Background Accurate prediction of water inflow in coal mines is essential for ensuring safe mining operations. Three-dimensional (3D) groundwater numerical models can be used to characterize hydrogeological conditions in mining areas and simulate the dynamic processes of water inflow, providing a theoretical basis for scientifically predicting mining-induced water inflow. However, there is a lack of a clear understanding of the impacts of the boundary conditions with surface water-groundwater interactions and the heterogeneity of aquifer permeability coefficients (K) on the simulation and prediction of water inflow during coal mining.

Methods This study developed a 3D transient groundwater numerical model for a typical coal mine. Through multi-scenario simulation experiments, this study quantitatively analyzed the impacts of rainfall infiltration recharge, infiltration process in the unsaturated zone, and river boundary configurations on the simulations of water inflow. Employing geostatistical methods, this study established 3D heterogeneous permeability coefficient fields to explore the impact of parameter heterogeneity.

Results and Conclusions The simulations using the initial model revealed that the average water inflow in the mining area decreased by 25.48% in the case where only the heterogeneity of rainfall infiltration recharge was considered. When the impact of vertical kinematic wave infiltration in the unsaturated zone was incorporated, the average water inflow increased by 18.68% and the water inflow was more prone to undergo periodic fluctuations compared to the initial model. Assuming that rivers were connected to the aquiclude via a tectonic fracture zone, the average water inflow in the mining face rose by 7.51% compared to the initial model, showing an accelerated growth trend. An increase in K heterogeneity corresponded to rapidly increasing water inflow. Meanwhile, the ratio of water inflow in the mining face to the average water inflow in the mining area increased to 85%-88% gradually, resulting in concentrated water inflow in the mining area. However, extreme K heterogeneity led to dispersed water inflow. Compared to the homogeneous generalized parameters, the random, heterogeneous K distribution was associated with more gentle growth trends in water inflow. Among all scenarios, the scenario where river channels lay the phreatic aquifer manifested a minimal impact on water inflow. In contrast, significant water inflow deviations from the initial model were observed under scenarios such as river channels passing through a locally fractured aquiclude, low-heterogeneity parameters, and vertical infiltration from the unsaturated zone. This study provides a critical reference for accurately simulating groundwater dynamics under complex mining conditions.