Construction of Urban Agglomeration Ecological Network Based on Multi‐Scale and Multi‐Method: A Case Study of Yangtze River Delta Urban Agglomeration, China

Abstract

Ecological networks (ENs) are vital for maintaining regional ecological security and preserving biodiversity. While various methods exist for constructing ENs, their effectiveness across different spatial scales, particularly in urban agglomerations, has not been thoroughly investigated. This study focuses on the Yangtze River Delta urban agglomeration (YRDUA), constructing ENs at three scales: urban agglomeration, metropolitan area, and city. Two methods were employed at each step, and the outcomes were evaluated and ranked using specific indicators. The results indicate: (1) For ecological source identification, the spatial distribution of ecological sources identified by different methods is consistent at the same scale, with the number of ecological sources identified at the three scales being around 600, 140, and 160, respectively. (2) For resistance surface construction, although there are differences between the two methods, the final resistance value shows relatively small changes. (3) Regarding the number of corridors, the quantities of the three scales are around 1470, 380, and 410, respectively. For specific indicators, α values of the three scales are around 0.71, 0.85, and 0.81, respectively; β values are around 2.42, 2.68, and 2.61, respectively; γ values are around 0.81, 0.90, and 0.88, respectively; Cr values are around 0.88, 0.80, and 0.68, respectively. Comparing and ranking all indicators can yield: at the urban agglomeration scale, ecological sources identified by the MSPA method, resistance surfaces constructed by Spatial Principal Component Analysis (SPCA), and corridors extracted by Linkage Mapper yielded optimal results. At the metropolitan and city scales, ecological sources identified by the MSPA method, resistance surfaces constructed by the Analytical Hierarchy Process (AHP), and corridors extracted by Graphab yielded optimal results. These findings provide methodological guidance for constructing ENs across different scales and offer new insights for landscape planning at multiple levels.