Spatiotemporal differences in climate change impacts on vegetation cover in China from 1982 to 2015.

The impacts of climate change on vegetation cover in different regions in China are not entirely clear because of the interference of non-climatic factors, such as human activity. This study aims to analyze the spatiotemporal differences in climate impacts qualitatively and quantitatively by applying trend, correlation, and multiple linear regression (MLR) analyses to the data of Normalized Difference Vegetation Index (NDVI) and two climatic factors (air temperature and precipitation) during 1982-2015 in China. The MLR equation linking two climatic variables with NDVI was used to identify the NDVI trend caused by climate change. We demonstrated that the central and eastern regions of China, dominated by deciduous and evergreen broadleaf forests, experienced a rapid increase in NDVI from 1982 to 2015. The response of NDVI to variations in temperature and precipitation exhibited large spatiotemporal differences across China, which was closely related to climatic conditions and vegetation types. Overall, warming, particularly the sharp rise in spring, was the main climatic driving force behind China's NDVI increase, and precipitation also influenced the NDVI increase in temperate grassland and desert regions due to the relatively arid climate, particularly in summer. The contributions of climate change to the total NDVI trend (CC) showed a large spatiotemporal heterogeneity across China. Overall, only 45% of the pixels (with a resolution of 8 km) in the study area showed that the MLR equations between NDVI and two climatic factors were significant at the 0.05 significance level during the growing season (April-October), and the average CC of these pixels was 38%. Among the eight vegetation sub-regions of China, the temperate desert and Qinghai-Tibet Plateau alpine meadow regions generally exhibited relatively larger CCs than other vegetation sub-regions in different seasons. At a national scale, the regional average CC reached 64% during the growing season. These results at multiple scales can help to deeply understand the mechanisms of regional environmental variation and sustainability.