Hydro-informer: a deep learning model for accurate water level and flood predictions

This study aims to develop an advanced deep learning model, Hydro-Informer, for accurate water level and flood predictions, emphasizing extreme event forecasting. Utilizing a comprehensive dataset from the Slovak Hydrometeorological Institute SHMI (2008-2020), which includes precipitation, water level, and discharge data, the model was trained using a ladder technique with a custom loss function to enhance focus on extreme values.

The deep learning model developed in this study demonstrates significant capabilities in predicting water levels, particularly in forecasting extreme values. The model's performance metrics-such as MSE of 18.07, RMSE of 4.25, MAE of 1.82, MAPE of 1.11%, and R2 of 0.8785-highlight its robustness and accuracy. These results indicate that the model can reliably predict both regular and extreme water levels, which is crucial for flood management and early warning systems.