Under the background of climate change, the Tibetan Plateau presents high spatial and temporal variability in land surface temperature (LST). To understand the spatiotemporal patterns of LST change, this study conducted a spatiotemporal analysis using the Mann-Kendall trend analysis method with time series of mean annual surface temperature (MAST) extracted from the MODIS/Terra daily LST product from 2000 to 2018. The analysis indicated that both daytime and nighttime MASTs show an obvious warming trend with the average rates of 0.028 K/year and 0.069 K/year, and the nighttime variation has larger spatial coverage. Areas ranging from 4500 to 5000 m exhibited the strongest warming effect. The Geodetector method was applied to detect the impacts from seven factors, including elevation, land cover type, latitude, normalized difference vegetation index (NDVI), precipitation, air temperature, and solar radiation on the spatial distribution of LST. The controlling effects of these factors were generally stronger in the nighttime than those in the daytime, and elevation was the most important factor with the contribution scores of 27.12% and 62.98% in the daytime and nighttime, respectively. In addition, the analysis revealed that the temporal changes of LST were mainly caused by surface properties (vegetation, snow cover, and water surface area) changes, radiant flux changes induced by cloud amount changes, and climate warming. In general, this study provides important insights into the spatiotemporal dynamics of LST in the TP since 2000 and helps to reveal the impact of climate change on ecoenvironmental conservation.

No abstract