A top-down deep learning model for predicting spatiotemporal dynamics of groundwater recharge

Xin Huang, Lei Gao, Nan Zhang, Russell S. Crosbie, Long Ye, Jian Liu, Zhaoxia Guo, Qinghua Meng, Guobin Fu, Brett A. Bryan

Abstract: Highlights•We proposed a top-down deep learning model (s-LSTM) for predicting groundwater recharge.•We compared s-LSTM with three bottom-up models in predicting spatiotemporal dynamics.•s-LSTM achieved the best prediction performance and required the shortest training time.•s-LSTM enables identification of regional-scale influential explanatory variables.•Extraction, wet-days, seasonal temperature, rainfall, and evaporation were influential.