimated as present value of future (recurring) benefits considering the life of the system. The life of system is usually considered as the minimum time period in which system will be functional. In case of drainage systems (canals, catch-pits, culverts, etc), it is taken as 20 years. The present values of future benefits are estimated as: PV = C G+d)t-1 = Cox “7 % dx(1+d) Where, PV = Present value Co = Cost (In this case average annual loss) d = discount rates (assumed 3%) t= time, years (assumed = life of the system 20 years) The outcome of the above equation is the present value of future benefits over the time (considered as the life of the system). The total present values of benefits with the proposed mitigation action is USD $ 127,505. Cost benefit Analysis As described above, various benefits and costs are considered in the cost benefit analysis to estimate the cost benefit ratio. The proposed rehabilitation of canal system is expected to benefit most of the area near Jacquezil and reduce flood losses to various exposure sectors. +. EMERGING 42 SSTAINABLE PR) Ses ERM 10 ESCI HAITI - APPENDIX 10: COST BENEFIT ANALYSIS [page 291] Table A10.4 evaluates this mitigation option in comparison to base case. The base case is status quo condition generally assumed without mitigation. Table A10.4: Cost benefit analysis of Canal Rehabilitation Other Information Life of the Canal System, Years | Grohrator | 145 | us | Flood AAL Residential Building AAL, USD 7,137 2,141 fi Commercial Building AAL, USD 5,088 1,526 métal uiéimgaat un | Present value of future Flood losses Residential Building AAL, USD 106,339 31,902 fi Commercial Building AAL, USD 75,810 22,743 métal Buiéimgaat un | Cost of Dredging, USD |__| 183,600 | rotalcos, usn |__| i8s600 Present Value of Benefits Residential Building, USD D 74,438 [| commercial Buiäimg, USD | | 53,067 | | industrial Buiaing, USD EE || rotalBeneñts,usD || a27s0s Cost benefit Ratio Cost benefit Ratio [| oæ | Base Case: The results from base case analysis of the probabilistic risk assessment have been used to establish this scenario. The results from the defined beneficiary area suggest that AAL for buildings due to floods is USD $ 12,225. The present values of future flood AAL to the buildings in the base case condition are estimated over a 20 year period to be USD $182,149. Cost benefit Ratio of Mitigation Option: Since the total cost of construction of dredging is estimated to USD $ 183,600. The decreased AAL and the present values of future flood losses with the proposed drainage improvements in place are given in Table A10.4. The cost benefit ratio (BCR) is computed by taking a ratio of present value of all benefits due to mitigation and total costs of mitigation. BCR Y Present Value of Benefits : Y Costs of Mitigation The total present values of benefits with the proposed drainage options are estimated to be USD $ 127,505 against the total costs of mitigation of USD $183,600, the BCR is 0.69. The cost benefit ratio, less than one, suggests that benefits are lower than the cost of the project, indicating that the project is a not a good investment (i.e. the benefits under weigh the costs). Considering the non-satisfactory cost benefit ratio, this option should not be taken forward for possible planning, pre-feasibility studies, etc. A10.5 Mitigation Option No. 4.: Upland Reforestation of the Trou Du Nord Watershed Deforestation and environmental degradation in Haïti has been caused by growing populations and a reliance on wood as source of energy, which places great strains on forests. The changed vegetation cover affects the 2< 4 GIDB si À .-: ERM 11 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 292] hydrological behavior of catchments and has resulted in recent times to an increase number of intense flood disasters throughout the country. The reforestation of Trou du Nord watershed is being proposed as an example of a sustainable mitigation measure to increase the interception of water in the upper reaches of watershed so as to decrease the peak and total storm flow during intense rainfall events. This mitigation measure will also work to prevent soil erosion and contribute to forest conservation which in turn enhancing carbon dioxide removal. Key considerations for the implementation of this measure are to understand that the potential benefits will be tied to a longer time horizon (i.e. storm flows depend on the maturity of the forest) and to ensure that the replanted areas remain undisturbed (i.e. enforcement). It is proposed that the barren hillsides (which are results of degradation) be planted with commercial crops like coffee. In Trou Du Nord watershed, lands that are barren are estimated to be about 487 ha. The areas are mostly in the upstream of the watershed (Figure A10.5), and contribute significant amount of flows to main stream of Trou Du Nord. Frs rreow sen J tr f è >. : SR CUIR M F5 Extent 100 Year Return Period CE L UN \ ] MM 5: to Raorstrations + 4 > s VA 2 Beneficiery Aress de a un] 0 1,5003,000 6,000 Meters ——— sen sen LL hu den rralen Figure A10.5 Trou du Nord Watershed and Re-Forestration Areas Approach The reforestation of Trou du Nord watershed is being proposed as an example of a sustainable mitigation measure to increase the interception of water in the upper reaches of watershed so as to decrease the peak and total storm flow in intense rainfall events. It proposed that the barren lands (which are results of degradation) may be planted with coffee. The planting of commercial crops will improve the water holding capacity of the soil and reduce the peak flows. The hydrological and hydraulic models developed in this study have been used to estimate the impact of the planting of upland crops on the reduction of flows. The abstraction parameters (initial losses, infiltration losses, etc.) have been estimated with improved land use and model simulations which suggest that there would be a reduction of flooding by 27%. The reduction is estimated higher in lower return periods (low magnitude events) as compared to higher return periods. With reduced flood inundation, it is expected the risk of flooding in the area of interest (Trou Du Nord Watershed) will also be reduced. Probabilistic risk assessment results derived earlier in this study have used to estimate the 2 $*. Nas BIDB Ke EC UN, Gris SR ERM 12 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 293] reduction in risk and overall benefits (losses avoided). The base case risk metrics (AAL, LEC and Loss Cost) have been used to identify and blocks within the area of interest which are likely to be benefited. As shown in the above map, about 67 blocks in the area of interest benefits from this mitigation action. Finally, the benefits to the area of interest are calculated by comparing the base case (AAL) and the AAL with the mitigation options in place. Cost benefit ratio measures the costs incurred and the benefits accrued from a policy or action - in this case the watershed re-forestation. Costs: The costs primarily include the costs of planting of coffee in upland areas. The project cost for crop planting in 487 ha area is estimated at about USD $ 1,217,500 considering a unit cost of USD 2,500/ha. The costs are assumed one-time investments with expected benefits spread over the life of the crop. The life of the crop has been taken as 20 years. Benefits: Different benefits due to mitigation are considered for the cost benefit analysis. For this project, a portion of the direct tangible benefits of a flood damage reduction is expected as inundation decreases. Mathematically, we compute this inundation-reduction benefit B; as: Bir = Xwithout — Xwith in which Xwithout — expected damages or economic impact without project; and Xwith - expected damage or economic impact if the project is implemented. The benefits include reduced losses to residential and commercial buildings due to reduction in flood flows. The benefits due to reduction in losses for various sectors (as shown in above equation) are estimated as the difference between present values of future flood AAL with the project in place and AAL without improvements to the drainage system. Since these benefits accrue over the life of the project (buildings/infrastructure/systems), it is important to discount them to a present value so that benefits accruing at different times can be made comparable. The benefits are estimated as present value of future (recurring) benefits considering the life of the system. The life of system is usually considered as the minimum time period in which system will be functional. Though, the average life of the crop is 40 years, but the benefits starts with the complete maturity of the crop and established of renewed hydrological cycle in the watershed. Hence, in this case life of the crop is taken as 20 years. The present values of future benefits are estimated as: PV=C G@+d)-1 = Cox = 9 dx(1+d)t Where, PV = Present value Co = Cost (In this case average annual loss) d = discount rates (assumed 3%) t= time, years (assumed = life of the crop 20 years) The outcome of the above equation is the present value of future benefits over the time (considered as the life of the system). The total present values of benefits with the proposed mitigation action is USD$2,772,785. Cost benefit Analysis As described above, various benefits and costs are considered in the cost benefit analysis to estimate the cost benefit ratio. The proposed watershed reforestation is expected to benefit most of the area in Trou Du Nord watershed and reduce flood losses to various exposure sectors. +. EMERGING 42 SSTAINABLE PR) Ses ERM 13 ESCI HAITI - APPENDIX 10: COST BENEFIT ANALYSIS [page 294] Table A10.5 evaluates this mitigation option in comparison to base case. The base case is status quo condition generally assumed without mitigation. Table A10.5 Cost benefit analysis of Watershed Reforestation Other Information Life of the Plantation, Years RS CE CES Flood AAL Residential Building AAL, USD 520,112 379,682 [| commercial Building AAL, USD 169,121 123,459 À industrial Buramgaat un | | Present value of future Flood losses Residential Building AAL, USD 7,749,665 5,657,255 [| commercial Building AAL, USD 2,519,908 1,839,533 À industrial Buramgaat un | | Cost of Plantation, USD | | 1217500 [| Total cost, usD || arssoo Present Value of Benefits Residential Building, USD D 2,092,410 | commercial guiaing, un | | 680375 | |" imdustril Buiaing, usD EE | | rotal Beneñis, usD | armes Cost benefit Ratio Cost benefit Ratio | | 223 | Base Case: The results from base case analysis of the probabilistic risk assessment have been used to establish this scenario. The results from the defined beneficiary area suggest that AAL for buildings (RES and COM) due to floods is USD $ 689,233. The present values of future flood AAL to the buildings in the base case condition are estimated at USD $ 10,269,572. Cost benefit Ratio of Mitigation Option: Since the total cost of planting is estimated to USD 1,217,500. The total present values of benefits with the proposed drainage options are estimated to be USD $ 2,772,785. The cost benefit ratio (BCR) is computed by taking a ratio of present value of all benefits due to mitigation and total costs of mitigation. BCR Y Present Value of Benefits _ X Costs of Mitigation The comparison of the total benefits of USD $ 2,772,785 against the total costs of mitigation of USD $ 1,217,500, the BCR is 2.28. The cost benefit ratio, greater than one, suggests that benefits are higher than the cost of the project, indicating that the project is a good investment (ï.e. the benefits outweigh the costs). Considering the satisfactory cost benefit ratio, this option may be taken forward for possible planning, pre feasibility studies, etc. A10.6 Mitigation Option No. 5.: Mangrove Reforestation in the Parc National Trois Baies Forested wetlands that are linked to the coastal environs serve as a natural buffer against storm surges, sea- level rise and wave action in particular. Natural features help to absorb large volumes of advancing water, and as a result, have a dissipating effect on wave energy. In this regard the wetlands, particularly the mangrove forests, that are part of the Parc National Trois Baies, are vital for mitigating the effects of coastal flooding in + EMERGING Lo “6 GIDB LR ERM 14 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 295] Northern Haiti. Mangrove forests in particular play an important ecological role while providing a variety of services that include protection from erosion, flooding cyclones and tidal waves. An increase in the spatial extent of mangrove forests in the Parc National Trois Baies will not only serve as a mitigation measure against coastal hazards and climate change, but also contributes to conservation Haïti ecological systems, natural environment and biodiversity that are becoming increasing important to everyone around the world. Along the coastal line of about 60 km of Parc National Trois Baies, it proposed that a 5km strip of mangrove be reforested. This would mean about 300 sq.km of area would be planted. The targeted area would benefit about 138 blocks from coastal flooding (Figure A10.6). rrsow raw nsson u F7 HUSEN wsen \ CN L'N di LD er “in dE PS, O7 à ne, De V4 ] CI D (el À j | ie — à FLAT LS met | 4 f 2 EE à Fa VE Dan à me pre Cu 17400N =" AVS Ai Vu F Le à TELL 7 sn en ÿ | RATES) 3 ns ls \ KT SN \ 517 Legend L | ’ y j 4: KI A || Beniiciaryy Ares ; RO. Ke CC] 3 80 Part Boundery ” ” - fe PR LS 1] Block Boundary — AN ne 74 Coastal Flood 100 Year Retrun Period |... L 17503500 7,000 Meters à ls APS ne: j L s Low rrden Fo Fe rrslon Figure A10.6 Beneficiary Areas from Mangroves Approach Natural features like Mangroves help to absorb large volumes of advancing water, and as a result, have a dissipating effect on wave energy. In this regard the wetlands, particularly the mangrove forests, that are part of the Parc National Trois Baies are vital for mitigating the effects of coastal flooding in Northern Haïti. It is proposed to reforest a 5km of strip of Mangroves along the coastal line of about 60 km of PN3B comprising 300 sq.km of area. With reduced flood inundation, it is expected the risk of flooding in the area of interest (in 138 blocks on northern coast of Haiti) will also be reduced. Probabilistic risk assessment results derived earlier in this study have used to estimate the reduction in risk and overall benefits (losses avoided). The base case risk metrics (AAL, LEC and Loss Cost) have been used to identify and blocks within the area of interest which are likely to be benefited (approximately 138 blocks). EMERGING = Lo «8% GIDB Métisties ERM 15 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 296] Finally, the benefits to the area of interest are calculated by comparing the base case (AAL) and the AAL with the mitigation options in place. Cost benefit ratio measures the costs incurred and the benefits accrued from a policy or action - in this case the Mangrove protection. Costs: The costs of this measure include the costs of planting mangrove trees. The project cost for crop planting in 300 sq.km area is estimated at about USD $ 6,750,000 considering a unit cost of USD 22,500/sq.km. The costs are assumed one-time investments with expected benefits spread over the life of the system. The life of the system has been taken as 50 years. Benefits: Different benefits due to mitigation are considered for the cost benefit analysis. For this project, a portion of the direct tangible benefits of a flood damage reduction is expected as inundation decreases. Mathematically, we compute this inundation-reduction benefit B, as: Bir = Xwithout — Xwith in which Xwithout — expected damages or economic impact without project; and Xwith - expected damage or economic impact if the project is implemented. The benefits include reduced losses to residential, commercial and industrial buildings due to the mangrove protection. The benefits due to reduction in losses for various sectors (as shown in above equation) are estimated as the difference between present values of future flood AAL with the project in place and AAL without improvements to the drainage system. Since these benefits accrue over the life of the project (buildings/infrastructure/systems), it is important to discount them to a present value so that benefits accruing at different times can be made comparable. The benefits are estimated as present value of future (recurring) benefits considering the life of the system. The life of system is usually considered as the minimum time period in which system will be functional. Though, the average life of the mangrove is generally more than 50 years, but the benefits starts with the complete maturity of the plats and established of renewed ecosystem cycle in the area. Hence, in this case life of the system is taken as 50 years. The present values of future benefits are estimated as: PV=C G@+d)-1 = Cox ——————— 9 dx(1+d)t Where, PV = Present value Co = Cost (In this case average annual loss) d = discount rates (assumed 3%) t= time, years (assumed = life of the system 50 years) The outcome of the above equation is the present value of future benefits over the time (considered as the life of the system). The total present values of benefits with the proposed mitigation action is UD USD $ 60,472,298. Cost benefit Analysis As described above, various benefits and costs are considered in the cost benefit analysis to estimate the cost benefit ratio. The proposed mangrove protection is expected to benefit most of the area in north coast and reduce flood losses to various exposure sectors. Table A10.6 evaluates this mitigation option in comparison to base case. The base case is status quo condition generally assumed without mitigation. 4x. EMERGING ) Rd =: ERM 16 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 297] Table A10.6 Cost benefit analysis of mangrove protection Other Information Life of the Plantation, Years À Growthactor 7 | 257 | 257 | Coastal Flood AAL Residential Building AAL, USD 4,403,444 2,421,894 [| commercial Building AAL, USD 825,462 454,004 | méustrielBuraimg aa un | | Present value of future Flood losses Residential Building AAL, USD 113,168,505 62,242,678 D Commercial Building AAL, USD 21,214,379 11,667,909 RE AS ES Cost of Plantation, USD |" 6750000 [| Total cost, usD | | 6750000 Present Value of Benefits Residential Building, USD l 50,925,827 | commercialguiaing,usD | | osaçan | |" imdustril Buiaing, usD ES | | Total eneñis, usD || coa7zzss Cost benefit Ratio Cost benefit Ratio [| ss | Base Case: The results from base case analysis of the probabilistic risk assessment have been used to establish this scenario. The results from the defined beneficiary area suggest that AAL for buildings due to floods is USD $ 5,228,906. The present values of future flood AAL to the buildings in the base case condition are estimated at USD $ 134,382,884. Cost benefit Ratio of Mitigation Option: Since the total cost of planting is estimated to USD 6,750,000. The decreased AAL and the present values of future flood losses with the proposed mitigation in place are given in Table A10.6. The total present values of benefits with the proposed drainage options are estimated to be USD $ 60,472,298. The cost benefit ratio (BCR) is computed by taking a ratio of present value of all benefits due to mitigation and total costs of mitigation. BCR Y Present Value of Benefits : Y Costs of Mitigation The comparison of the total benefits of USD $ 60,472,298 against the total costs of mitigation of USD $ 6,750,000, the BCR is 8.96. The cost benefit ratio, greater than one, suggests that benefits are higher than the cost of the project, indicating that the project is a good investment (i.e. the benefits outweigh the costs). Considering the satisfactory cost benefit ratio, this option may be taken forward for possible planning, pre feasibility studies, etc. A10.7 References Kunreuther, H., Cyr, C., Grossi, P., Tao, W., Using Cost-benefit Analysis to Evaluate Mitigation for Lifeline Systems, Wharton Risk Management and Decision Processes Center, The Wharton School, University of Pennsylvania, 2001. #8 GIDB nn? \ ERM 17 ESCI HAITI — APPENDIX 10: COST BENEFIT ANALYSIS [page 298] APPENDIX 11: IDB Water Study - Simulation Model Development Results NORTHERN DEVELOPMENT CORRIDOR, HAITI Appendix EC EU ERM [page 299] Development of a Flood Simulation Model for Infrastructure Planning s«and-Design'at the Parc Industriel de. ( Caracof (PIC) in- Northern Haiti ”\.. x N > AG + = 1e 3 C res [page 300] Background ° The PICis located in a low-lying area, as well as downstream in a major river basin. ° Naturally flooded environment, characteristic of areas near coastal wetlands. * Several flooding events have been reported at the PIC ° There is a need to get the flooding situation at the PIC under control as the site continues to develop. [page 301] Z A. EE ET E PE PSE. . - < L - 1 bé le Le. AN $ | ben. y 4 19 s | AZ \ , ; | LORS / M Léat 27 cl ru k ' A : d Fr HAL DC 1 voue ». "© à PP} 5 PTE Cor - & NORDEÆSY \ ef" rt SUR Da 4 ». ’ Le J Le 44 Le } 2 12649 1 É Gi 1 C7 74 em) ST Cmirieur Départerenis ! Æ Er] FAC AMGET. 3 Lt ets » = Sem [__] Lane Secton Coran | @ ee ht ve TR TR L'ARRNEE LT ITA LE Fe ‘ vai f- Aisque Dinnondation #- LÉ be use Al A7: TE EST 7 (> see Fu Dù 167 AT a 9 LAN ER E NE [page 302] REGIONAL FLOOD RISK (NATHAT) f Quartier , J Ta ‘q À umonmde | D; : Terrier Rouge [a P Da “al ATroudu /#, LA OUT ES 22 # WN\ 2\ | Nord te, TT SR" = & A AN PERS EN CUT R 7 Ë ee 72 ù . . i} À ; » N fi Mn) gAcul Sim} ñ : ! k - LEGEND SCALE Average Flood Risk —— Rivers and Waterways Ocean ré Lagoon Low Flood Risk #) Cities Coral Reef = Development Corridor 95 5 Kilomètre -- Floodplains Mangroves ( ) Dominican Republic [page 303] About this Project ° Title: Water availability, quality and integrated water resources management in northern Haïti. * Objective: to quantitatively assess current and future water availability and quality and water demand by all stakeholders as key inputs to integrated water resources management (IWRM) in northern Haiti [page 304] Project Activities ° 1. Institutional and governance analysis of in- country water resources management ° 2. Data gap analysis and compilation of available modeling data in the PIC andits contributing watershed ° 3. Development of hydrologic models ° 4. IWRM Plan for the Trou du Nord - MTA System [page 305] Simulation Model Development ° Part of Activities 2 and 3 ° Support sustainability of water resources and other natural and built infrastructure in the area of the PIC. ° A particular focus is flooding and stormwater management. [page 306] Simulation Model Development ° What does a “simulation model” do? ° “Thinking” role ° “Diagnostics” role ° “Analysis” role ° “Solutions” role ° “Education” role [page 307] For this type of problem, we conceived “Hydro-BID” An integrated and quantitative system to simulate hydrology and water resources management in the LAC region, under scenarios of change (e.g., climate, land use, population) which allows to evaluate the quantity and quality of water, infrastructure needs, and the design of strategies and adaptive projects in response to these changes. [page 308] The Hydro-BID Simulation Syst fi pan J MN CEA LE] re LA Me œe NE L oi PA ANT ma ns #4 nv, United States ne CT + AZ. $ AE 2 yo Atlantic sd MS Part sc Ocean TX à GA! : ° _”200,000+ catchments/stream/segments throughout LAC À Fe 7 7e 1 LA = ° Average size: 80 km?, 10 km « D [page 309] . \@ - ne F | # UC » x À é . TRS ; | = Satellite Data : Precipitation à, River Gauge Data Weather Estimates Radar Data Observations Precipitation és Snow Cover! Forecasts Melt Data ps _ DA Ce” £a SRTM Terrain Data R LA ; aSArs ——— 71 OU I ELATNS De. ll , Climate Predictions gr panigit - éditions nn } . . Decisions [page 310] Componentes de Hydro-BID (current) Input Data Soil type Analytical Land Use GWLF Hydrography Î Rainiall Rainfall-Runoff Dataset (AHD) W Temperature Reference flows Basin delineation (Er GEÈIEN EN) and drainage network(s) Time series of flow Water rates in each basin Risk analysis and ro specs for adaptive Mo infrastructure (eg S HRGORE RON Water demand(s) WEAP) Prices/costs [page 311] AHD: Analytical Hydrographic Dataset + ANDis a database that is available for the entire LAC region. De # + __ Completely derived from a digital elevation map (DEM) using SRTM . (NASA) a m average ADN SD Des) LS + This data was processed to se En delineate basins and determine RTS ON RL > the drainage network and L NS FN. ce LE, DE PSS : connectivity. Da SRE CRE EE à PA [page 312] ITOU-QU-INOrTQ VVatersnEeu , LUE TPE. es EU [page 313] Massacre Transboundary Aquifer (basins delineated in Haiti-AHD) PÉCPUS CE NAT OR SVM, PES € VAS TI FAC a +. 2 Cr Ée 1/1 E- — om one { 9 7 ! à & La , à ff, A ot à “LR ? LAT | . \ LA S + a Ë 4 \ A Se p: | 14 4 [page 314] Haiti-AHD (entire country) De aise s : à TS ne E 109. ANR CAS ee LE REN SSSR re D om to Ten ï D nes = RS ee ARE RS NAN LE AC KT ANRT NL Ve RARE ho en nd ne nee Ne ESS _ ne RAS RE ie PRINT PE GE DNS f = N s ne , ss * Ke TIRE Le ui) #-. Se ; D. Use © QGis 2014 [page 315] ° _ Covers the entire LAC region: useful to organize and aggregate scarce data; the whole is greater than the sum of its parts * _ Spatial and temporal resolution suited for planning and design of water resources infrastructure °__ Simulates basin hydrology driven by climate in a modular, flexible and scalable way; robust hydrologic model formulation that is able to interact with just about any type of climate model or data source * Tailored to simulate water resources at all time-scales: near term, intra/inter-annual, decadal and beyond * _ Developed using web-based architecture: runs from a browser-like (app) interface * _ Open-source: designed to be community-driven, opening the doors to a rich development and improvement process *__ Available soon at http://hydro-bid.net [page 316] About “Hydro-BID 2D” ° Simulates flood depths and flow rates (velocities) for a given combination of rainfall, topography, soil, land use, and infrastructure components. ° Two-dimensional (2D), mesh-based formulation, latest graphical processing unit (GPU) technology (runs very fast). * Developed by Hydronia LLC and Universidad de Zaragoza (Spain) [page 317] LOW TIOWS réak VISCNATEE : Êl EE — 7 A TTOMITT [page 318] 4 F. Nil ê AE 247 a. U: LE # | 5 - ; _ — . + SERA VAVATA RES SA à # N IA Ar ; « . LE SX) Sel LVAVAN" CÉRRESE VA ÿ l N A HIRRE Se _ AS SACREEE RE VS Q: fl é« nt DE DH LEFEE Æ KE À SR en ». ZA), a site NUE EN NES | a ee RO EERE Re KA ZZ | { 7. DOUTE AS FRA _ dde AS h ere Le NE Sn . a Re HA Fu PRES “ ZA; DER DC D FUIUAESS EE © C2) NET a Fe d, =. se . me re HN) K£ k \ k LTÉE ARR TS : oo PÉTER SEXE RÉ É … Re VAVAVA (RooOoe AA [page 319] Data inputs * Topography: digital elevation model with 2m spatial resolution (from IDB-ICES project) * Soils: designated areas based on aerial photography (from IDB-ICES project) * Rainfall data (SNRE) * Resolution: spatial (2 m min; 50 m max); temporal adjustable (a few secs) ° Design storm calculation: SCS Type Ill curve; good for hydrograph generation in the Caribbean basin [page 320] Monthly Rainfall Data Seasonality Cap Haitien Rain 250.00 200.00 150.00 100.00 50.00 0.00 + r T r r r [l r r r r r r 1 Source: SNRE [page 321] L] Monthly « Region + Daily M: Port-de-Paix — 2 AM RUME Cap-kHgitien Ponte Cristit 24e Es Para : ; LUE N Valverds ee, SL Monde “ [ GOMAVES ’ \ Nord-ESK Sar ÆArtibonte Daison Santiago'R St Louis de Gonzague (Se CP , ee faiti Centre à 5 € 7m “ >. Sanuan \ z à a, L | Le = Kz San Wuan defa/Mag Jerémie CPE L'on à ’ ue ON ITR N SN (Sa rT/ DS © ; MOues UE MGNnce Bahonuco : SE Nippes : Independencia PR j SUR F SES! ÿ _ ne PART» Lcmele c / Barahona < Les Cayes : } ds Fe PeGoogleéarth Stations in study region have only monthly data. Daily datasets are mostly clustered around Port-au-Prince [page 322] Rain SLGonzague 250 200 -+ 150 - 100 - ATCLLRTRRRE — Cap Haïitien Monthly —=SLGonzague monthly rain 20 15 10 5 0 -5 -10 SD comparison of calculated SL Gonzague monthly data with Cap Haitien monthly dataset [page 323] Synthetic vs Measured Monthly Rainfall at Cap Haitien — Cap Haitien Monthly Model ——Cap Haitien Monthly 20 15 Tv. -10 [page 324] Simulated Scenarios ° “Sunny Day” ° Rainfall ° Multiple return periods ° Water depths and velocities * Dilution of wastewater treatment plant discharge [page 325] “Sunny Day” Scenarios [page 326] Sunny Day Scenarios ° Hydrographs from upstream of the basin obtained by hydrologic analyses * Return Period (years) = 1,2, 5, 10, 25, 50, 100, 200 ° No rainfall on the PIC * Flooding comes from upper watershed river flow [page 327] Max Depths. Sunny Day, 10 year event s } En T4 . PE À *, > pa ‘ g 0.00 [page 328] Depths. Sunny Day, 10 year event RiverFlow2D Depths TIME: 0000:00:00:02 DDDD:HH:MM:5S 2178582.0 2177923.8 [=] [em] ET (nu [I] 10.00 9.47 = En 8.95 En77265.8 [7 8.42 > 7.89 es 7.37 6.84 S = 6.32 à £ 5.79 = 5.26 & 4.74 2176607.5 à 421 3.68 3.16 2.63 2.11 1.58 1.05 0.53 2175949.5 0.00 8124508 813108.8 813766.9 8144250 815083.1 X(m) [page 329] Depths. Sunny Day, 25 year event RiverFlow2D Depths TIME: 0000:00:00:02 DDDD:HH:MM:5S 2178582.0 2177923.8 [=] [em] ET (nu [I] 10.00 9.47 = En 8.95 En77265.8 [7 8.42 > 7.89 es 7.37 6.84 S = 6.32 à £ 5.79 = 5.26 & 4.74 2176607.5 à 421 3.68 3.16 2.63 2.11 1.58 1.05 0.53 2175949.5 0.00 8124508 813108.8 813766.9 8144250 815083.1 X(m) [page 330] Inundation times. Sunny-day, 50-year event Sn . Ua nil L 0 | 23.750 LITE eee trs 19.050 PRE V Se 2 16.700 EL] Le Es 14.350 on | 5 12.000 + 9.650 LG "or 7.300 48/7 4.950 bé 2.600 1, SRE 0.250 [page 331] Depths. Sunny Day, 50 year event RiverFlow2D Depths TIME: 0000:00:00:02 DDDD:HH:MM:5S 2178582.0 2177923.8 [=] [em] ET (nu [I] 10.00 9.47 = En 8.95 En77265.8 [7 8.42 > 7.89 es 7.37 6.84 S = 6.32 à £ 5.79 = 5.26 & 4.74 2176607.5 à 421 3.68 3.16 2.63 2.11 1.58 1.05 0.53 2175949.5 0.00 8124508 813108.8 813766.9 8144250 815083.1 X(m) [page 332] Inundation times. Sunny-day, 100-year event | 7 le SE hrs ÿ on u # L. | 23.750 te Th _ r- =. ùi | 21.400 DUT “ref rene) 19.050 A #) | dés 16.700 tre à À : 14.350 . Sn 12.000 Fr 9.650 Æ … a 7.300 » LEA 4.950 . 2.600 sa LS 0.250 [page 333] Depths. Sunny Day, 100 year event RiverFlow2D Depths TIME: 0000:00:00:02 DDDD:HH:MM:55 2178582.0 2177923.8 [=] [= (es) —Ü] 10.00 9.47 = = 8.95 Én7r2658 C7] 8.42 > 7.89 _ 7.37 5.84 = s = 6.32 £ 5.79 = 5.26 & 474 2176507.5 à 421 3.68 346 2.63 241 1.56 1.05 0.53 2175949. 0.00 8124506 813108.8 813766.9 8144250 8150831 X (m.) [page 334] Inundation times. Sunny-day, 200-year event | + # $ _ L Le | Ê re times 14 > 2 | | #) 23.750 1e Sn nn” - / 21.400 Ê [1 1: + ef ERA 19.050 . = nr 4 ( 16.700 oaus D ) | 14.350 | 12.000 & d 9.650 ÈS Peu 7.300 1& A 4.950 MUST NA - 2.600 \ 0.250 [page 335] Depths. Sunny Day, 200 year event RiverFlow2D Depths TIME: 0000:00:30:00 DDDD:HH:MM:SS 2178582.0 2177923.8 [=] [—.] [mme hi} 1.02 0.96 = sn 0.91 E 21772658 [7 0.86 > 0.80 0.75 =. 0.70 s 0.64 € 0.59 = 0.54 2048 2176607.5 à 0.43 0.38 0.32 0.27 0.21 0.16 0.11 0.05 2175949. 0.00 812450.6 813108.8 813766.9 814425.0 815083.1 X (m.) [page 336] Rainfall Scenarios [page 337] Rainfall Scenarios ° Hydrographs from upstream of the basin obtained by hydrologic analyses * Return Period (years) = 1,2, 5, 10, 25, 50, 100, 200 * Rainfall events occur on the PIC * Flooding comes from upper watershed river flow plus local rainfall-runoff. [page 338] 40 35 30 25 £ £ £ £ Z £ 20 £ £ & LI] = ü & 15 10 0 tllill [| D snnnnnnnennnnnne F On Hu Nu mu + un un un vw un R un œ un nu ou du NN Mu + UN UN UN LUN UN UN NUN OU AUN QUN MAN + ES TA A Où 6 Rd diet dm Tu Un te do À SN SN Q À 6 À È ETATATITETITETITET TARA à £ Ê Time (hr) [page 339] Inundation times. Rainfall, 1-year event SR £ r dire imes Ee sé fi IT . 23.750 mo. — el ES 21.400 € TES à. pee es 19.050 EL A al 7 16.700 Es. Le < 14.350 , Pa À 12.000 — 9.650 A À of 7.300 s. 4 ‘À 4.950 2.600 3 0.250 [page 340] Depth infall, 1 r event (© ©) S. Ra all, L year eve RiverFlow2D Depths TIME: 0000:00:45:00 DDDD:HH:MM:S5S 2178582.0 2177923.8 LI [em | L _] = Il 0.46 [7 L_ 0.43 = = 0.41 Ê2177265.8 [| 0.39 > 0.36 0.34 > 0.31 = s 0.23 € 0.26 0.24 80.22 2176607.5 A019 0.17 0.14 0.12 010 = 0.07 / 0.05 / 0.02 2175949.5 0.00 8124506 813108.8 813766.9 8144250 815083.1 X Cm.) [page 341] 100 90 80 70 £ 60 È E Es 2 50 2 É £ = È £ 40 [4 30 20 | | | | 0 LIL | D snnnnnnnnnnnmnnn FTOuHUNNUN MU S+UNUNUNLvuNNR UN x UN NU oO UN AN AN MN MN HAN UN UN LUN RUN uN NN ON LAN QUN M UN + È s Li LI mn LA ui s = co Hot AT TmT TA ART 1 a NS N SN Q NN oi È Sata ti at St ST SAN AN à E Ê Time (hr) [page 342] Inundation times. Rainfall, 2-year event ji Fr + « à — - Inundation times : © D D: ne 23.750 uns P—— 16 À F | 21.400 EDS 4 : 19.050 da 4 s à », 14.350 Re, 12.000 F3 " 9.650 _# À Pa 7.300 CS * L à 4.950 el 2.600 "5 0.250 [page 343] L] Depths. Rainfall, 2 year event RiverFlow2D Depths TIME: 0000:00:30:00 DDDD-HH:MM:SS 2178582.0 2177923.8 [=] — LT 1} 0.40 0.38 = on) 0.36 E 2177265.8 [} 0.34 > 0.31 0.29 = 0.27 s 0.25 € 0.23 = 0.21 20.19 2176607.5 8017 0.18 0.13 0.10 0.08 0.06 0.04 0.02 2175949. 0.00 812450.6 813108.8 813766.9 814425.0 815083.1 X (m.) [page 344] 160 140 120 = 100 £ £ £ £ E 2 80 2 £ & = £ 5 É 60 40 | | | | 0 till [] Ellonnnnnnnennnnnnm Fou HN Nu Mu s+unununvunR un x un nu oUuN AU AN UN M UN SUN UN UN LUN RE un œ UN Nu OU du Qu Mu + Et St At dm + Où D Rd aie Tatin due in te do NS N SN Nm N È STATATITEATATETETATITERT TAN ® Ê Ê Time (hr) [page 345] Inundation times. Rainfall, 5-year event 5 +. , << Fi 23.750 oo all ; en à 21.400 0e Sr = 19.050 AUTRE ? È 16.700 FRS 4 ; 14.350 — LE 12.000 «] aix à 9.650 g ? Col: 7.300 ; G LA 4.950 AL 2.600 Si LR 0.250 # ) [page 346] h infall Depths. Rainfall, 5 year event RiverFlow2D Depths TIME: 0000:00:30:00 DDDD'HH:MM:SS 2178582.0 2177923.8 [=] |. LT 1} 0.47 0.44 = us 0.42 E 2177265.8 [} 0.39 > 0.37 0.35 ) 0.32 me =. - 0.30 € 0.27 = 0.25 £0 22 2176607.5 à 0.20 0.17 0.15 0.12 { 0.10 0.07 0.05 / 0.02 2175949.5 0.00 8124506 813108.8 813766.9 814425.0 815083.1 X (m.) [page 347] 10-yr Storm Event 180 160 140 120 £ £ £ Æ£ 100 E ia E 5 £ = 80 £ CI È 3 [2 60 40 | | | | 0 til | Llsnnnnnnnnnnnnmne © 1 M Nu mu + un un un À un RO un © nn a un © un dun Ni A M un + M A un © un R un un on un © un du Qu m un + È s “ a m + wi 5 = oo tot ai mt ni AR Ad a NS NN a Nm E STATATAT ETATS TETE ENS ANRT ri € Ê Time (hr) [page 348] Inundation times. Rainfall, 10-year event RL. ll | Be 23.750 PE — ES + : & 21.400 æ, [TT EN CR RE A | 19.050 cs x + k 3 16.700 / LES La _ Ve 14.350 nt, , 12.000 ! a j 9.650 ‘@. ST 7.300 à de 4.950 TE 2.600 : # rè 0.250 1.4 [page 349] L] Depths. Rainfall, 10 year event RiverFlow2D Depths TIME: 0000:00:00:02 DDDD:HH:MM:5S 2178582.0 2177923.8 [=] [em] ET (nu [I] 10.00 9.47 = En 8.95 Ê2177265.8 [7 8.42 > 7.89 es 7.37 6.84 S = 6.32 à £ 5.79 = 5.26 & 4.74 2176607.5 à 421 3.68 3.16 2.63 2.11 1.58 1.05 0.53 2175949.5 0.00 8124508 813108.8 813766.9 8144250 815083.1 X (m.) [page 350] 250 200 £ 150 £ £ E à ê 2 £ £ Li] < 100 L4 50 | 0 LL antilil funntesssusrs…… F ON Hu Nu mu + un un un vu Run un un © un æ un Nu M un + un nn un Lu R un © n A un © N du Qu M un æ ES À ON Om OS Où OR à om S ti dû tot ui dun À og À oi N SN SN oi À £ STATATITETITS TETE TIR NR AR ® E Ê Time (hr) [page 351] Inundation times. Rainfall, 25-year event R F4 N » Lie 4 + 1 4 » Inundation times # DE Le L, u J ù “ (hrs) EL, ; = til p | le ) Li 23.750 ve ES 21.400 4 ui 1: pee my nn 19.050 | 4 = 4 LT > 16.700 N us. . d À à +. 14.350 1 E 7 | ; 12.000 : AT à L. 9.650 @ LA ol: 7.300 4 Le LA 4.950 RAT F9 2.600 x. 0.250 [page 352] h infall Depths. Rainfall, 25 year event RiverFlow2D Depths TIME: 0000:00:30:00 DDDD:HH:MM:SS 21785682.0 2177923.8 [=] — LT 1} 0.65 0.62 = = 0.58 E 21772658 [} 0.55 > 0.51 0.48 = 0.45 =. 0.41 € 0.38 = 0.34 £0 31 2176607.5 © 0.27 0.24 0.21 } 0.17 0.14 0.10 0.07 0.03 2175949.5 0.00 812450.6 813108.8 813766.9 814425.0 815083.1 X (m.) [page 353] 300 250 200 £ £ £ £ E 2 150 2 £ £ = £ 5 L4 100 50 | 0 a nntlil fitness. © 1 M Nu mu + un un un À un RO un © nn a un © un dun Ni A M un + M A un © un R un un on un © un du Qu m un + E ST À Om + Où 6 OR d ae td toi ts du TG AR À 0 NS NN Q N mi £ ATATAITAITETITETIT ÉTAIT NUS & E Ê Time (hr) [page 354] Inundation times. Rainfall, 50-year event cs mA s; ER V7. È 2: 1 : | L * és »“ 7. rundstontimes na _ " ( ( p): 23.750 à g— an À. gd un # 21.400 4 NI PRE Ms/ “T4 19.050 Ê ALLIE % À 16.700 > | 4 7 À 14.350 ” » - & 12.000 ñ : HO Aix Fe 9.650 © À PU 7.300 Ke L Le 4 sl 4.950 F) ” #8 ï) 2.600 Ton "és Le 0.250 es LT: à [page 355] h infall Depths. Rainfall, 50 year event RiverFlow2D Depths TIME: 0000:00:30:00 DDDD:HH:MM:SS 21785682.0 2177923.8 [=] — LT 1} 0.76 0.72 = = 0.68 E 21772658 [} 0.64 > 0.60 0.56 a 0.52 me =. — 0.48 € 0.44 = 0.40 £0 36 2176607.5 © 0.32 0.28 0.24 0.20 0.16 0.12 0.08 0.04 2175949.5 0.00 812450.6 813108.8 813766.9 814425.0 815083.1 X (m.) [page 356] 300 250 200 £ £ £ £ E 2 150 2 £ £ = £ ü [2 100 50 | 0 || Lan © 1 M Nu mu + un un un À un RO un © nn a un © un dun Ni A M un + M A un © un R un un on un © un du Qu m un + EST TA mn Sn 8 RO D RABAT AUTRUI S ANA IR INT NN ANAQN MA E STATATAT ETATS TETE ENS ANRT ri E Ê Time (hr) [page 357] Inundation times. Rainfall, 100-year event à à 4 * ee 4 / Le Inundation ti À AU | s #2 d ere Rs à "= l'an -» DR 1 23.750 + Er > 79 rip À. w/: ; j 21.400 À D'PAE D A - : 19.050 # 3 LL # 16.700 | a a, 16 9.650 AC D de À 7.300 KI | D” | 4.950 AT #8 ÈS 2.600 4 0.250 [page 358] L] Depths. Rainfall, 100 year event RiverFlow2D Depths TIME: 0000:00:30:00 DDDD:HH:MM:S5S 2178582.0 2177923.8 LI [em | L _] — Il 0.89 ns 0.84 = oi 0.79 En77265.8 [| 0.75 > 0.70 0.65 > 0.61 = 5 0.56 à É0.51 0.47 0.42 2176607.5 à 0.37 0.33 0.28 0.23 019 2 0.14 / 0.09 / 0.05 2175949.5 0.00 8124506 813108.8 813766.9 8144250 815083.1 X Cm.) [page 359] 350 300 250 £ Ê Ë 200 E ia E 2 £ & 150 = £ ü [2 100 50 0 | (laut. © 1 M Nu mu + un un un À un RO un © nn a un © un dun Ni A M un + M A un © un R un un on un © un du Qu m un + È s “ a m + wi 5 = oo tot ai mt ni AR Ad a NS NN a Nm E STATATAT ETATS TETE ENS ANRT Fi € Ê Time (hr) [page 360] Inundation times. Rainfall, 200-year event ) pr LL) L : e Ee & f f à. rundaion times eu Pa | LE: F 23.750 re ii À. din. e 21.400 | d'I VIE" # 19.050 h} w 4 16.700 Le / a 14.350 é 2 ‘= P à 12.000 î ECS mt 9.650 mp, | CÆ 7.300 Ÿ Ù FC « 4.950 2 € Na l 2.600 ; NN 0.250 [page 361] L] Depths. Rainfall, 200 year event RiverF low 2D PR TIME: 0000: 00:30:00 DDDD:HH:MM:5S 2178580 2717938 5 | =, ! LL] Eznr2658 l 0.85 > = 0.20 Se us à 0.4 2176807 5 à 0.4 Le. 0.3 02 021 0.16 0.11 2172495 0.00 8124506 813108. 813766.9 siu250 810831 X (m.} [page 362] Peak discharges in Trou-du-Nord River [page 363] 1.000 0.900 0.800 & 0.700 Éd Es rl A T5 # 0-600 —Tr=-1 S —Tr 2 È Tr=s S 0.500 Ê ——Tr =10 (°] ee = —Tr=25 5 0.400 Ir =50 È —Tr =100 & 8 5 —Tr =200 © 0300 0.200 0.100 0.000 0.00 50.00 100.00 150.00 200.00 250.00 300.00 350.00 Distance from discharge (m) [page 364] Total N dispersion simulations [page 365] Total N Dispersion Scenarios ° Treatment Plant — Q = 0.04 m°/s — Total N = 10 mg/l ° Trou-du-Nord River — Q = 0, 1,5, and 20 m°/s — Total N = 0.5 mg/l [page 366] nAydaro-BID 2D model refined mesh MO CN ERE Rn NOR. CU de À ee. [page 367] n 3 Irou-du-Nora Kiver Ü.m°/s &: ONE ET, HOTTES 4, CP nr: ue "7 Éd PTE ie = ss Por: d e P [> 1 F ° Fe * . Le. . nd ° Ds CT A LEE AE CINE SEL ee 3: mt... =. “ e s SOS TE CUS CEE UT OR PENSE PR 4 Res CRAN 2e SON RE Le = à FR % ne Ler RUES | # 54 Fo SA PA F p . k [page 368] l 2 Irou-dau-Norda Kiver 1 m°/s PRET Lies # TRE CERSRERINNR He CRAN NS GO à PARENT a ? {a \ a , . LÉ < . ee > P PA : "#4 [ Pr > she: Lot F? 12e, Per : : “A Pan PEER LR 4 . ve S E , [page 369] Trou-dau-Nord Kiver 5 m°/s | F7 ni : NO D: PONT fase Te ÉÉRR EE : [page 370] n ») Trou-du-Nord River 20 m°/s NES ÉCN , Pr = Ne. gris ASE A PL Act à. Fr ge Se Fe ES [page 371] Total N Concentrations from discharge point 35 3 F Trou-du-Nord River Discharge z 2 > —— 0 m3/s z — 1m3/s ë 15 —— 5 m3/s ‘ —— 20 m3/s 1 0.5 0 à 50 100 150 200 250 300 350 Distance (m) Treatment Plant Discharge [page 372] Summary of Key Results ° The PICis highly susceptible to flooding, even without rainfall occurring onsite (25-50 yr upstream; 1 yr onsite). ° The model can be used to design onsite drainage improvements in detail, as well as infrastructure to prevent/mitigate inflow from the river floodplain. * It does not appear that the WWTP will have a significant effect on the river’s water quality, or on the downstream discharge. [page 373] Real-time hydroclimatic data assimilation TR NET 3 QUE MAPS. ue Re ve PA ei re NO fees ES SNS HOVAR 18Z 11 August 2005 (mm) ESS *__ Rainfall input from satellite information (TRMM + other satellites [TT MPA]) * _ Runoff generation °_ Hydraulic Routing o 1/8‘ degree o 50 deg. N-50 deg. S o 3-hrtime steps http:/flood.umd.edu/ [page 374] Pan the map Rainfall (1—-day accum.) [mm] 00Z070ct2014 Y 50N 40N [#] 3ON Zoom in 20N [ttt] 10N Zoom out EQ 10S Plot time series for an individual point (lat, lon): 20S (Tips: Zoom in to 40s o ess | 508 Tfoozn0c01 | 120W 9OW GOW 30W o 30E BUE QUE 120€ 190 180 | |12: {0707000014 | See time series | CS SE CO, 5 10 25 50 100 150 300 [mm] Plot different variable: | Plot | Surttime Endtime [anima ] a [page 375] Rainfall (Instantaneous) [mm/h] 06Z09Apr2014 06Z30Apr2014 5 45 ë Reported rainfall ; 4 atthe PIC: “8-11 35 inches in 2-3 hours” | | 2.5 | 2 1.5 e | | 0,5 | 0: den à 0,5 L Au 1SAPR 15APR T7APR 1S9APR 21APR 253APR 25APR 27APR 29APR [page 376] ,Rainfall (1-day accum.) [mm] 06Z09Apr2014 06Z30Apr2014 16 14 12 19 8 6 al qe en œ oo 2 —2 RER HSRPR 1APR TAPR HOAPR ZAR 2HPR 2SHPR 274R 204PR [page 377] ,Rainfal (3-day accum.) [mm] 06Z09Apr2014 06Z30Apr2014 16 14 J 12 19 8 6 4 GREEN Rs NN rsserssesssssesqusn : | —2 : DRPR (SR 1SAPR (ZAR TOÂPR ZtAPR 23PR 25APR 27APR 204PR [page 378] Rainfall (1-day accum.) [mm] 06Z010Oct2010 06Z040ct2014 160 140 € 120 k 100 9 " Î 80 i 5 : Ï : 60 a Si éi 4 Ô ( ss FT |e : 40 4 x a . ti _ 1. | Il do G |D (99 à pi " 6 EH $ 0 Ô (L. a : @ $ jl Lo 9 |. © # ET é ( P ° Q LA foe 5 ef LL % l Ÿ al -20 JAN JUL JAN JUL JAN JUL JAN JUL 2011 2012 2013 2014 [page 379] Rainfall (1-day accum.) [mm] 06Z010ct1980 06Z040ct2014 160 140 120 | 9 100 pl 80 ral 60 || 88 6x Go io [IN —20 1985 1990 1995 2000 2005 2019 [page 380] Moving forward * Conceive potential flood alleviation measures * Design simulations for updated/new PIC infrastructure ° Impacts on flooding and water quality ° Using the model to assemble and support economic analysis of options * Connect PIC model to basin scale modeling and IWRM plan for Trou du Nord watershed ° Ownership building and model transfer to in- country parties [page 381] LAC-AHD (Analytical Hydrography) Drainage Drainage Country Area: Sq. Area: Sq. Name ID Source Station River Latitude Longitude Km. (Gauge) Km. (AHD) Chile 35 UNESCO Corneche Rapel —33.98 —71.58 13,186 13,782 Colombia 51 UNESCO Pte Pusmeo Patia 1.7 —77.61 13,147 13,172 C] Ecuador 59 UNESCO D.J.Palmira Paute —2.55 —78.55 5,162 5,172 PES Le Lx EE nas RO à V2 Se HÉRRUR EC? so SJ NU DE ANS Un ce | ue PE Se SON AZENNE) . AU Pie Eu Rapel Patia IX FR y D re Lun. DES Leu ANS PEN Faute er TES ET > TS A OT PAS der D de, \Y RTE [page 382] Hydrologic Model (GWLF) 2 À Ki ty R+M,+08D Unsaturated Potential Evaporation ge Infitration D Runoff T +273 l Base Flow G, _ rs, Water Balance U,, =U, +R +M, —Q, —E, — P Su =S, + P —-G, -D, [page 383] D | / f 7 RE \ / Be Le” .. ë "- | € { 1 sd L + F] AM Pneu à. | se. | PO ) (e- fl ) Le *\. | Legend + Y se. ir À se à Lu Fn # f TA169 [ ; ri } A NS Gauges SAS : ae, ] x n Water Fe 7 d e à 7 Es a — —fus D FA 4 EM Med Forest DL r LA a \ CH { b Pasture/Hay \ } > FM Ro* crops : / \ : LL Soils f Ç B7TT271 a” } | = > [] Emergent Hemaceous Wetiands EE 7 PE vou | AR f Ts A | FA 2 . n : __ Le DS À | in \ \ F FS d eu: re | è ARRETE ANA 1 SE PAT SU NL NE 72 IRANCLSN LEE. l À sTins VA, Dons Ç ei CA Né ASS À ITU £ > L NS. CAIN SE D SI LS Nas SÈ À ANT 77 À } 5. an " { _ « À NES + 1 1f — sn ‘5 Ææ "4 LS 2 / { RE ER VE \ A 1 2085240900 À CLR RE & £ & Argentina / ù | "af F AT y BRTI | (TR, Nes sh c / = à ource: FAO NX + D Psp | R- ] / { Saun amesea D ns À | » | = à ÈT \ Le NC à si j [page 384] Network, grid and mesh based models ° “Network”: 1D, OK for design of urban drainage systems, evaluate existing infrastructure. ° “Grid”: 2D, OK for floodplain assessments, can be used for urban drainage using high resolution grids. ° “Mesh”: 2D, more flexible to handle existing urban infrastructure, can use smaller grids reducing significantly computer run times. [page 385] 2D Shallow Water Approximation Ci = L = |, [page 386] 2D Shallow Water Equations ôn OUH OVH 7 Ox F dy oU OU OU ON Tor ot L Ox L Oy Er 5 pH OV OV OV On Ty y pr 7, 5 US +! 0y ‘Oy pH ge U VU? + V? grvV VU? + V? Tor — H413 Ty — 143 ° Unknowns: n, U, V f(x, y, time) 88 [page 387] e em eme Len Len Lens 4710.00 | 4708.12 4710.00 4709.00 4707.36 | 4707.67 4706.30 a [en em le _ = À one | 4710.11 | 4709.64 | 4708.72 | 4708.01 | 4707.10 | 4705.92 | 4705.18 89 [page 388] re 7 ee ne. — EL \ Z\7 \7 CA SARL HA ve TER Es IX PERLE RER XE te /V RER Ke 90 [page 389] Simulated time: 0.00 À. Simulated time: 0.00 h. GPU wall time: 0.00 minutes CPU wall time: 0.00 minutes é =. NOEL DER - = 7 = [page 390] . out of 20(0.10 [hrs]) A 15.0 ft/s .0 ft/s Automatic À Google Ke à | à Earth D A: 2 + À | Le 4 Fa à [e] @ HERRERA [page 391] Max Depths. Sunny Day, 25 year event Î E = à en etes ns Dept [page 392] Max Velocities. Sunny Day, 25 year event [page 393] Max Depths. Sunny Day, 50 year event | CR : É 2 | J 4 [page 394] Max Velocities. Sunny Day, 50 year event | al ES >. {le | IA DE "TS ut, > -- 300 ‘ = nn . é . 1.20 [page 395] Max Depths. Sunny Day, 100 year event LL} À te RD: f n:! [page 396] Max Velocities. Sunny Day, 100 year event =. | 111 d: F TR "LP Ÿ M 4 . [page 397] Max Depths. Sunny Day, 200 year event we | ii es RU LT 3 LÉ | % M D de si ” [page 398] Max Velocities. Sunny Day, 200 year event =: . L M Velocity en, j ' [page 399] Vidx LEPINS. RdiNrall, L VEAr ÊVE ù “ Ë , + à AR 4 L PER [page 400] Max Velocities. Rainfall, 1 year event e CS FE IT YU: é [M : FR éme." pen | fe _ D. nn “7 (m/s) Il é Pine. | [page 401] (m) 1.00 0 30 40 0 1.60 || 120 || 80 En 40 |. 00 FA [page 402] IVIdX VEIOCITIES. KdINTall, 4 YEAr EVENT & 9 4 = | "4 ” 2 « À LA, re « a [page 403] IVIaX LEPINS. RdiNTall, > ÿYÉAr EVENT ” d AE > à > ê , h { 44" , | F-. Lo di | Q 1. L … . = [page 404] Max Velocities. Rainfall, 5 year event nn: { 4P ; Lœ | a Fi L ue TS ns ed { ÿ » F 2.70 \ nn. 1 # p” 240 C7 DL F4 0 &: = Re " n ee L , " [page 405] Max Deptns. Rainfall, 10 year event CU _, 82’ Ç 174 ee, ,. 8 | 1. * [page 406] Max Velocities. Rainfall, 10 year event 4 LT M P- a | | | 8 & : 1? D sg 5 en D. $ : à EEE 0.00 [page 407] Max Depths. Rainfall, 25 year event + d ÿ | | [page 408] Max Velocities. Rainfall, 25 year event À, d * . + a | e à F jus h Re 4 | = 4 : [page 409] Max Depths. Rainfall, 50 year event | LP. LERX, . [page 410] Max Velocities. Rainfall, 50 year event &. Ps a La | | 4 E [T & | | D : ss 10 |. 0 | » he ns Nr -* : sl Velocit ” HORS < : [page 411] Max Depths. Rainfall, 100 year event | FES . [page 412] Max Velocities. Rainfall, 100 year event mg © , Pan. , a à A _# | # | eau ed pr e. ——.. LR | » | : . . [page 413] : Max Depths. Raintall, 200 year event ” et ft ['e = 4 [page 414] Max Velocities. Rainfall, 200 year event 9 | £ %., (e + Off . M: 8 Dmpines .— L 4 = | + Velocity à _— +. cu J1 Fi J | 2.70 l ÿ 240