Integration of MODIS-derived indices and field observations to estimate surface soil moisture at regional scales

Abstract

Surface soil moisture (SSM) is a critical factor in monitoring climate change, soil fertility, flood, and runoff modeling. Integration of satellite earth observations and field measurements data is a reliable approach to estimate environmental parameters in remote sensing applications. In this paper, a method is proposed to estimate the SSM by integrating synoptic weather stations and MODIS imagery on a regional scale. The data were adopted from the Soil Climate Analysis Network of the United States (US-SCAN) stations that were routinely collected in more than 220 stations from 2012 to 2015. The proposed method is a regression model composed of spectral indices including Normalized Difference Water Index (NDWI), Visible and Shortwave infrared Drought Index (VSDI), land surface temperature (LST), and estimated surface soil temperature using ordinary Kriging (OK). This method has inspired a simple integration method that used a linear combination of the remote sensing and field LST measurements. Compared to the inspired method, the proposed method has shown a 24% improvement in SSM estimation.