Resource information
Monitoring of soil moisture dynamics provides valuable information about grassland degradation, since soil moisture directly affects vegetation cover. While the Mongolian soil moisture monitoring network is limited to the urban and protected natural areas, remote sensing data can be used to determine the soil moisture status elsewhere. In this paper, we determine whether in situ and remotely sensed data in the unaccounted areas of Southwestern Mongolia are consistent with each other, by comparing Soil Moisture and Ocean Salinity (SMOS) first passive L-band satellite data with in situ measurements. To evaluate the soil moisture products, we calculated the temporal, seasonal, and monthly average soil moisture content. We corrected the bias of SMOS soil moisture (SM) data using the in situ measured soil moisture with both the simple ratio and gamma methods. We verified the bias-corrected SMOS data with Nash–Sutcliffe method. The comparison results suggest that bias correction (of the simple ratio and gamma methods) enhances the reliability of the SMOS data, resulting in a higher correlation coefficient. We then examined the correlation between SMOS and Normalized Difference Vegetation Index (NDVI) index in the various ecosystems. Analysis of the SMOS and in situ measured soil moisture data revealed that spatial soil moisture distribution matches the rainfall events in Southwestern Mongolia for the period 2010 to 2015. The results illustrate that the bias-corrected, monthly-averaged SMOS data has a high correlation with the monthly-averaged NDVI (R2 > 0.81). Both NDVI and rainfall can be used as indicators for grassland monitoring in Mongolia. During 2015, we detected decreasing soil moisture in approximately 30% of the forest-steppe and steppe areas. We assume that the current ecosystem of land is changing rapidly from forest to steppe and also from steppe to desert. The rainfall rate is the most critical factor influencing the soil moisture storage capacity in this region. The collected SMOS data reflects in situ conditions, making it an option for grassland studies.
