Soil temperature and moisture are the important parameters in land surface process, and they are also important physical parameters in boundary conditions of atmospheric numerical model. This paper tried to obtain spatial-temporal evolution of soil moisture of the model through the machine learning method according to the memory characteristics of soil moisture. Considering the influence of soil temperature on soil moisture, the soil temperature and moisture of ERA5 reanalysis at depths of 0-7, 7-28, 28-100, 100-289 cm are used as predictors to predict changes of soil moisture on a monthly and seasonal scale based on convolutional neural networks (CNN). The results show that the method proposed in this paper is reliable and can effectively predict soil moisture 6 months in advance. The mean bias of predicted soil moisture in the shallow layer (0-28 cm) and deep layer (28-289 cm) is less than 0.05 and 0.02 m³·m^-3, respectively. In the humid area, the mean bias is basically within 0.03 m³·m^-3, showing a good effect.The prediction method and results presented in this paper can be used for both soil drought prediction and the initial and boundary conditions for numerical models.

A Tibetan Plateau Vortex (TPV) generated on the western Tibetan Plateau on 14 August 2006 was simulated using the Weather Research and Forecasting (WRF) regional model. Based on the evaluation of the performance of the WRF driven by different reanalysis/analysis datasets including the ERA5 and ERA-Interim data from the European Centre for Medium-Range Weather Forecast, and the final analysis data (FNL) of the global forecasting system of the National Centers for Environment Prediction (NCEP) on the simulation of TPV, the kinematic and thermodynamic structures during the genesis of TPV were investigated through the equation of  vorticity budget, apparent heat source and apparent moisture sink. The results show that ERA5 and ERA-Interim could well reproduced the genesis of TPV. The ERA5 had the best ability to simulate the low-level cyclonic circulation, the spiral rainbands of TPV, and it produced the heaviest rainfall among three simulations. The FNL data could not portray the genesis of this TPV and its rainfall pattern. The development of positive vorticity, heat and moisture budgets during TPV generation were closely related to corresponding vertical transportation in lower troposphere. Before the genesis of TPV, surface sensible heating played an important role. During the development of TPV, strongest upward motion and most significant rainfall occurred, which was contributed mainly by obvious latent heating of condensation and eddy vertical transport process of heat and moisture, and conducive to the development of simulated vortex.