An ecosystem-based analysis of urban sustainability by integrating ecosystem service bundles and socio-economic-environmental conditions in China

Urban areas are human-natural systems sustained by fundamental ecosystem services (ESs), which interact in complex ways with the dynamic conditions of cities. To enhance our understanding of urban sustainability, it is important to investigate how the four types of ES (provisioning, regulating, cultural, and supporting services) are integrated with social, economic, and environmental conditions at the city scale. By adopting the concept of ecosystem service bundles (ESBs), this study investigated how the selected ES indicators were correlated with multiple ESs and clusters of conditions from 1995 to 2015 in Ordos, which is a rapidly growing city in northern China. Pearson's parametric correlation test was used to demonstrate the trade-offs and synergies of ESs. Self -organizing maps were used to delineate ESBs and the condition clusters. Overlap analysis demonstrated the spatial relationships of the six obtained ESBs to reveal the co-occurring clusters of conditions. The provisioning services exhibited trade-offs with regulating and supporting services. Synergies exhibited relationships with regulating, cultural, and supporting services. Out of the six acquired ESBs, the provisioning bundles increased from three to four between 2005 and 2015. In comparison, the regulating bundles decreased from three to one between 1995 and 2015. The extent to which the environmental cluster overlapped with ESBs declined from 63.82% to 50.02% between 2005 and 2015. In comparison, the overlap between the social cluster and ESBs increased from 13.25% to 49.98% during the same period. These two clusters represented the main stressors of multiple ESs in the city. Thus, coal mining, solid waste recycling, and pollution control should be integrated in future decision making. The integrative analysis presented here is expected to help improve the current un-derstanding of urban sustainability by combining spatiotemporal variations in ESBs with the socio-economic-environmental conditions of cities.