Soil erosion susceptibility mapping for current and 2100 climate conditions using evidential belief function and frequency ratio

Abstract

Soil erosion is a global geological hazard which can be mitigated through better future land-use planning. In the current research, a Dempster-Shafer-based evidential belief function (EBF) and frequency ratio (FR) were used to map the soil erosion susceptible areas and their outcomes were compared subsequently. These methods were selected due to their efficiency and popularity in natural hazard studies. Moreover, the application of EBF is poorly examined in this area of research. Nine conditioning factors belonging to the current time, and rainfall intensity for the two time periods of current time and 2100 based on the A2 scenario CSIRO global climate model, were utilized in this research. The main aim was to estimate and compare the soil erosion hazards at Southern Luzon in the Philippines under two time periods, current time and 2100. This region has been highly affected by erosion and has not received much attention in the past. The area under the curve outcomes indicated that the FR model produced 70.6% prediction rate, while EBF showed superior prediction accuracy with a rate of 83.1%. The results also project that soil erosion hazards in the Philippines will increase due to changes in rainfall patterns by 2100.