Quantifying nonlinear driving forces of land surface temperature dynamics in the Tarim River Basin under climate warming
摘要
Land surface temperature (LST) is crucial for assessing ecological responses to climate change in arid basins. Using MODIS daytime LST data from 2001 to 2024, the spatiotemporal trends in China's Tarim River Basin were analyzed via Sen's slope and Mann–Kendall tests. Twelve potential driving factors including vegetation, hydroclimatic, topographic, and anthropogenic variables were evaluated using Pearson correlation analysis and the Geodetector method. The results show that: (1) annual LST displays seasonal fluctuations but no long-term trend. LST variability is higher in high alpine regions and lower at low elevations. High LST occurs along low-altitude desert margins. Low LST concentrates in mountainous regions. Summer LST is relatively stable compared to variable winter conditions; (2) daytime LST exhibited a weak basin-wide cooling trend (− 0.01 °C/yr) during 2001–2024. Significant changes occurred in 8.69% of the basin, with cooling dominating oasis–desert ecotones and alpine meadows; (3) elevation, slope, precipitation, and soil moisture correlate negatively with LST, while air temperature and potential evapotranspiration (PET) show positive correlations; (4) spatial LST distribution is primarily controlled by natural factors, with PET, air temperature, and elevation exhibiting the strongest explanatory power. Irrigation and nighttime lights have minimal impact. Natural topography and moisture dominate LST patterns in arid regions, especially mountainous areas. PET exhibits the strongest interactive effect, highlighting the role of coupled hydrothermal conditions and surface properties. This study enhances the understanding of long-term LST dynamics, supporting regional ecological restoration and climate adaptation strategies.