Time-lagged response of vegetation to the 2009 extreme drought in the Tibet Autonomous Region

Global warming has led to an increasing frequency of extreme drought events, posing a severe threat to vegetation growth. A thorough understanding of the mechanisms by which vegetation responds to extreme drought is crucial for assessing its impact on terrestrial ecosystems. However, the response characteristics of vegetation to extreme drought in the Tibet Autonomous Region, the lag times of different vegetation types, and the relative importance of influencing factors remain unclear, limiting our understanding of vegetation-climate interaction mechanisms in arid regions. This study used the Standardized Precipitation Evapotranspiration Index (SPEI) to reveal the spatiotemporal distribution of extreme droughts in the TAR from 2000 to 2020, and employed Pearson correlation, Random Forest, and Shapley Additive Explanations (SHAP) methods to evaluate the time-lagged effects of the 2009 extreme drought on vegetation and the importance of its influencing factors. The results indicate that the frequency of extreme drought events in Tibet has continued to increase, with significantly higher occurrence in the western and eastern regions compared to other areas. During the 2009 extreme drought, the lag times of vegetation responses were mainly concentrated within 1-3 months, accounting for 68.1% of cases. Overall, the average time lags of the four vegetation types in response to extreme drought were shrubland (2.57 months) > forest (2.49 months) > grassland (2.38 months) > cropland (2.23 months), indicating that cropland is the most sensitive to drought. Regarding the dominant factors influencing vegetation responses to drought, potential evapotranspiration (PET) was the primary influencing factor common to all four vegetation types. Forest drought responses were also regulated by precipitation, temperature, and deep soil moisture (100-289 cm); grasslands depended on both precipitation and shallow soil moisture (0-7 cm); whereas croplands exhibited higher sensitivity to topographic factors. This study enhances the understanding of climate-vegetation relationships in the TAR and provides theoretical support for addressing extreme drought risks under climate change.