Key Laboratory of Mapping from Space of State Bureau of Surveying and Mapping,Chinese Academy of Surveying and Mapping,青海省基础地理信息中心,Key Laboratory of Mapping from Space of State Bureau of Surveying and Mapping,Chinese Academy of Surveying and Mapping,
Qinghai Lake Basin (QLB) belongs to the alpine grassland ecosystem. The grassland degradation is an effective detector to reflect the ecological environment condition in QLB. In this study, the spatial distribution and change information of the degraded grassland in QLB were firstly mapped and extracted using remote sensing techniques. Then the spatial-temporal changes of the degraded grassland between the year 2000 and 2010 were analyzed in different perspectives using Markov model, index of dynamic variation degree, landscape pattern index and grassland change index. Results show that the grassland in QLB was in good condition in 2010, and the non-degraded meadow and non-degraded grassland dominated QLB and covered about 38% of this region. The grassland has little variation during these 11 years, and the index of dynamic variation degree is -0.0384. The grassland was improved after the first degradation, but still had slight degradation in the overall trend. The spatial landscape pattern of the degraded grassland was stable in this study. The non-degraded meadow is dominantly distributed in the central region of QLB. The spatial-temporal changes of this land cover type were stable during these 11 years, and its dynamic variation degree was 1.82%. The change of moderate-degraded meadow was the most significant. During the period of 2006-2010, even the grassland in QLB was obviously improved; there was almost 7.4 ten thousand hectare of heavy-degraded meadow from moderate-degraded meadow, covering 65% area of land cover types from moderate-degraded meadow. Light-degraded meadow, moderate-degraded meadow and heavy-degraded grassland were dominantly located in northwest region of QLB, north and surrounding area of Qinghai Lake. The spatial changes of these three land cover types evolved from and into other types were the most significant, and their dynamic variation degree floated between 15% and 20% from the year 2006 to 2010.