A semi-distributed groundwater recharge model for estimating water-table and water-balance variables

A semi-distributed groundwater recharge model is presented, which estimates water-table fluctuation and water-balance variables. The model is expressed by the water-balance concept linking atmospheric and hydrogeological parameters to different water uses (industrial, agricultural, domestic, etc). It was calibrated and validated using 5 years of data collected in the Dogo Plain in Japan. A 3-year dataset, from 2000 to 2002, was used in the calibration, while a 2-year dataset, from 2003 to 2004, was used for the validation. Calibration of the model was achieved by the shuffled complex evolution automatic optimization of model parameters to match simulated results with measured water-table depth. Square roots of relative error (R (2)) are 0.88 and 0.90 for calibration and validation processes, respectively. Monthly evolution of water storage change was then estimated and the water-table drawdown in different pumping scenarios was simulated. Finally, the groundwater-pumping amount planned by the government for future sustainable groundwater utilization was evaluated. The government-planned groundwater-pumping amount is feasible in most regions while the midstream region should be paid more attention. This study offers a scientific basis to control and prevent depletion of groundwater resources.