The spring discharge is difficult to simulate in Karst areas under human activity. An integrated model (MODFLOW-ANN) was developed by combining the merits of the numerical groundwater flow model - MODFLOW and the artificial neural network model (ANN). Based on the application in the Xiaonanhai spring catchment of a karst region, Henan Province, the principles and algorithms of the integrated model were discussed, and the results were compared between MODFLOW and MODFLOW-ANN. The coefficient of determination, relative error and correlation were 0.79、4.98% and 0.84 for MODFLOW, respectively, and 0.88、1.22% and 0.89 for MODFLOW-ANN, respectively. The results show that this integrated approach could take advantage of the groundwater numerical analyzing capacity of MODFLOW and the nonlinear approximation ability of ANN, thus precisely predicting the peaks and troughs of spring discharge. This model improved the predicting accuracy and was successfully applied to model the spring discharge dynamic. This study could provide reference and guidance for further exploitation of groundwater in the spring catchment of Karst areas.
