Transformer neural networks for interpretable flood forecasting

Floods are one of the most devastating natural hazards, causing several deaths and conspicuous damages all over the world. In this work, we explore the applicability of the Transformer neural network to the task of flood forecasting. Our goal consists in predicting the water level of a river one day ahead, by using the past water levels of its upstream branches as predictors. The methodology was validated on the severe flood that affected Southeast Europe in May 2014. The results show that the Transformer outperforms recurrent neural networks by more than 4% in terms of the Root Mean Squared Error (RMSE) and 7% in terms of the Mean Absolute Error (MAE). Furthermore, the Transformer requires lower computational costs with respect to recurrent networks. The forecasting errors obtained are considered acceptable according to the domain standards, demonstrating the applicability of the Transformer to the task of flood forecasting. • As far as we know, this paper represents the very first work in the literature in which Transformers are applied to the task of flood forecasting. • Transformer outperforms the state-of-the-art Recurrent Neural Network (i.e. more than 4% in terms of RMSE and 1.7% in terms of the MAE. • Transformer requires lower computational costs with respect to Recurrent Neural Networks for both training and inference. • The attention mechanism adds interpretability to flood forecasting by focusing on the most critical upstream branches of the river.