Spatiotemporal Dynamics and Nonlinear Associations of Eco-Environmental Quality in Mining Areas Using MRSEI and an Object-Based XGBoost-SHAP Framework

Long-term mining activities can cause complex ecological and environmental problems, including mineral surface exposure, dust-related disturbance, and vegetation degradation, while conventional ecological indices may not fully characterize the distinctive ecological conditions of mining areas. This study developed a mining-area Modified Remote Sensing Ecological Index (MRSEI) by incorporating the Lithological Mineral Index (LMI), Normalized Dust Difference Index (NDDI), and Vegetation Health Index (VHI) into the Remote Sensing Ecological Index (RSEI) framework. Object-based analytical units were constructed using Landsat time-series data from 2000 to 2024 and G-means clustering, followed by the development of an Object-Based XGBoost-SHAP (OB-XGBoost-SHAP) framework to quantify the relative model contributions and nonlinear associations of explanatory factors. The results showed that, in the Daye mining area (DMA), MRSEI provided clearer differentiation of ecological quality among land-use types than RSEI and Mining-Specific Eco-Environment Index (MSEEI), with an overall gradient of forest land > cultivated land > construction land > mining land. Annual Principal Component Analysis (Annual-PCA) and Global Principal Component Analysis (Global-PCA) results showed high consistency (Pearson r = 0.985). Across the seven evaluation years, the mean MRSEI of the DMA increased from 0.5493 in 2000 to 0.6035 in 2024, representing an overall change of 9.87%. The object-based model consistently outperformed the regular-grid model under random splitting, while model performance decreased under spatially separated validation but retained moderate predictive capability. Particulate Matter (PM10), Digital Elevation Model (DEM), Temperature (TEM), and Land Use Intensity (LU) showed relatively high model contributions, with their relative importance varying across different periods. This study integrates mining-specific ecological characterization, object-based spatial analysis, and explanatory-factor contribution analysis, providing a methodological reference for ecological quality assessment and long-term change monitoring in mining areas.