Exploring habitat patch clusters based on network community detection to identify restored priority areas of ecological networks in urban areas

Abstract

Ecological connectivity is the foundation of maintaining urban biodiversity and ecosystem health. Identifying and managing ecological (connectivity) networks can help maintain the stability of urban ecosystems. However, few studies have explored the cluster effect in the ecological network caused by the imbalance in connectivity strength between habitat patches, which is not conducive to the in-depth restoration of ecological networks. In the present study, a typical urban area, Shenzhen, was used as an example to analyze the important habitats in the city based on the focal species and to identify an ecological network. Habitat patch clusters in the ecological network were explored based on random walk network community detection. These are clusters of closely connected habitat lands. Finally, we analyzed existing urban policies for the protection of clusters and the points to be repaired in the network. The results showed that 50 ecological corridors connected 39 habitats in the study area, which further formed seven habitat patch clusters. Most of the clusters were well-protected by existing policies. Nineteen barrier points were identified between the clusters, and their restoration helped strengthen the connectivity between clusters. This study provides a reference for future urban ecological restoration.