Jinghe River Basin is chosen as a case to study vegetation degradation in Loess Plateau in this paper. Firstly

remote sensing based vegetation index （VI） and climatic aridity index are used to estimate potential vegetation index by regression model approach. The spatial distributions of current vegetation and potential vegetation in Jinghe River Basin are then obtained according to supervised classification of NDVI remote images and potential vegetation index. The status or severity of vegetation degradation is analyzed by means of the transfer possibility between vegetation types. The result shows that the main potential vegetation types of the Jinghe River Basin are broadleaved and coniferous mixed sparse forest （32.44%）

broadleaved deciduous forest （31.28%） and shrub（23.71%）. It is found that 25.08% and 13.32% of the broad leaves forest potential region were converted to cultivated land and broadleaved and coniferous mixed sparse forest

while 13.04% and 14.22% changed to shrub and aridity shrub with only 25.09% potential region remained broad leaves forest. Potential region of the broadleaved and coniferous mixed spare forest changed to cropland（26.01%）

aridity shrub（20.99%）and meadow （17.12% ） whereas potential region for shrub was dominant by meadow （30.29%） and grassland （43.21%）. Vegetation degradation is most serious in loess gully region

followed by loess hilly region in northern basin whereas vegetation degradation in mountain region of Ziwuling in eastern basin and Liupanshan in western basin are relative slight.