Linking Historical Dynamics, Spatial Explanatory Factors, and Cross-Scenario Consistency of Ecosystem Service Value in Giant Panda National Park

Ecosystem services in national parks are sensitive to land use/land cover (LULC) change, yet their long-term dynamics, spatial explanatory factors, and potential changes under future are rarely evaluated together at the scale of an entire park. This limitation is evident in the Giant Panda National Park (GPNP), where previous assessments of ecosystem service value (ESV) have largely focused on selected regions. This study therefore assessed park-wise ESV dynamics in the GPNP from 2005 to 2025 using the equivalent factor method. The Optimal Parameter-based Geographical Detector (OPGD) was used to distinguish the explanatory factors associated with annual ESV density patterns from those associated with long-term ESV changes. Future ESV dynamics under three SSP–RCP scenarios for 2030 and 2050 were further evaluated, together with cross-scenario consistency in spatial changes. Forests consistently covered more than 84% of the park, while water bodies, grasslands, and croplands declined. Total ESV decreased by −1.07% from 2005 to 2025, indicating that gains associated with forest expansion did not fully offset losses from other land-use types, particularly water bodies. Annual ESV density patterns were mainly associated with topography, vegetation productivity, and climate, whereas long-term ESV changes showed the strongest spatial associations with initial elevation and changes in land-use intensity. Interactions among factors generally increased their explanatory power. Scenario-based estimates indicated potential increases in total ESV under all three future scenarios, mainly due to projected forest expansion. Spatially, consistent ESV gains and relatively stable areas dominated, although localized scenario-sensitive areas showed greater dependence on alternative development pathways. These findings highlight the need to protect high-value forest ecosystems while strengthening conservation of vulnerable water bodies, grasslands, and shrublands, with adaptive monitoring focused on park margins and scenario-sensitive areas. By integrating historical change, differentiated explanatory analyses, and cross-scenario future assessment, this study provides a framework for ecosystem conservation and spatial management in national parks.