Seasonal dynamics and regional features of MODIS-derived land surface characteristics in dust source regions of East Asia

Atmospheric mineral dust is an important component of the climate system; however, representation of dust production in the climate models poses significant challenges. Satellite remote sensing has the potential to aid in determining the surface characteristics of active dust source regions that are of importance to dust emission modeling. This study uses data from the Moderate-resolution Imaging Spectroradiometer (MODIS) in conjunction with soil texture to investigate linkages of spatial distribution of surface characteristics related to dust emission, and their dynamics, at the seasonal time scale. In addition to standard MODIS land products such as surface albedo and NDVI which are strongly linked to dust emission, we introduce a roughness index (RI) and an arid soil surface index (ASSI) to aid in land surface characterization. Three regions of northwestern China known for dust emission, the Taklamakan, Badain Jaran, and Gurbantunggut Deserts, are examined for spatial and temporal changes of surface characteristics during winter-spring-summer transition February-July 2005. A new methodology is proposed by introducing regional masking derived from MODIS Band 10 surface albedo. The analysis demonstrates regional unique temporal and spatial characteristics in the 2005 seasonal transition for these areas. Seasonal modes of response are clearly present. The soil texture correlation results demonstrate that clay fraction has a consistently high negative correlation to albedo, as does vegetation. The analysis also demonstrates that RI is a dynamic characteristic changing both with season and on much shorter time scales. (C) 2010 Elsevier Inc. All rights reserved.