Analysis of Seasonal Grassland Change and Its Drivers during 1982-2006 in Xinjiang

Abstract

Xinjiang is the largest inland province of China, where the ecological environment is vulnerable. The relationship between grassland growth and single climatic factor and howdistinctly the climatic factor trends have influenced grass trends is poorly understood. To address these questions, the Global Inventory Monitoring and Modeling Systems (GIMMS) Normalized Difference Vegetation Index (NDVI) dataset from 1982 to 2006 is used to evaluate the seasonal changes of NDVI over progressively longer periods from 18 to 25 yr. Multiple regression analysis is used to detect the relation between NDVI and the drivers. Partial correlation analysis and Pearson correlation are used to discuss the difference between the two coefficients. Trend analysis and the Mann-Kendall test are used to detect the implication of climatic factor in the change of grass growth. The results indicate that grassland in Xinjiang represented a slight increasing trend over the 25 yr, with an increasing rate of 7.0 × 10-4 yr-1. The increase of livestock has little effect on NDVI and regional productivity. The partial correlation analysis indicates that both precipitation and temperature are important to grassland dynamics in each season. But precipitation, especially for growing season and summer, plays a more crucial role in influencing the grassland change. However, the precipitation is more influential in a 1-yr scale. In addition, there is a time lag of about 1 mo between the NDVI and precipitation fluctuation in this area. The reasons why the NDVI increases so slowly include the trend variation before and after the turning point of the growing season precipitation trend and the decrease or stalling of precipitation in some areas. The conclusions from this study provide guidance for preserving the ecosystem in inland dry areas and should be considered when carrying out vegetation countermeasures. The methodologies can be transferred to other areas with different climatic conditions. © 2017 The Society for Range Management. Published by Elsevier Inc. All rights reserved.