Objective Coal mining-induced disturbance leads to water redistribution in the vadose zones of mining areas. The sensitivity analysis of the factors influencing this redistribution holds critical significance for the reconstruction of eco-geological environments and the sustainable economic development in mining areas.

Methods Against the engineering background of the Yushen mining area characterized by wind-blown sandy terrain in northern Shaanxi Province, this study analyzed the primary factors influencing water redistribution in the vadose zone under coal mining disturbance. Accordingly, 18 COMSOL Multiphysics numerical models for water redistribution in the vadose zone under coal mining disturbance were constructed through orthogonal experiments. Using these models, the theoretical average saturation along three monitoring lines at burial depths of 6.2 m, 7.0 m, and 7.8 m was derived to recover the coal mining disturbance of the pre-mining stage. Through the range analysis and multiple linear regression analysis, this study clarified the sensitivity of the factors influencing water redistribution in the vadose zone under coal mining disturbance and further determined the weights of the primary influential factors using the analytic hierarchy process (AHP).

Results and Conclusions From the perspective of stratigraphic structure, hydrogeological conditions, and mining conditions, the primary factors influencing water redistribution in the vadose zone under coal mining disturbance included the ratio of the sandy loam thickness to the sandy clay thickness (a), initial porosity (p), initial number of fractures (l), the matrix permeability coefficient of the vadose zone (K₁), fracture permeability coefficient (K₂), and coal seam mining height (b). Under coal mining disturbance, the overall water redistribution in the vadose zone was primarily influenced by factors a and K₂, along with minor influence from factors l, b, and K₁. Based on the range analysis and multiple linear regression analysis of the orthogonal experimental results, the sensitivity of these factors decreased in the order of a, K₂, p, b, l, and K₁. Following the sensitivity ranking of these factors, the weights of the primary factors were determined at 38.25% (a), 25.04% (K₂), 15.96% (p), 10.06% (b), 6.41% (l), and 4.28% (K₁) using the AHP. The sensitivity ranking was consistent with the weight ranking, with both substantiating each other. The results of this study hold great significance for reconstructing the ecological security patterns and constructing green mines in ecologically vulnerable areas within western China.