Ecological quality–risk integration reveals spatial management priorities in rapidly urbanizing high-biomass regions

Rapid urbanization in high-biomass regions has intensified conflicts between ecological conservation and urban expansion, yet conventional assessments may overlook areas where high ecological quality coexists with elevated landscape structural risk. However, the long-term spatial mismatches between ecological quality and landscape ecological risk, together with their nonlinear associations with natural and anthropogenic factors, remain insufficiently understood in rapidly urbanizing high-biomass regions. To address these gaps, this study integrated an improved Remote Sensing Ecological Index (IRSEI) with a Landscape Ecological Risk (LER) model to evaluate ecological quality–risk relationships across Guangdong Province, China. To improve ecological assessment in densely vegetated areas, IRSEI replaced NDVI with the kernel normalized difference vegetation index (kNDVI) and incorporated the abundance index (AI) and net primary productivity (NPP) to represent habitat-support conditions and ecosystem productivity. Standardized IRSEI and LER values were combined through quadrant partitioning, and extreme gradient boosting coupled with SHapley Additive exPlanations (XGBoost–SHAP) was used to examine nonlinear relationships with natural and anthropogenic factors. From 2001 to 2022, ecological quality remained generally stable with a slight upward trend, whereas landscape ecological risk showed substantial spatial changes. The integrated assessment revealed persistent quality–risk mismatches, particularly in areas with relatively high values of both indicators. These Priority Regulation Zones expanded from 6.36% to 8.14% of the study area, suggesting that landscapes with relatively high ecological quality but structural vulnerability may be underestimated by quality-based assessments alone. SHAP results indicated that topographic factors had the strongest associations with both ecological quality and landscape ecological risk, while anthropogenic factors showed nonlinear and stage-dependent relationships. The IRSEI–LER framework thus provides a practical basis for differentiated ecological management in rapidly urbanizing subtropical high-biomass regions.