On the urban compactness to ecosystem services in a rapidly urbanising metropolitan area: Highlighting scale effects and spatial non–stationary

Understanding how urban compactness is linked to ecosystem services is crucial for sustainable spatial planning and allows decision–makers to better assess the strengths and weaknesses of compactness options. Yet it is still unknown whether urban compactness is conducive to supporting various ecosystem services. In this study, urban compactness is conceptualised into physical space, social space, urban function and social activity organisation. Drawing on a case study of the Wuhan metropolitan area in China, we aim to address the impacts of urban compactness on ecosystem services (food production, carbon storage, habitat quality and recreation service) utilising ordinary least squares, geographically weighted regression and multiscale geographically weighted regression models at the grid scale. The results indicate that urban compactness has the highest negative correlation with food production, followed by carbon storage and habitat quality. In contrast, urban compactness is positively correlated with recreation service. In relational interpretation, the geographically weighted regression model effectively identifies spatial heterogeneity and improves the explanatory power compared to ordinary least squares. Furthermore, the multiscale geographically weighted regression model with different bandwidths as the effective method reveals local relationships between urban compactness and ecosystem services, as well as spatially non–stationary associations, which are particularly critical in the investigation of complex and heterogeneous urban landscapes. The contributions of different urban compactness factors to ecosystem services vary considerably. For example, the impacts of building density, information entropy of land use structure, land development intensity and public service density on ecosystem services are on a global scale. In contrast, the bandwidth of the percentage of ecological land in total land is smaller, indicating a higher level of spatial heterogeneity in its impact. These results contribute to decision–making criteria for sustainable urban development policies and urban ecological landscape management.