Data sources of Unveiling nonlinear vegetation dynamics and driving mechanisms in the karst region of Southwest China using BFAST and XGBoost-SHAP

The NDVI data utilized in this research were sourced from the MOD13A1 product, provided by the National Aeronautics and Space Administration (NASA) (http://glovis.usgs.gov/). This dataset offers a spatial resolution of 250 meters and covers the period from 2000 to 2021, with data collected every 16 days.Vegetation dynamic evolution is subject to multifactorial synergistic effects. The driver framework of this study integrates seven-dimensional spatial variables encompassing topographic, climate, soil, rivers, and anthropogenic disturbances (Table 1; Figure 2). Temperature, precipitation, evapotranspiration, soil moisture were obtained from the 1 km resolution raster products were from the National Tibetan Plateau Data Center (https://data.tpdc.ac.cn). Topographic analysis utilized the ASTER GDEM v3 digital elevation model (30 m resolution, https://www.gscloud.cn), and extract the terrain slope from it. Nighttime light and GDP datasets originated from the Resource and Environment Science and Data Center (https://www.resdc.cn/). Rivers, residences, and road data were sourced from National Fundamental Geographic Database of the National Geographic Information Resource Directory Service System (https://webmap.cn/). Population density data were obtained from open-access spatial demographic statistics and research (https://www.worldpop.org/).