Basin characteristics had limited impacts on the performance difference between the LSTM model and the SIMHYD model. The results show that although the LSTM model significantly outperforms the SIMHYD model in the calibration period, it has significant performance degradation in the validation period. This study compared the flood simulation capabilities of the SIMHYD hydrologic model and the LSTM machine learning model in 232 basins with different climate conditions. Few studies have compared the flood simulation capabilities of machine learning models and hydrologic models. Machine learning models have been widely used for flood simulation. Nevertheless, without further model advancements, the presented models only provide robust discharge predictions for small and medium magnitude floods in low altitude catchments with warm temperate climate. We conclude that random forest provides a low-cost and, yet, competitive alternative to conventional rainfall-runoff models in large-scale flood discharge simulation. Relating catchment characteristics to model skill, we found that primarily climatic conditions and elevation affect the flood simulation capability. However, both models exhibit inaccuracies for higher flood events. Our analysis showed that random forest is competitive to hydromad in the simulation of low and medium flood magnitudes. We comparatively evaluate the predictive performance of random forest against the conceptual hydrological modeling package ‘hydromad’ and assess the influence of catchment characteristics on model performance. In this study, we simulate flood event and peak discharge on a daily time scale for 95 study basins in Canada and the USA. Yet, the applicability of random forest to flood discharge simulation requires further exploration, especially with respect to heterogeneous catchments and daily temporal resolution. Due to low setup and operation cost, random forest could represent an alternative approach to physical and conceptual hydrological models for large-scale hazard assessment in multiple catchments. The machine learning algorithm ‘random forest’ has been applied in many areas of water resources research including discharge simulation.
0 kommentar(er)
