Urban flood susceptibility modelling using AHP and GIS approach: case of the Mfoundi watershed at Yaoundé in the South-Cameroon plateau

Abstract

Floods are considered as the natural hazards that affect the world's major metropolises the most. Thus, the present study aimed at evaluating the sensitivity to flood risks of the Mfoundi watershed (96.5 km2) located in the heart of the Cameroonian political capital in a tropical humid forest zone, more precisely in the South Cameroon plateau. The methodological approach adopted was to identify the factors that most favor the risk of flooding in the area from intense literature review and field investigations; the analysis of these factors and the calculation of the Flood Harzard Index (FHI) using the Analytical Hierarchy Process (AHP) approach coupled with the Geographical Information System (GIS) environment. The results reveal that among the ten parameters of the natural environment (elevation, drainage density, rainfall, slope, distance from the river, topographic humidity, hydraulic conductivity, groundwater level, geology and land cover) selected, the land cover, elevation and the geology are the factors that most influences the flooding phenomenon in the area. The value of Flood Hazard Index (FHI) varied from 4.16 to 9.16, the higher the value, the more sensitive the area is to the risk of flooding. Five main classes of flood susceptibility are highlighted: very low, low, moderate, high and very high, representing 9.50, 26, 23, 22 and 19.5%, respectively of the study area. To validate the efficiency of the obtained flood susceptibility map, the adopted Area Under the Curve (AUC) method shows a very good accuracy (0.84 or 84%). The results of this study constitute a basic tool for decision-making for environmental management by public authorities and decentralised territorial authorities with territorial jurisdiction.