Flood susceptibility assessment using machine learning approach in the Mohana-Khutiya River of Nepal

Abstract

Nepal, known for its challenging topography and fragile geology is confronted with the constant threat of floods leading to substantial socio-economic losses annually. However, the country's efforts in planning and managing flood risks remain insufficient, especially in the vulnerable Mohana-Khutiya River. Therefore, this study focused on the Mohana-Khutiya River and utilizes the Maximum Entropy (MaxEnt) model to comprehensively map flood susceptibility and fill crucial gaps in flood risk assessments. This study employed a combination of 10 geospatial environmental layers and field-based past flood inventory to implement the MaxEnt machine learning model for flood susceptibility modeling. The available past flood data were divided into two sets, with 75% allocated for model construction and the remaining 25% for model validation. This study demonstrated that the proximity of the river had a significant impact (33.1%) on the occurrence of the flood. Surprisingly, the amount of annual precipitation throughout the year exhibited no detectable contribution to the flood event in the study site. About 4.9% area came under the high flood susceptible zone followed by 12.75 % in the moderate zone and 82.34% in the low-risk zone. The model exhibited excellent performance with an Area Under Curve (AUC) value of 0.935 and a low standard deviation of 0.018, indicating accurate predictions and consistent precision. These results highlight the model's reliability and its significance for developing disaster management policy by local government in the study site. Future research should refine the MaxEnt model by including more variables, validating against observed flood events, and exploring integration with other flood modeling approaches.