Coupling relationships among natural resources, socioeconomic activity, and spatial morphology at the grid scale: a case study of Shanghai

Abstract This study investigates how natural-resource value, socioeconomic activity, and spatial morphology interact at the grid scale in Shanghai, with particular attention to functional differentiation, non-linear transitions, and spatial heterogeneity. Multi-source data were harmonised to a 400 m grid, and composite indicators were constructed through factor analysis and aggregation of normalised indicators. Natural-break classification, unsupervised machine-learning clustering, and Lee’s L analysis identified spatial patterns, whilst bidirectional XGBoost models interpreted with SHAP examined relationships between natural-resource value and socioeconomic activity. Aggregate activity and ground coverage show the strongest negative associations with natural-resource value, but technology and public services remain positively associated with natural-resource value, with parallel tendencies for culture, tourism, and finance. The associations of young-worker concentration and building height with natural-resource value are U-shaped. Gas and hydrological regulation shift from positive to negative associations with socioeconomic activity as provision intensifies, distinguishing ecological landscapes embedded in the built-up area from regional ecological service systems. The three-way coupling is asymmetric, non-linear, and spatially differentiated. Its planning significance lies in distinguishing land-intensive growth from knowledge- and service-oriented renewal, and urban ecological landscapes from regional ecological service systems. Super-large and mega cities should therefore coordinate load and land-occupation control, functional and vertical renewal, and the distinct urban and regional roles of ecological resources.