Causal relationship between seasonal variation in surface roughness and dust and sandstorm occurrence in arid and semi-arid Mongolia

Abstract

In this study, the causal relationship between seasonal variations in surface conditions and DSS generation was investigated through the integrated use of the C-band backscatter coefficient (J 0) of the Sentinel 1 microwave satellite from the optical satellite sensor (Terra/MODIS). Because of the very high reflectance of the background soil in sparsely vegetated bare land, it has been difficult to rely solely on optical satellite data for accurate surface condition assessment in previous studies. Therefore, we used SAR and optical data to represent the spatial variability of backscatter coefficients for different surface conditions in the Gobi Desert and Steppe areas, and employed the C-band backscatter coefficient from the Sentinel-1 microwave satellite as a key parameter. The results show that in the Gobi Desert region, the correlation between the seasonal variation of backscatter coefficients and NDVI is highly significant, with a maximum correlation coefficient in fall (R=0.645, p<0.001) an inverse correlation in summer (R=-0.216, p<0.001), indicating little seasonal variation between NDVI and backscatter coefficients. In conclusion, surface roughness (σ0) is relatively low and stable from winter to the following spring and increases with plant growth in summer. In the Gobi Desert region, surface roughness changes little throughout the year, but in the steppe region, surface roughness increases with vegetation growth during the rainy season, which suppresses the generation of DSS.