Deep Learning for Streamflow Regionalization for Ungauged Basins: Application of Long-Short-Term-Memory Cells in Semiarid Regions

Rainfall-runoff modeling in ungauged basins continues to be a great hydrological research challenge. A novel approach is the Long-Short-Term-Memory neural network (LSTM) from Deep Learning toolbox, which few works have addressed its use for rainfall-runoff regionalization. This work aims discuss application of LSTM as regional method against traditional (FFNN) and conceptual models practical framework with adverse conditions: reduced data availability, shallow soil catchments semiarid climate, monthly time step. For this, watersheds chosen were located on State Ceará, Northeast Brazil. streamflow regionalization, both FFNN better than model used benchmark, however, quite superior. The methods also showed ability aggregate process understanding different performance networks trained regionalization single catchments.