Responses of vegetation growth to temperature during 1982-2015 in Xinjiang, China.

Abstract

Combined with the normalized difference vegetation index (NDVI) dataset, vegetation type data, and meteorological data, we revealed the variation of vegetation growth responses to air temperature in the growing-season during 1982-2015 in Xinjiang, using the moving-windows based partial correlation analysis, the unitary linear regression analysis and GIS spatial analysis. Results showed that, in the whole growing-seasons of study period, there was a significant downturn trend in the responses of vegetation growth to temperature. At the seasonal scale, the downturn trend was obvious especially in summer and autumn, while it was in adverse in spring. During the whole gro-wing season, the responses of different vegetation types to air temperature change showed a decreasing trend. Seasonally, the responses of grassland and forest to temperature change showed a significant increase, while that of shrubland and desert were exactly the opposite in spring. The responses of all natural vegetation (grassland, shrubland, desert and forest) to temperature change showed a significant decreasing trend in summer, whereas their responses in autumn had no significant statistical characteristics. Spatially, the decreasing influence of temperature on the vegetation growth during the growing season in Xinjiang was universal, which might be due to the change in precipitation and solar radiation.