Abstract
Ground surface temperature (GST, depth of 5 cm) is essential for understanding the climate-induced impacts in the Earth Critical Zone, especially in cold regions. This work aims to detect the GST variability from fine (2 to 16 m) to local and landscape scales (2 and 50 km2) and along an 800-m elevational transect. The study area is located in the southern part of the Headwater Area of the Yellow River (HAYR), a representative area of discontinuous alpine permafrost on northeastern Qinghai-Tibet Plateau. From 2019 to 2020, GST was recorded at 39 sites under similar and varied environmental conditions. A statistically significant correlation was identified between mean annual GST (MAGST) with landcover (0.37, p < 0.05) and elevation (−0.76, p < 0.001). The analysis of variance (ANOVA) showed significant differences in MAGST between sites on bare ground and with vegetation. At the local/landscape scales, MAGSTs in meadows (−0.6 to 0.6 °C) were higher than those in swamp meadows (−1.4 to 0.7 °C) and bare grounds (−2.0, − 1.2 °C). At the fine scale, GST revealed a high similarity between plots (R2 > 0.95 for 90 % of the sites) but with an RMSE ranging from 0.25 to 1.5 °C. The intra-plot difference in MAGST was below 0.5 °C for sites with similar landcover in both plots and up to 1–2 °C when vegetation was compared to the bare ground. The decrease of MAGST with rising elevation was more pronounced for the sites in bare ground (R2 = 0.83, p < 0.001) than vegetated ones (e.g., meadow: R2 = 0.64, p < 0.003). These results help identify the patterns of permafrost distribution and thaw at a high spatial resolution and better understand their relationship with vegetation development, soil organic carbon, and the rapid landscape changes in the HAYR.