Vegetation–Climate–Topography coupling in a cold temperate region: Evidence from the Genhe Basin (2000–2019)

Vegetation in cold temperate regions is shaped by a mix of climate pressures and terrain influences, often in uneven and place-specific ways. This study explores how vegetation, climate, and topography interact across the Genhe Basin from 2000 to 2019, using MODIS NDVI data, both station-based and gridded temperature records, and high-resolution terrain metrics. Warming was most evident in spring and winter, while precipitation showed a moderate upward trend. Vegetation greenness increased slightly (0.014 per decade, p < 0.05) but significantly, mainly in high-elevation forested areas. In the warm southern lowlands, NDVI stagnated or slightly declined despite surface warming, especially in the agro-pastoral belt where increasing heat, moisture limitation, and land management interact. Topographic analysis further shows that greening responses are enhanced at mid-to-high elevations and on 5°–15° slopes with higher moisture convergence, whereas the low, warm agro-pastoral belt remains comparatively unresponsive. Pixel-wise Pearson and partial correlations indicate that precipitation exerts the dominant independent control on interannual NDVI variability, while localized negative NDVI–LST relationships persist in warmer, drier lowlands. Block-wise variance partitioning attributes a larger unique share of NDVI trend variance to terrain (Δ R 2 =0.066) than to climate (Δ R 2 =0.022), indicating that terrain explains roughly three times more of the spatial differences in vegetation change than climate alone. Rather than a uniform trend, vegetation responses reflected local climatic stress and terrain-mediated buffering, pointing to a need for spatially nuanced understanding of ecosystem dynamics under climate change. • Winter–spring warming is strongest; summer–autumn trends are weak. • Modest greening; southern NDVI stagnation is not driven by LUCC. • Precipitation dominates NDVI variability; terrain explains more trend variance.