Optimizing Ecological Management in China: Insights from Chongqing’s Service Projections

Abstract

The assessment of ecosystem service (ES) supply–demand relationships is critical for addressing regional sustainable development challenges, yet systematic studies integrating spatial drivers analysis and multiscenario forecasting in rapidly urbanizing mountainous regions remain scarce. This study focuses on Chongqing as a representative case to investigate spatial patterns, driving mechanisms, and future trajectories of ES supply–demand dynamics. Through spatial quantification of four key ES (food provision, water retention, soil conservation, carbon fixation) and statistical analysis of socioeconomic datasets from 2010 to 2020, geographical weighted regression modeling was employed to identify spatially heterogeneous drivers. Long-term projections (2030–2060) were developed using climate–economy integrated scenarios reflecting different global development pathways. The results demonstrate three principal findings: First, while regional ecosystem quality maintains stable with an improved supply–demand ratio (0.260 to 0.320), persistent deficits in carbon fixation capacity require urgent attention. Second, spatial mismatches exhibit intensifying polarization, with expanding deficit zones concentrated in metropolitan cores and their periurban peripheries. Third, thermal-hydrological factors (aridity index, temperature) coupled with land intensification pressures emerge as dominant constraints on ES supply capacity. Scenario projections suggest coordinated climate mitigation and sustainable development strategies could maintain the supply–demand ratio at 0.189 by 2060, outperforming conventional development pathways by 23.5–41.2%. These findings provide spatial decision support frameworks for balancing ecological security and economic growth in mountainous megacities, with methodological implications for cross-scale ES governance in developing regions.

Keywords

• ecosystem services
• supply–demand
• spatial–temporal mismatch
• geographically weighted regression model
• SSP–RCP scenarios