Abstract: Vegetation change serves as a dual indicator of shifts in both climate and the ecological environment. Quantifying regional-scale vegetation changes and assessing the effectiveness of ecological restoration have become common scientific issues in the study of terrestrial ecosystems. Based on long-term normalized difference vegetation index (NDVI) data and climate, topography, and human activity factors, this study employed Theil-Sen Median trend analysis, Mann-Kendall test, temporal stability, and geographic detector methods to examine the dynamic variation characteristics of vegetation NDVI in China's first-level watersheds from 2000 to 2020. It also identified representative watersheds based on temporal stability and Investigated the factors driving the spatial variation of vegetation NDVI in these representative watersheds. The results indicate that: (1) Over time, both the vegetation NDVI values in China and the first-level watersheds show a significant increasing trend. Among them, the Yellow River Basin had the highest growth rate at 0.0055 a⁻¹, followed by the Pearl River Basin at 0.0033 a−1, and the Huai River Basin with the lowest growth rate of only 0.0011 a⁻¹. Spatially, the NDVI exhibited a decreasing trend from southeast to northwest, with the dominant vegetation change being improvement (accounting for 52.02%), and the flow of vegetation cover mainly directed towards higher grades. (2) Spearman rank correlation analysis showed that vegetation coverage in China has significant temporal stability (P<0.05), with the southeastern rivers identified as representative watersheds of the nine first-level watersheds in China. (3) The top three explanatory factors for vegetation NDVI in the southeastern rivers' watersheds are elevation > slope > land use type. Additionally, population density and nighttime lights have continuously increased since 2000. Climate and topographic factors are the main interactive controlling factors of NDVI changes, with synergistic effects observed between these factors, mainly in a dual-factor enhancement pattern. The findings are significant for understanding the regional vegetation growth patterns and for protecting regional ecological environments.