Several key issues in the application of MODFLOW-CFP software to the numerical simulation of karst water systems

The MODFLOW-CFP software effectively addresses the coupling issue between pipe flow and groundwater seepage in porous media, as well as the transition from laminar to turbulent flow in pipe water movement. It provides a valuable numerical simulation platform for the quantitative study of karst conduit groundwater systems and has been applied in practical engineering by some researchers in recent years. However, due to limitations in survey methods, particularly for linear projects such as railways, highways, and water conservancy, the survey accuracy is relatively low. It is challenging to accurately characterize the structure of karst conduit systems in the presence of fractures and dissolution channels in karst aquifers, and the simulation results are often unsatisfactory. To better utilize and promote the application of this software in the quantitative evaluation and study of karst groundwater systems, and to avoid the blind use of new methods in practical applications and improve the simulation accuracy and prediction precision of numerical models for complex karst water systems, further development of quantitative hydrogeological research in karst geology is needed. Based on an analysis of the basic principles of MODFLOW-CFP simulation software and the authors' extensive experience in karst hydrogeology, this study conducts an in-depth analysis in various aspects. These include determining the model scope and boundary conditions, quantifying karst conduit structures, acquiring hydrogeological parameters, and selecting objective functions. Additionally, this study discusses key technical issues and possible solutions in the simulation process, taking the Daiye cave karst groundwater system in Yongshun county, Hunan Province, as a case.

The research results indicate as follows. ① Detailed investigations of karst hydrogeology, including groundwater tracing, drilling, and geophysical exploration, are essential for understanding the hydrogeological conditions, finely partitioning the karst water system, and establishing accurate conceptual hydrogeological models as the basis for numerical simulations. ② For the determination of the model scope, we should first consider selecting a complete groundwater system based on engineering requirements. In terms of the boundary conditions, we should then consider the changes in groundwater system equilibrium caused by the engineering activities, especially for soft boundaries such as watersheds. ③ Conducting groundwater tracing test, obtaining breakthrough curves of tracer concentration, and using the Qtracer2 model can effectively characterize the conduit structures and acquire the relevant parameters required by the CFP module. ④ Comprehensive and long-term monitoring data on rainfall, flow rate, water level, water chemistry, temperature, etc. are crucial to accurately obtain hydrogeological parameters, select appropriate objective functions, and improve the simulation accuracy of the model. ⑤ Currently, model identification mostly relies on the groundwater flow rate or the groundwater level as the sole objective function. Incorporating multiple conditions simultaneously as objective functions for model identification is less common but requires further research to improve the predictive accuracy of models.