The spatial pattern, trade-off and synergy relationship, and driving mechanism of ecosystem services in Chengdu under multi-model fusion

Against the background of rapid urbanization, the intensity of human activities has continued to increase, significantly reshaping land-use patterns and ecological processes, and consequently leading to a series of ecological security issues, including habitat degradation, carbon storage loss, and intensified soil erosion. Taking Chengdu City as the study area, this study integrates multi-source spatiotemporal data from 2000 to 2020 and applies the InVEST model, the XGBoost machine learning algorithm, SHAP-based explainable analysis, and principal component–based indirect gradient analysis to systematically investigate the spatial patterns, temporal dynamics, and driving mechanisms of four key ecosystem services: habitat quality, carbon storage, water yield, and soil retention. The results indicate that ecosystem services in Chengdu exhibit a pronounced “high-in-the-west and low-in-the-east” spatial differentiation pattern. Habitat quality, carbon storage, and soil retention show strong spatial coupling, with high-value areas mainly concentrated in the western mountainous regions, while low-value areas are predominantly distributed in the eastern plains and central urban areas. In contrast, the spatial pattern of water yield is relatively fragmented and is primarily controlled by precipitation conditions. The indirect gradient analysis reveals significant synergies among habitat quality, carbon storage, and soil retention, whereas water yield generally shows trade-offs with these regulating ecosystem services. Further XGBoost-SHAP analysis demonstrates that natural topographic factors (elevation and slope) dominate the variation in all four ecosystem services, precipitation is the key controlling factor for water yield, and human activity indicators such as population density, GDP, and nighttime light intensity exert significant negative effects on habitat quality and carbon storage. These findings provide a scientific basis for understanding the synergy–trade-off relationships among ecosystem services and their nonlinear driving mechanisms under rapid urbanization.