Accessing the Impact of Meteorological Variables on Machine Learning Flood Susceptibility Mapping

Machine learning (ML) algorithms have emerged as competent tools for identifying areas that are susceptible to flooding. The primary variables considered in most of these works include terrain models, lithology, river networks and land use. While several recent studies average annual rainfall and/or temperature, other meteorological information such snow accumulation short-term intense rain events may influence the hydrology area under investigation not been considered. Notably, Canada, inland flooding occurs during freshet, due melting an accumulated snowpack coupled with heavy rainfall. Therefore, this study impact climate along various hydro-geomorphological (HG) were tested determine their inclusion. Three tests run: only HG variables, addition temperature precipitation (HG-PT), inclusion six datasets (HG-8M) on five across Canada. In HG-PT, both selected important every area, while HG-8M a minimum three each area. added, many initial dropped from selection set. accuracy, F1, true skill Area Under Curve (AUC) marginally improved when data was added parallel random forest algorithm (parRF). When model is applied new data, estimated accuracy prediction higher HG-8M, indicating relevant, local improves result.