Causal Inference Reveals Temperature as the Dominant Driver of Vegetation Greening on the Qinghai-Tibetan Plateau over the Past Two Decades

The Qinghai-Tibetan Plateau (QTP) is a globally vital ecological security barrier and highly climate-sensitive region, where vegetation dynamics serve as key indicators of climate change and anthropogenic activities. However, conventional methods provide limited causal interpretability, which severely constrain our understanding of vegetation dynamic processes. To address this limitation, this study employed causal inference methods to investigate the causal relationships underlying climate change and human activities affecting vegetation dynamics on the QTP. The results show that: (1) From 2001 to 2020, the QTP has shown an overall improvement in its ecological environment, with the most notable greening seen in mid-altitude regions. (2) The average time lags of vegetation responses to precipitation (Pre), temperature (Tem), soil moisture (SM), land surface temperature (LST), and surface solar radiation (SSR) were 0.81 ± 0.79, 0.74 ± 0.68, 0.59 ± 1.02, 1.26 ± 0.89, and 1.93 ± 0.66 months, respectively. With increasing altitude, the lag effects tended to weaken. (3) Causal inference revealed temperature as the dominant driver of vegetation greening on the QTP. Furthermore, a unidirectional causal relationship between human activities and vegetation changes was detected. The results improve the causal interpretability and offer a novel perspective for attributing changes in surface vegetation.