Analysis of interrelated characteristics between ecosystem services and ecosystem health in the Guangdong-Hong Kong-Macao Greater Bay Area.

Abstract

Ecosystem health (EH) underpins the capacity of vegetation ecosystems to provide essential ecosystem services (ESs), which together are fundamental to regional sustainability. In regions undergoing rapid urbanization, the interrelationships between EH and ESs become increasingly complex, yet they remain largely unexplored in previous studies. This study integrates the VOR and InVEST models to quantify EH and four key ESs in the Guangdong-Hong Kong-Macao Greater Bay Area (GBA) from 2000 to 2020 and further analyzes their interrelationships using a bivariate spatial autocorrelation model and the XGBoost-SHAP approach. The results indicate that: (1) From 2000 to 2020, low-value areas of most ESs and EH expanded, regions of EH deterioration accounted for 71.75% of the study area, indicating the profound impact of rapid urbanization. (2) EH showed strong positive global spatial correlations with CS and NPP, but weak negative spatial correlations with FP and WY. (3) Interrelationships between ESs and EH can be divided into stable synergy type and dynamic trade-off type based on their differing ecological processes; climate factors can significantly impact the interrelationships primarily by affecting the dynamic trade-off type. This study integrates spatial analysis and machine learning approaches to examine the relationships between EH and ESs, thereby advancing the understanding of ecosystem states and functions and providing a theoretical basis for formulating ecological restoration targets.