Spatiotemporal Trend Analysis of Soil Moisture Retrieved From Three NLDAS-Based Advanced Land Surface Models over the United States: A Comparative Study

Soil moisture content is an important hydrological parameter, which is of great significance for drought evaluation. In this paper, soil moisture simulated from three NLDAS-based Advanced Land Surface Models including Community Land Model version 4.0 (CLM4.0), Noah LSM with Multiphysics options (Noah-MP) and Catchment LSM-Fortuna 2.5(CLSM-F2.5) are used to study the trend in the past 35 years (1980-2014) over the conterminous US (CONUS). The water content of the soil surface layer (0-10cm) and root layer (0-100cm) is the focus of this study. The Mann-Kendall mutation test, and Theil-Sen approach are used to analyze the trend of soil moisture. The results from CLM4.0 are different from those from CLSM-F2.5 and Noah-MP. The change of surface soil water content is similar to that of root soil water content, and the surface soil water was affected by the meteorological conditions. The trend of soil moisture in autumn is quite different from that in the other three seasons over CONUS, especially in the central plain area.