Abstract: The Shendong mining area, located in the transition zone between the Loess Plateau hilly and gully region and the Mu Us Sandy Land, is characterized by a semi-arid continental monsoon climate and a fragile native ecological environment. High-intensity coal mining activities have further exacerbated regional vegetation degradation and habitat destruction. Clarifying the spatiotemporal dynamics, response patterns, and driving mechanisms of vegetation coverage in this region is a prerequisite for ecological restoration and sustainable management. Systematic revelation of the spatiotemporal differentiation characteristics of fractional vegetation coverage (FVC) in the Shendong mining area, its response patterns to major environmental factors, and multi-factor synergistic mechanisms can provide a scientific basis for ecological restoration in semi-arid coal mining areas. Based on multi-source remote sensing data including Landsat−9OLI, ERA5−Land, CHIRPS, and SMAP acquired via the Google Earth Engine (GEE) platform from 2021 to 2025, the Pixel Dichotomy Model, Geodetector, and Coefficient of Variation method were adopted to analyze the spatiotemporal patterns, seasonal fluctuation, driving factor intensity, and multi-factor interaction effects of FVC. The results indicate that spatially, the FVC exhibits a heterogeneous pattern of “high in the southeast and low in the northwest”, with mining activities creating a “high−low” mosaic distribution; the overall vegetation ecological environment showed an improvement trend from 2021 to 2025. Seasonally, vegetation coverage presents characteristics of “winter low coverage dominance, summer medium-high coverage prominence, and spring-autumn transition”, with a significant increase in the proportion of medium-high coverage in the summer of 2025. In terms of variability, low fluctuation zones accounted for 63.1%, showing a significant spatial difference of “low fluctuation in underground mining areas and high fluctuation in open-pit mining areas”. Regarding driving mechanisms, single-factor explanatory power shows significant seasonal differentiation; soil moisture had the strongest explanatory power during the growing season (spring and summer) with mean q-values of 0.46–0.64, superior to precipitation, while temperature was the dominant factor in winter with a mean q-value of 0.39. Multi-factor interactions all exhibited enhancement types; coupling of soil moisture and precipitation dominated in the growing season, while terrain and temperature showed nonlinear enhanced interactions in winter. The results clarify the dominant driving mechanisms of FVC changes in different seasons, providing a scientific basis for implementing seasonal, differentiated ecological restoration strategies and the coordinated development of resource exploitation and ecological protection in semi-arid mining areas.