Vegetation variations and its driving factors in the transition zone between Tibetan Plateau and arid region

The Qilian Mountains (QLMs) has functioned as an ecological barrier area; water connotation area; priority area for biodiversity conservation; practice area; bearing area; and demonstration area for the construction of ecological civilization in China; and its ecological function and demonstration role cannot be ignored. To investigate the response of vegetation to climate change and human activities, the GIMMS NDVI3g, MODIS NDVI, meteorological data, and nighttime light data were analyzed using Theil-Sen trend analysis, Mann-Kendall significance test, Hurst index, and correlation analysis to explore the spatial and temporal characteristics and trends of vegetation in the QLMs from 1982 to 2020. The results show that from 1982 to 2020, the NDVI in the QLMs showed an increasing trend in the interannual, growing season (0.0011/a), and through the four seasons-spring (0.0003/a), summer (0.0015/a), autumn (0.0007/a), and winter; and demonstrated a high spatial pattern in the southeast and low spatial pattern in the northwest. The NDVI showed a gradual increase in the improved area (46.43% -* 79.32%), a gradual decrease in the unchanged area (50.81% -* 18.23%), and a stable degraded area (2.76% -* 2.45%), but the distribution of the degraded area shifted from the northern part of Daxue Mountain to the Qinghai Lake and Datong River basin, and the intersection of Qinghai South Mountain and Laji Mountain. Additionally, the vegetation cover was not persistent, and the area of unclear future was more than half and distributed in the central and western parts of the QLMs, with obvious anti-sustainability characteristics. Temperature was the dominant climatic factor for vegetation growth in the QLMs, and from the spatial distribution, the area controlled by temperature (63.81%) was significantly higher than that controlled by precipitation (29.95%). On a temporal scale, NDVI during spring, autumn, and winter was positively correlated with the temperature and negatively correlated with precipitation; and vegetation growth was influenced by temperature for a longer period of time during the year. NDVI has an obvious lagging effect on climate change, with a maximum response to temperature and precipitation lagging by approximately 2-3 months and 0-1 month, respectively. Human activities have both promoting and inhibiting effects on vegetation. In human activity areas, NDVI shows an increasing trend, and the ecological environment is improving annually.