Nonlinear dynamics of ecosystem productivity and its driving mechanisms in arid regions: A case study of Ebinur Lake Basin

Arid zone ecosystems exhibit significant sensitivity to climate change and human activities, often demonstrating pronounced nonlinear characteristics in their response processes. This study focuses on the typical inland basin of the Ebinur Lake Basin in the arid region of northwest China. Utilizing long-term remote sensing and meteorological data from 1982 to 2018, combined with an improved BFAST algorithm, the XGBoost-SHAP analytical framework, and Partial Least Squares Structural Equation Modeling (PLS-SEM), we systematically investigated the nonlinear variation characteristics of Gross Primary Productivity (GPP) and its underlying driving mechanisms. The results reveal that tipping points in GPP changes were detected in 99.78 % of the Ebinur Lake Basin, with the "Negative Reversal" type (initially increasing and then decreasing) being the most prevalent, accounting for 35.91 % of the cases. The majority of GPP tipping points occurred between 1998 and 2002, with the highest frequency observed in 1998 (10.54 %). Vapor pressure deficit (VPD) was identified as the primary factor controlling GPP changes in the Ebinur Lake Basin. However, in areas where the ecosystem exhibited a recovery trend, temperature surpassed VPD as the dominant driving factor, indicating that improved temperature conditions are critical for productivity restoration. These findings enhance our understanding of the nonlinear characteristics and underlying mechanisms of arid zone ecosystems, providing a scientific basis for adaptive ecosystem management in the region.