Grassland degradation and restoration monitoring and driving forces analysis based on long time-series remote sensing data in Xilin Gol League

Land degradation is a process during which the land's productive capacity declines and eventually becomes completely lost under the influence of natural forces and human activities. With the development of remote sensing technology, long time-series of vegetation parameters has become available. In this study, time-series annual net primary production (NPP) datasets covering Xilin Gol League, Inner Mongolia, China during 2001 to 2012 were established based on an improved Carnegie-Ames-Stanford Approach (CASA) model. Then, the areas of grassland degradation and restoration were determined using the Sen+Mann-Kendall method. Finally, the driving forces of grassland degradation and restoration in this area were distinguished over the past 12years through multiple and partial regression methods. The results showed the following five major findings: (1) From 2001 to 2012, areas showing degraded and restored trends were 2.36% and 9.37%, respectively, at the confidence level of α=0.1. There was a significant restored trend in Otindag sandy land and its surroundings, which indicates that some ecological engineering projects have achieved significant results. (2) Based on the combined analyses of multiple regression and partial regression, the main driver of grassland degradation in Xilin Gol League was identified as human activities, whereas climate change had a small influence. The effects of both human activities and climate change were the main drivers of grassland restoration; the single effect of human activities also played an important role in grassland restoration. (3) By comparing land use types in 2000 and 2010, we found that urban expansion and road construction occupied a major portion of grassland in Xilin Gol League in the past 10years. Under the influence of the human activities, 3.2% of grassland experienced degradation and became bare land. In contrast, some areas showed vegetation recovery: 7.1% of bare land transformed to grassland. (4) By analyzing vegetation changes in the key nature reserves and coal mining areas, we found that vegetation in the earlier exploited mining areas was influenced seriously by human activities and showed a degradation trend; vegetation in the earlier protected nature reserve showed a restored trend under human activities. Simultaneously, in the new nature reserve, grassland productivity is improving. (5) The proposed methods of grassland degradation and restoration monitoring and driving force analysis were suitable for long time-series vegetation indicators datasets at the regional scale. These results may allow the local government to develop land degradation control strategies and provide a basis for using this improved method to study the influence of global climate change on land degradation.