Abstract:Accurate prediction of regional terrestrial water storage change (TWSA) is of great significance for water resources planning and management, and early warning of extreme climate disasters. Aiming at the problem that the conventional methods on prediction of TWSA time series are difficult to be accurate, the six typical regions are selected in China as examples, including the upper reaches of the Yangtze River (UYR), the southwest region (SWR), the Liaohe River Basin (LRB), the North China Plain (NCP), the Qinghai-Tibet Plateau (QTP), and the Pearl River Basin (PRB). The mascon product from GRACE/GRACE-FO provided by CSR is used to extract TWSA time series in six typical areas. The improved Back Propagation (BP) neural network, Long Short-Term Memory (LSTM) neural network and the latest Bidirectional LSTM (BiLSTM-attention) neural network model based on attention mechanism are proposed to predict and analyze the regional TWSA. In the experiment, the selection of the optimal model parameters such as the number of hidden layer nodes and the number of hidden units of the neural network model is tested and analyzed in detail. Meanwhile, the model prediction results are compared with the traditional least squares method and random forest (RF) prediction method. The root mean square error (RMSE), determination coefficient (R 2 ), Nash–Sutcliffe efficiency coefficient (NSE) and mean absolute percentage error (MAPE) were used to evaluate the accuracy of the predicted results. The results show that the improved BP, LSTM and Bi-LSTM-attention neural network models all achieve higher prediction accuracy in UYR and SWR areas. RMSE is less than 2.641 cm, R 2 is as high as 0.8 or more, NSE is above 0.6, and MAPE is within 0.1. Compared with the least square method, the RMSE of the predicted results from three neural network decreased by 0.998 cm, 0.700 cm and 0.7563 on average, and the R 2 increased by 81.75%, 69.89% and 72% on average. Compared with RFML method, the RMSE from three neural network is reduced by 0.601 cm, 0.316 cm and 0.360, and R 2 is increased by 38.20%, 24.60% and 27.06% on average. NSE and RMSE are improved to varying degrees in the above regions. It shows that the improved BP, LSTM and BiLSTM-attention model used can effectively predict TWSA. The research methods and results in this paper can provide important reference for the rational utilization of regional water resources and disaster risk assessment.

Filling the gap between GRACE- and GRACE-FO-derived terrestrial water storage anomalies with Bayesian convolutional neural networks

Improving prediction of the terrestrial water storage anomalies during the GRACE and GRACE-FO gap with Bayesian convolutional neural networks

Comment on Hess-2022-168

Bayesian convolutional neural networks for predicting the terrestrial water storage anomalies during GRACE and GRACE-FO gap

Reconstructing the data gap between GRACE and GRACE follow-on at the basin scale using artificial neural network

Bridging the gap between GRACE and GRACE-FO missions with deep learning aided water storage simulations

Bridging the gap between GRACE and GRACE-FO using a hydrological model

Combining Physically-Based Modeling and Deep Learning for Fusing GRACE Satellite Data: Can We Learn from Mismatch?

Filling GRACE data gap using an innovative transformer-based deep learning approach

Analysis of gap filling techniques for GRACE/GRACE-FO terrestrial water storage anomalies in Canada

Simulation and Driving Factor Analysis of Satellite-Observed Terrestrial Water Storage Anomaly in the Pearl River Basin Using Deep Learning

Bridging the Temporal Gaps in GRACE/GRACE–FO Terrestrial Water Storage Anomalies over the Major Indian River Basins Using Deep Learning

The changes prediction on terrestrial water storage in typical regions of China based on neural networks and satellite gravity data

Hydrological Droughts of 2017–2018 Explained by the Bayesian Reconstruction of GRACE(‐FO) Fields

A Fast Generative Adversarial Network Combined With Transformer for Downscaling GRACE Terrestrial Water Storage Data in Southwestern China

A data-driven approach to generate past GRACE-like terrestrial water storage solution by calibrating the land surface model simulations

Bridging the Terrestrial Water Storage Anomalies between the GRACE/GRACE-FO Gap Using BEAST + GMDH Algorithm

The Reconstruction and Extension of Terrestrial Water Storage Based on a Combined Prediction Model

Improving the accuracy of reconstructed terrestrial water storage during the GRACE/GRACE-FO gap based on a new variable filter reconstruction model

Analysis of watershed terrestrial water storage anomalies by Bi-LSTM with X-11 time series prediction combined model

Optimal feature selection for improved ML based reconstruction of Global Terrestrial Water Storage Anomalies