Simulation and Uncertainty of Active Layer Thickness of Permafrost by Land Surface Model CAS-LSM under Different Atmospheric Forcing Data

Permafrost plays an important role in the climate system, which is of great significance in revealing the temporal and spatial variation of frozen ground to increase the understanding of land–air interaction. In this study, the Chinese Academy of Sciences Land Surface Model (CAS-LSM) was adopted, which contains the dynamic changes of the soil freezing and thawing fronts. Four different atmospheric forcing data, CRU-NCEP (Climatic Research Unit-NCEP forcing data), GSWP3 (Global Soil Wetness Project forcing dataset), Princeton (Princeton meteorological forcing dataset), and WFDEI (water and global change forcing data methodology applied to ERA-Interim data), were used to drive the CAS-LSM. The simulation time of the four groups of the simulation experiments was from 1960 to 2009 with a resolution of 0.9°×1.25°. The variation trend and uncertainty of permafrost active layer thickness under different atmospheric forcing were studied. The simulated active layer thickness compared well with the observed data. Results reveal an increasing trend of the active layer thickness under different atmospheric forcings. The variation trend of the forcing data WFDEI simulation is the largest. The regional average active layer thickness and variation trend range of different atmospheric forcing simulations were 1.1–1.25 m and 0.27–0.51 cm/a, respectively, with the relative changes of 11.2%–23.5%. The range of regional average active layer thickness and variation trend in Qinghai–Tibet plateau region, North America, and north of Eurasia are 2.26–2.81 m, 1.07–1.31 m, 1.32–1.48 m, and 0.47–1.04 cm/a, 0.29–0.48 cm/a, 0.25–0.55 cm/a. The difference in air temperature in the atmospheric forcing data is the main reason for these variations.