Abstract: Based on the concept of structured vegetation cover index (C_s) and by using TM images as the information source, the extraction way of C_s for Loess
Area in North Shaanxi by using remote sensing techniques was explored. In study area, C_s was better than the traditional projected vegetation overage index in expressing the relationships between vegetation structure and soil erosion. The C_s was closely related to the remote sensing vegetation indices, such as green indices NDVI (Normalized Difference Vegetation Index) and MSAVI (Modified Soil Adjusted Vegetation Index), and yellow indices NDSVI (Difference Senescent Vegetation Index) and NDTI (Normalized Difference Tillage Index). The combination of the green and yellow indices could better express the effects of vegetation on soil erosion, compared with the single index. Among these remote sensing vegetation indices, the MSAVI and NDTI could be the ideal green and yellow vegetation indices for the extraction of C_s from TM images. It was possible to extract the C_s from remote sensing data through the regression analysis of C_s and remote sensing vegetation indices. However, this method was just valid
ated and applied to the study area. Whether it could be applied to other regions was needed to be further validated due to the phonological differences from one region to another.

Key words: vegetation index, vegetation coverage, soil and water loss, vegetation structure, water spinach (Ipomoea aquatica),  , cultivar, rhizosphere, soil chemical characteristics, low Cd accumulation.