Distributed Hydrological Model Based on Machine Learning Algorithm: Assessment of Climate Change Impact on Floods

Rapid population growth, economic development, land-use modifications, and climate change are the major driving forces of growing hydrological disasters like floods water stress. Reliable flood modelling is challenging due to spatiotemporal changes in precipitation intensity, duration frequency, heterogeneity temperature rise changes. high-resolution data distributed model can solve problem. This study aims develop a using Machine Learning (ML) algorithms simulate streamflow extremes from satellite-based data. Four widely used bias correction methods were compared select best method for downscaling coupled intercomparison project (CMIP6) global (GCMs) simulations. A novel ML-based was developed runoff corrected satellite rainfall Finally, future downscaled GCM projected climate. The Johor River Basin (JRB) Malaysia considered as case area. ML showed Nash–Sutcliffe efficiency (NSE) values 0.96 0.78 Root Mean Square Error (RMSE) 4.01 5.64 during calibration validation. simulated flow analysis that river discharge would increase near (2020–2059) far (2060–2099) different Shared Socioeconomic Pathways (SSPs). largest be SSP-585. extreme indices, such Total Rainfall above 95th Percentile (R95TOT), 99th (R99TOT), One day Max (R × 1day), Five-day 5day), Intensity (RI), 5% SSP-119 37% SSP-585 base period. results socio-economic development cause an frequency extremes, causing larger events.