Answers are grounded in this paper and cite their sources.
Drought is an increasingly important climatic stressor that affects vegetation functioning and ecosystem stability, particularly on the climate‐sensitive Tibetan Plateau (TP). However, most previous studies have relied on a single vegetation index, limiting comparisons of vegetation stability across different indicators. Based on a multidimensional framework comprising resistance, resilience and temporal stability, this study compared vegetation stability under drought across the TP from 1992 to 2020 using gross primary productivity (GPP), Leaf Area Index (LAI), Vegetation Health Index (VHI), and Kernel Normalized Difference Vegetation Index (kNDVI). We further integrated XGBoost‐SHAP with structural equation modelling to characterize the nonlinear statistical associations between climatic variables and vegetation stability and to examine their potential direct and indirect pathways of association. Results revealed a pronounced trade‐off between resistance and resilience across most of the TP. GPP exhibited relatively low resistance (2.66) but high resilience (1.85) and rapid recovery (5.59 months), whereas LAI showed higher resistance (2.74) but lower resilience (1.49) and slower recovery (5.73 months). Long‐term trends indicated increasing resistance and temporal stability for productivity‐ and health‐related indices, whereas resilience displayed divergent trajectories among indices. Mean vegetation stability dimensionality (DS) ranged from 0.73 to 0.77 across the four indices, with mixed forests exhibiting the highest mean DS of 0.84. Relative humidity, precipitation and solar radiation emerged as the dominant climatic drivers, modulating vegetation stability through nonlinear water‐energy interactions. Our findings demonstrate the value of comparing multiple vegetation indices and integrating XGBoost‐SHAP with SEM for interpreting multidimensional vegetation stability under drought on the TP.
No takes yet. Share an insight, caveat, or question.
Du et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: