Spatio-temporal patterns and driving mechanisms of ecosystem services in mountainous regions: A multi-scale analysis of the Yanshan-Taihang mountain area

Investigating the spatio-temporal dynamics of ecosystem services and their underlying driving factors in mountainous regions is of critical importance for ensuring regional ecological security and promoting sustainable development. Focusing on the Yanshan-Taihang Mountain area as a case study, this research applied the InVEST model to evaluate the temporal trends of four key ecosystem services-carbon storage, habitat quality, water yield, and soil conservation-between 2000 and 2023 across multiple spatial scales (1 km, 5 km, 10 km, and 15km). The study further analyzed the spatial distribution characteristics and trade-off/synergy relationships among these services. To identify key driving factors, the XGBoost-SHAP model was employed. Key findings include: (1) From 2000 to 2023, all four ecosystem services exhibited significant spatial heterogeneity. Carbon storage showed a steady upward trend, with a concentric pattern of higher values in central areas and lower values at the periphery. Habitat quality continuously declined, although relatively high levels were maintained in the central Taihang Mountains and the Yanshan region. Water yield and soil conservation capacity both followed an initial increasing and subsequent decreasing trend. Water yield displayed a south-to-north decreasing gradient, while high soil conservation capacity was predominantly concentrated in the southern part of the study area. (2) Spatial statistical analysis revealed significant clustering of high-value areas for all four services. As the spatial scale increased, the patterns of hot and cold spots became more spatially homogeneous. (3) In terms of interactions, habitat quality, carbon storage, and soil conservation demonstrated strong synergistic relationships, whereas water yield exhibited trade-off effects with the other services. These trade-off/synergy relationships were scale-dependent; while the direction remained consistent across scales, the intensity varied significantly. (4) Driving factor analysis indicated that slope exerted a substantial influence on all four services. Population density primarily affected carbon storage, habitat quality, and water yield. NDVI had a significant impact on carbon storage and soil conservation, while annual precipitation was a key determinant of water yield and soil conservation. The relationships between ecosystem services and their driving factors also varied with the spatial scale of analysis. These findings provide a scientific basis for optimizing the spatial configuration of ecosystem services and formulating targeted ecological conservation strategies in the region.