Liu, Junrong ， Zhang, Qiang ， Zhang, Zhuoyu
2025-12-31 GEOCARTO INTERNATIONAL 2025 40(卷), 1(期), (null页)
Mining activities drive socio-economic growth but cause severe disturbance to vegetation and ecological stability, especially in arid and semi-arid regions with weak natural resilience. This study applied an normalized difference vegetation index (NDVI)-dead fuel index (DFI) pixel trichotomy model - comprising photosynthetic vegetation (PV), non-photosynthetic vegetation (NPV) and bare soil (BS) - to the Bulian'gou mining area of Inner Mongolia (2007-2024) using multi-temporal Landsat data validated by UAV imagery. The model outperformed the traditional NDVI-based dichotomy, achieving R-2 values of 0.72, 0.60 and 0.76 for PV, NPV and BS, respectively, with an overall RMSE < 0.18. The NDVI increased steadily after 2012 while DFI stabilized, indicating improved vegetation structure and soil stability under the 'mining-while-restoring' strategy. The PV expanded from fragmented southern patches to continuous central-southern zones, and the NPV shifted toward higher coverage. The NDVI-DFI pixel trichotomy model thus offers a robust, scalable framework for monitoring vegetation dynamics and assessing ecological restoration in mining-disturbed arid landscapes.