Intercomparison of spring soil moisture among multiple reanalysis data sets over eastern China

Statistical characteristics of spring soil moisture in different reanalysis data sets of ERA-Interim, Modern Era Retrospective-Analysis for Research and Applications (MERRA), Japanese 25-year Reanalysis, Climate Forecast System Reanalysis, and National Centers for Environmental Prediction/National Center for Atmospheric Research-Reanalysis 1 (NCEP/NCAR-R1) are intercompared with each other and with the observations over China. The spring soil moisture is largest in southeastern China and smallest in northwestern China in climatology. It exhibits a pronounced interannual variability with opposite variation in the midlatitude zone and northeastern China. There exist a wet trend at midlatitudes and a dry trend in northeastern China. The intercomparison shows that, except NCEP/NCAR-R1, the reanalyses can reproduce the observed gradual increases of climatological soil moisture in China from the northwest to the northeast and to the southeast. MERRA presents the best climatological soil moisture. Only ERA-Interim can well represent the interannual variations of observed soil moisture. The first empirical orthogonal function mode of observed soil moisture demonstrates that the variability of soil moisture is most robust in the midlatitude zone of eastern China and the ERA-Interim is the best in reproducing the spatial and temporal features. The reasons causing differences between reanalyses of soil moisture are also investigated in terms of two main factors affecting soil moisture, precipitation and evaporation. The ERA-Interim can well reproduce the precipitation and evaporation from observations as well as their relations to soil moisture, resulting in a preferable ability to represent the spatial and temporal characteristics of observed soil moisture. Although the other four reanalysis data sets reproduce precipitation well, their poor ability to describe the evaporation causes large differences of soil moisture between their simulations and observations.