Search

Climate change and urbanization have increased the frequency of floods worldwide, resulting in substantial casualties and property loss. Accurate flood forecasting can offer governments early warnings about impending flood disasters, giving them a chance to evacuate and save lives. Deep learning is used in flood forecasting to improve the timeliness and accuracy of flood water level predictions. While various deep learning models similar to Long Short-Term Memory (LSTM) have achieved notable results, they have complex structures with low computational efficiency, and often lack generalizability and stability. This study applies a spatiotemporal Attention Gated Recurrent Unit (STA-GRU) model for flood prediction to increase the models' computing efficiency. Another salient feature of our methodology is the incorporation of lag time during data preprocessing before the training of the model. Notably, for 12-h forecasting, the STA-GRU model's R-squared (R2) value increased from 0.8125 to 0.9215. Concurrently, the model manifested reduced root mean squared error (RMSE) and mean absolute error (MAE) metrics. For a more extended 24-h forecasting, the R2 value of the STA-GRU model improved from 0.6181 to 0.7283, accompanied by diminishing RMSE and MAE values. Seven typical deep learning models—the LSTM, the Convolutional Neural Networks LSTM (CNNLSTM), the Convolutional LSTM (ConvLSTM), the spatiotemporal Attention Long Short-Term Memory (STA-LSTM), the GRU, the Convolutional Neural Networks GRU (CNNGRU), and the STA-GRU—are compared for water level prediction. Comparative analysis delineated that the use of the STA-GRU model and the application of the lag time pre-processing method significantly improved the reliability and accuracy of flood forecasting. [ABSTRACT FROM AUTHOR]

Mitigating the harmful effects of flooding is essential in the current climate change and urban growth scenario, where these natural phenomena are expected to occur more often. Addressing this problem, we present a novel graph-based forecasting model for predicting urban flooding and showcase its application in the Bow River in the City of Calgary, Alberta, Canada. The proposed model is based on graph theory and deep learning paradigms and was used to forecast flooding occurrences up to 24 h ahead. Our new approach developed the SAGE algorithm, a hybrid learning and planning approach that accurately predicted floodings for various analyzed forecasting horizons, increasing the forecasting performance by up to 44% compared to the persistence model. The SAGE model also returned superior or competitive results compared to other models in the literature. The superiority of the SAGE model was highlighted for longer forecast horizons of 24 h ahead, where it reached a maximum improvement of 79%, suggesting its superiority in capturing spatiotemporal information in the dataset. These results indicate that the SAGE approach is a cutting-edge tool for flooding forecasting, and it could be used to develop early flood warning systems to reduce potential flooding impacts. [ABSTRACT FROM AUTHOR]