Spatial and temporal variations of soil moisture and its impact factors in the Golmud River Basin

Soil moisture is a relative variable which can characterize soil water content. It is the core and connection among the groundwater, surface water and atmosphere water cycle system. It is of great significance to study the distribution characteristics of soil moisture and its impact factors on the ecological environment protection in arid regions. Based on MODIS data, this paper uses the apparent thermal inertia (ATI) method to invert the temporal and spatial variation of soil moisture in the Golmud River Basin and analyzes its impact factors. In order to examine the reliability of the results, the Mann-Kendall method is used to verify the trend test analysis. The CLDAS data and the field measured soil water content data are used to verify the remote sensing inversion results. The results show that from 2002 to 2016, the soil moisture in the Golmud River Basin generally shows an increased trend, and the Mann-Kendall test show that the mutation occurred in 2008 and the increasing trend is obvious. According to interpretation of different land types, the northern salt pond has the highest average ATI value of 0.088. The average ATI value, 0.028, of the bare soil in the middle and southern areas is the lowest. The average ATI values of the natural vegetation, farmland and forest land are 0.034 and 0.033, respectively. The correlations between the remote sensing inversion soil moisture results and CLDAS data and field measured soil water content data are 0.85 and 0.59, respectively, and the high correlation indicates that the remote sensing inversion results are of certain reliability. The soil moisture has good response relationship with vegetation and evapotranspiration, and the correlation coefficients are 0.92 and 0.79, respectively.