ROYAL ACADEMY OF OVERSEAS SCIENCES First Young Researchers Overseas’ Day 16 December 2014, Brussels, Belgium ISSN: CODE HERE© 2014 The Authors. Published by KAOW-ARSOM Selection and peer-review under responsibility of KAOW-ARSOM Scientific Committee Risk assessment in the Caribbean: modeling a GIS-based Flood risk Tool for Jamaica Hanne Glas 1* , Maxine Jonckheere 2 , Philippe De Maeyer 2 , Greet Deruyter 1 1 Department of Engineering and Architecture, Ghent University, Ghent, 9000, Belgium 2 Department of Geography, Ghent University, Ghent, 9000, Belgium * Corresponding Author. Email: hanne.glas@ugent.be Abstract The Caribbean is known to be one of the most hazard-prone regions in the world. Although the growing intensity of these disasters increases the concern of decision makers, researchers have not yet succeeded in developing an accurate multi-hazard risk assessment tool to locate the high-risk areas. Many single-hazard risk analyses adequately estimate the risk of one hazard, but the complexity of the relation between the different types of hazards causes difficulties in developing one risk analysis to assess all hazards. This research aims to develop such a model for the Caribbean. In a first step, a flood risk tool was developed for Jamaica. After optimizing this tool, the methods will be used for other hazard types. In a final step, all single-hazard tools will be combined into one multi-hazard risk assessment. To generate the flood risk tool, a new methodology based on minimizing risks rather than building water defences, was used. In the Flemish Region in Belgium, this method is already used in a tool called LATIS, and has proven that using a risk-based methodology helps tremendously in finding the most cost- efficient measures to reduce risk. For Jamaica, the lack of data was and is a big concern. Since there was only minimal rainfall data available, flood hazard maps could not be generated. Therefore, a risk map could not yet be computed, only a vulnerability map and a damage map could. Furthermore, the available damage functions did not cover all elements at risk. Before regenerating the damage map, these functions will have to be reassessed. However, the vulnerability map that was produced shows promising results in indicating the high-risk areas, which are the most important factor in the decision making process. Further research will focus on the