Spatial Identification for Ecological Restoration: Integrating Ecosystem Resilience and Networks

Ecosystem resilience is fundamental to sustainable development in high-density urbanized areas, which face mounting pressures from global climate change and intensified human activities. Ecological restoration projects are vital for strengthening ecosystem resilience in both natural and engineered systems, thereby safeguarding ecological integrity. However, existing methodologies for identifying ecological restoration areas often fail to capture ecosystem complexity and uncertainty, particularly the functional resilience required under dynamic disturbances. This study proposes a novel identification approach that integrates ecosystem resilience assessment with ecological network optimization to pinpoint restoration priorities and clarify restoration objectives. Resilience was assessed using a 5-dimensional quality-importance-connectivity-stability-sustainability framework, while the ecological network provided a spatial quantification pathway. The proposed approach was validated through a case study in the Guangdong–Hong Kong–Macao Greater Bay Area. The results showed that urgent restoration priorities covered 8,408.00 km 2 in the northwestern region, characterized by extensive forest areas but limited water availability, along with 616.85 km of ecological corridors and 25 key points in the central and southeastern areas requiring ecosystem service improvement. The core challenges to strengthening ecosystem resilience include connectivity fractures, critical node vulnerabilities, and ecological islanding effects. This approach effectively reveals latent ecological risks, such as connectivity fractures and critical node vulnerabilities, which are objectively present but difficult to visualize. It also identifies areas with high vegetation coverage but low ecological function. Overall, this approach overcomes the limitations of traditional hierarchical resilience frameworks, which typically fragment ecosystem structure and function.