Assessing the Sensitivity of Vegetation Cover to Climate Change in the Yarlung Zangbo River Basin Using Machine Learning Algorithms

Comprehensive evaluation system for vegetation ecological quality: a case study of Sichuan ecological protection redline areas

Dynamic monitoring and evaluation of vegetation ecological quality (VEQ) is indispensable for ecological environment management and sustainable development. Single-indicator methods that have been widely used may cause biased results due to neglect of the variety of vegetation ecological elements. We developed the vegetation ecological quality index (VEQI) by coupling vegetation structure (vegetation cover) and function (carbon sequestration, water conservation, soil retention, and biodiversity maintenance) indicators. The changing characteristics of VEQ and the relative contribution of driving factors in the ecological protection redline areas in Sichuan Province (EPRA), China, from 2000 to 2021 were explored using VEQI, Sen's slope, Mann-Kendall test, Hurst index, and residual analysis based on the XGBoost (Extreme gradient boosting regressor). The results showed that the VEQ in the EPRA has improved over the 22-year study period, but this trend may be unsustainable in the future. Temperature was the most influential climate factor. And human activities were the dominant factor with a relative contribution of 78.57% to VEQ changes. This study provides ideas for assessing ecological restoration in other regions, and can provide guidance for ecosystem management and conservation.