Simulation of soil moisture and its variability in East Asia

Soil moisture and related hydrological process play an important role in regional and global climates. However, large-scale and long-term observation of soil moisture is sparse. In this study, the latest NCAR Community Land Model is used to simulate regional soil moisture in East Asia for recent 25 years with the atmospheric forcing provided by NCEP/DOE reanalysis. A 50-year simulation has been conducted with the first 25 years as the model spins up for soil moisture to reach steady state. The last 25 years simulation provides a soil moisture dataset with physical consistency and spatio-temporal continuity. Our analysis focuses on spatial and temporal variability of the regional soil moisture based on the last 25-year modeling. Additionally, The trend in the regional soil moisture and its possible link to climate warming is examined. The main conclusions can be summarized as follows: 1. Simulated soil moisture exhibits clear sensitivity to its initial condition. Such sensitivity is a function of soil depth. This study indicates that the equilibrium time of soil moisture increases with the depth of soil layers. It takes about 20 years to reach equilibrium below 1.5m. Therefore either a longer spin-up (20 years or more) or accurate initial soil moisture is necessary for a quality land surface modeling. 2. In comparison with the reanalysis and in-situ measurements, the model reproduces the observed large-scale structure reasonably well. The simulation shows mesoscale spatial variation as well. 3. Linear trend analysis shows that soil has become drier in most areas of East Asia in recent years except southern China and the Tibetan Plateau where soil gets wetter. Further analysis indicates that such dry trend may have a close link to warming surface climate through enhanced evaporation.