Streamflow Prediction in Ungauged Basins (PUB) Using an Ensemble Modelling Approach

Hydrological modelling is a widely used tool for different water resources management related purposes. In this regard, the availability of hydrological data can be very scarce in remote zones, which makes calibration of the model parameters unfeasible. Therefore, streamflow prediction in ungauged catchments is a challenging issue in hydrology. One possible approach to address this problem is by transferring parameters from a gauged (donor) basin to the ungauged (receptor) basin. In the present study, developed in 24 catchments in the northwest of Spain, we propose to apply a combination of a regression-based method and an ensemble modelling approach for streamflow prediction in ungauged catchments. Preliminary results showed that the method obtained values for the Nash-Sutcliffe coefficient greater than 0.5 in 79 % of the cases and determination coefficients greater than 0.67 for the 24 basins. However, those cases in which the proposed method fails should be further studied to improve the results in future work