Leveraging machine learning and open-source spatial datasets to enhance flood susceptibility mapping in transboundary river basin

Floods pose devastating effects on the resiliency of human and natural systems. flood risk management challenges are typically complicated in the transboundary river basin due to conflicting objectives between multiple countries, lack of systematic approaches to data monitoring and sharing, and limited collaboration in developing a unified system for hazard prediction and communication. An open-source, low-cost modeling framework that integrates open-source data and models can help improve our understanding of flood susceptibility and inform the design of equitable risk management strategies. This study integrates open-source datasets and machine -learning techniques to quantify flood susceptibility across the data-scare transboundary basin. The analysis focuses on the transboundary Gandak River Basin, spanning China, Nepal, and India, where damaging and recurring floods pose serious concern. flood susceptibility is assessed using four widely used machine learning techniques: Long-Short-Term-Memory, Random Forest, Artificial Neural Network, and Support Vector Machine. Our results exhibit the improved performance of Artificial Neural Network and Support Vector Machine in predicting flood susceptibility maps, revealing higher vulnerability in the southern plains. This study demonstrates that remote sensing and machine learning can improve flood prediction, hazard mapping, and susceptibility analysis in a data-scare environment.