Improving streamflow prediction for hydrological drought forecasting through a machine learning approach : A case study of Rhine River, the Netherlands

Drought forecasting is crucial for effectively managing water resources. However, streamflow forecasting remains challenging in regions like the Rhine Delta in the Netherlands, where forecasting accuracy and lead time are still limited. To address these issues, we present a machine learning (ML) approach based on the Long Short-Term Memory (LSTM) algorithm. We examine various combinations of univariate and multivariate configurations to improve streamflow forecasts for the Rhine River, focusing on sub-seasonal to seasonal lead times. In various experiments, we explore stacked LSTM setups with... Mehr ...