Author(s): Joseph ADERMUS, Nyanokna Gonomy, Olivier Carlier D’odeigne, Yves Zech, Yves Zech
Keywords: Lack of data; Flood mitigation; Flood modelling; Drone survey; Analytic Hierarchy Process;
Abstract: Haiti is one of the poorest countries in the world, and prone to worst natural hazards. Among these hazards, inundations and mudflows, frequent in this hurricane zone with soils destroyed by intense deforestation, recurrently threaten people and goods. Technical approaches of event forecasting, flood modelling, risk mapping and alert organization, that are classical in developed countries become challenging in Haiti, where data are missing or incomplete and public administration is often ineffective and without adequate means. Moreover, even in case of improved technical tools, a large part of the difficulty relies in appropriation of the problem by the local population, who is often fatalistic. The paper describes a comprehensive process of technical study in link with sociological approach in a watershed severely affected by Hurricane Matthew in 2016: the watershed of Cavaillon River in the Southern Province of Haiti. The lack of hydrological data was circumvented by the use of conceptual model ATHYS and a smart use of the scarce data acquired immediately after the hurricane. A drone driven survey that appeared to be efficient, while cheap, addressed the lack of topographical and bathymetric information. Regarding the appropriation of the problem by the population, a vulnerability map was progressively built, using an Analytic Hierarchy Process and its transposition in terms of risk confronted with the perception by the local population.