GIS-Based Multi-Criteria Analytical Hierarchy Process Modelling for Urban Flood Risk Analysis and Assessment, Accra Metropolis

Abstract A series of devastating flood events within the Accra Metropolis over the last decades have demonstrated the urgent need for urban resilience. The present work aimed to develop a GIS-based model to analyse and assess floods in the Accra Metropolis. The framework is based on the Multi-Criteria Analytical Hierarchy Process within the GIS platform. Relevant thematic layers including LULC, Elevation, Slope, Soil, and Drainage density were generated within the GIS environment. In the context of objective weight assignments, the AHP algorithm was successfully applied to create flood hazard indices, and then the urban flood risk maps were constructed for 2007, 2010, 2015 and 2020. The findings revealed that areas under high and very high flood risk have increased from 55.163–60.043% over the last decade and a half. A flood inventory map was produced by randomly mapping 256 flood test locations to determine the model's performance. These locations consisted of 128 flooded and 128 non-flooded points extracted from historical flood data, ground truth data and high-resolution satellite imagery. The computed Receiver Operating Characteristic (ROC) curve showed an Area Under the Curve (AUC) value of 0.916, indicating an excellent correlation between the analysed flood risk areas and the ground truth data. Findings from the study will assist decision-makers in formulating medium-long term mitigation measures to reduce flood-related damages and employ proper future land-use planning.