Using an Improved SWAT Model to Simulate Karst Sinkholes: A Case Study in Southwest China

Abstract

Hydrological simulation of the karst area is significant for assessing water resources accurately and exploring the relationship in the hydrologic cycle. However, the existence of sinkholes causes the spatial heterogeneity of aquifers and changes the distribution of surface water as well as groundwater, which makes the traditional hydrogeological model difficult to quantitatively characterize the hydrological processes of the sinkhole. Hence, improving the hydrological model for the karst area is a necessary direction at present. The soil and water assessment tool (SWAT) is one of the most widely used semi-distributed hydrological models right now in the world. In this study, we focused on the upper course of the South Panjiang River and used the pond module of the SWAT model to simulate karst sinkholes, modifying the source code to realize the rapid response to the recharge in karst sinkholes. After the improvement, the surface runoff, especially the peak value of the Xiqiao Hydrological Station at the outlet, has been reduced, while the baseflow of modified subbasins has been increased and the water yield is under a state of water balance. In addition, the model evaluation factor R2 was strengthened from 0.76 to 0.83 and NSE was strengthened from 0.66 to 0.79 of the Xiqiao Hydrological Station during the validation period. The improved model was used to analyze the spatial distribution of hydrological components. Also, it was found there are spatial relations between runoff modulus–slope and baseflow–surface runoff–land use types. The analysis demonstrated that the improved SWAT model could effectively change the hydrological components and simulate the rapid replenishment of karst sinkholes.