A multi-scale framework for understanding spatial scale effects on ecosystem service heterogeneity, interactions, drivers and their socio-ecological impact pathways for adaptive management

Abstract

Given the hierarchical complexity of ecosystems, understanding the interactions and mechanisms influencing ecosystem services (ESs) across multiple scales is essential for effective sustainable management. This study proposed a multi-scale analytical framework that identified spatial scale characteristics, summarized patterns, analyzed causes, and provided management suggestions. Focusing on the Yangtze River Delta, the study quantified five typical ESs at grid, county, and city scales, identifying their characteristics through spatial autocorrelation, trade-offs, synergies, and ES bundles. XGBoost and random forest models were used to identify dominant socio-ecological factors at different scales, while the partial least squares-structural equation modeling (PLS-SEM) revealed the impact pathways of these drivers and their direct and indirect effects on ESs. Key findings included: (1) Strong spatial autocorrelation was observed across all ESs, with carbon sequestration (CS) demonstrating the greatest scale stability and food supply (FS) exhibiting the highest variability; (2) Trade-offs were concentrated in northern cultivated areas, while synergies thrived in southwestern mountains, with synergies increasing and trade-offs decreasing as spatial scale expands (3) Natural factors (e.g., climate, geography, vegetation); primarily influenced ESs, but socio-economic and landscape factors increasingly shaped FS and CS at larger scales; (4) Geographic conditions had the strongest positive impact, climate effects diminished with increasing scale, and urbanization, particularly the proportion of built-up land, negatively influenced ESs through its impact on vegetation cover. These insights underscore the need for tailored, scale-specific strategies to support sustainable ecosystem management and urban development.