Prediction of terrestrial water storage changes by using GRACE data over Nile river basin

This research involved training two deep learning prediction models: Long Short-Term Memory (LSTM) and Dipper Throated Optimizations Fitness Grey Wolf-LSTM (DTOFGW-LSTM), utilizing data obtained from remote sensing to reconstruct and predict the Terrestrial Water Storage Changes (TWSC) over Nile River Basin (NRB). We evaluated factors including Terrestrial Water Storage Changes (TWSC) and Groundwater Storage Changes (GWSC), identified through the Gravity Recovery and Climate Experiment (GRACE) and GRACE-Follow-On (GRACE-FO), alongside precipitation data collected by the Global Precipitation Climate Change Program (GPCP) to analyze the patterns of change within the research area. We utilized the LSTM and DTOFGW-LSTM algorithms to rebuild the TWSC and GWSC from 2018 to 2024. We utilized the precise model to forecast the GRACE gap from 2017 to 2018 and the TWSC from 2024 to 2030. The findings demonstrated the superiority of the suggested model (DTOFGW-LSTM) with a root mean square error (RMSE) of 0.51, a coefficient of determination (R²) of 0.99, and a mean absolute percentage error (MAPE) of 0.21.