Improving spatial representation of soil moisture by integration of microwave observations and the temperature–vegetation–drought index derived from MODIS products

Abstract

The microwave observations of land surface soil moisture have been widely used for studying environmental change at large spatial scales. However, the coarse spatial resolution of the products limits their local-scale applications. In this paper, we developed a new method, which integrates the coarse spatial resolution soil moisture derived from microwave sensors and the temperature–vegetation–drought-index (TVDI) derived from the Moderate-resolution Imaging Spectroradiometer (MODIS) products, to spatially downscale soil moisture data from 25-km resolution to 1-km resolution. First, we assessed the quality of the remotely sensed soil moisture by comparing their values with field measured soil moisture at three temporal scales and two spatial scales. Second, we analyzed the robustness of the developed approach namely the PKU method by comparing its performance with the results of three published methods (i.e., the triangle-based method, the Merlin method, and the UCLA method) at the Magqu soil moisture monitoring network located in the northeastern Tibetan grasslands. The modeling results showed that by integrating the contextual information from the relatively fine spatial resolution MODIS products, spatial soil moisture representations were significantly improved. The PKU method produced the most accurate spatially disaggregated soil moisture among the four methods. In conclusion, the PKU method developed in this study is a practical and efficient approach for improving spatial representations of the coarse spatial resolution soil moisture data derived from microwave remote sensors. Within the PKU method, our refined method for estimating the parameters of the dry-edge outperforms the traditional method.