Vegetation dynamics and responses to recent climate change in Xinjiang using leaf area index as an indicator

There is a strong signal showing that the climate in Xinjiang, China has changed from warm-dry to warm-wet since the early 1980s, leading to an increase in vegetation cover. Based on a regression analysis and Hurst index method, this study investigated the spatial-temporal characteristics and interrelationships of the vegetation dynamics and climate variability in Xinjiang Province using the leaf area index (LAI) and a gridded meteorological dataset for the period 1982-2012. Further analysis focused on the discrimination between climatic change and human-induced effects on the vegetation dynamics, and several conclusions were drawn. (1) Vegetation dynamics differ in mountain and plains regions, with a significant increasing trend of vegetation cover in oases and decreasing trend of vegetation growth in the Tienshan and Altay Mountain. The Hurst exponent results indicated that the vegetation dynamic trend was consistent, with a sustainable area percentage of 51.18%, unsustainable area percentage of 4.04%, and stable and non-vegetated area ratio of 44.78%. (2) The warm-dry to warm-wet climatic pattern in Xinjiang Province since the 1980s mainly appeared in the western part of the Tienshan region and North Xinjiang. Temperatures increased in all seasons over the majority of Xinjiang, and precipitation showed a significant increasing trend in the mountainous regions in spring, summer and autumn, whereas the rate of precipitation change was higher in the plains region in winter compared with that in other seasons. (3) A correlation occurs between the climate variables (precipitation and temperature) and mean LAI, and this correlation varies at the seasonal and regional scales, with coniferous forest, meadow and grassland more correlated with precipitation in spring and summer and not correlated with temperature, which indicated that precipitation was the dominant factor affecting the growth of mountain vegetation. The mean LAI of vegetation in the plains exhibited significant correlation with precipitation in winter and temperature in spring and summer. (4) A residual analysis showed a human-induced change that was superimposed on the climate trend and exhibited two effects: vegetation regeneration in oases throughout Xinjiang and desertification in the meadow located in the mountainous area of the western Tienshan Mountains and Altay Mountains. (5) Grassland is the most sensitive vegetation type to short-term climatic fluctuations and is the land-use type that has been most severely degraded by human activity; thus, local governments should take full advantage of this climatic warm-wet shift and focus on protecting vegetation to improve this fragile arid environment.