Abstract: 【Objective】 Grassland is an important part of terrestrial ecosystem in the world. Dynamic monitoring of grassland will help to understand grassland status and provide important theoretical support for sustainable development and management of ecosystem. 【Methods】 Using the 2010-2021 Landsat images of the Kulusitai grassland in Xinjiang,seven vegetation indices were calculated,and the grassland degradation index(GDI) was constructed based on field measurements of grassland height,cover and biomass. An inversion model of degradation in Kulusitai grassland was established based on optimal vegetation index and GDI. The grassland degradation status in the study area was graded and graded,and finally the change in Kulusitai grassland was comprehensively analyzed from three aspects of grassland degradation grade,change direction and change intensity. 【Results】 (1) The weight of aboveground biomass,coverage and height in grassland degradation monitoring was determined by factor analysis in descending order:coverage(37.6%) > aboveground biomass(34.3%) > height(28.1%).(2) NDVI had the highest correlation with GDI among the seven vegetation indices,with R² of 0.854,and the model of grassland degradation monitoring based on NDVI was GDI=0.860×NDVI+0.038.(3) From 2010 to 2021,the severely degraded area of Kulusitai grassland decreased,and the change direction was mainly in the degraded recovery type and the unchanged type,while the degraded enhanced type showed a decreasing trend,and the change intensity was mainly in the unchanged intensity and the slow intensity. 【Conclusion】 As a whole,the degradation process of the Xinjiang Kulusitai grassland has been weakening in the past 12 years.