Objective Influenced by natural environmental changes and human activities, habitat fragmentation has become a widespread phenomenon in protected areas. Constructing ecological corridors is an effective measure to mitigate habitat fragmentation. However, current research on ecological corridors often focuses on urban areas and uses historical data for the spatial delineation of corridors at a single level, while there is a significant difference between national parks, which aggregate valuable ecological resources, and urban areas. Besides, the analysis of current research also reveals several challenges with respect to ecological corridor, such as failing to well adapt to future environmental changes, failing to correspond to the multi-level and multi-scale attributes of ecosystems, and failing to clarify the specific goals and measures for construction management and control. Therefore, it is of great significance to explore a ecological corridor system for national parks that adapts to future environmental changes, corresponds to the multi-level and multi-scale attributes of ecosystems, and implements differentiated construction management and control.

Methods This research proposes a ecological corridor system for national parks that considers future environmental changes, corresponds to the hierarchy of ecosystems, and implements differentiated construction management and control. The research establishes a framework for multi-level ecological corridor system construction based on future scenario simulation. Taking Nanling National Park as an example, this research first sets the overall goal of improving the connectivity of habitats within Nanling National Park and the spatial continuity of ecosystem service functions between Nanling National Park and peripheral areas outside the park at a larger scale. Based on the overall objective, the proposed levels are determined to be the ecosystem level corresponding to the region where the national park is located (regional level) and the community level corresponding to the park (park level). At the regional level, ecological source areas are identified based on the importance of ecosystem service functions, and resistance surfaces are created using factors that hinder the flow of ecological materials. Ecological corridors are extracted based on the circuit theory model. At the park level, ecological source areas are identified based on species habitat suitability. The method of future scenario simulation is introduced in the identification of ecological source areas, and the future environmental data are obtained through two climate models (ssp126 and ssp585) and related plans. Based on the current environmental data and future environmental data, the MaxEnt model is used to identify the ecological source areas. Next, the research uses the GeoDetector for spatial stratified heterogeneity analysis to identify the ecological background characteristics that affect the generation of ecological source areas. The width of ecological corridors is determined based on these characteristics, and ecological corridors are integrated and organized. Then, ecological stepping-stones are used to connect the ecological corridors at different levels. Finally, based on the ecological background characteristics, the landscape heterogeneity between the ecological corridors and the ecological source areas is analyzed, and based on the principle of reducing the difference between the ecological corridors and the ecological source areas, zoning is implemented for construction management and control of ecological corridors, and the specific contents of construction management and control are formulated.

Results In the ecological corridor system of Nanling National Park, 392 ecological source areas and 990 ecological corridors are identified at the regional level; 85 ecological source areas and 154 ecological corridors are identified at the park level. The system connects the regional level and the park level through 52 ecological corridors and 13 ecological stepping stones. Based on landscape heterogeneity between ecological source areas and corridors within the park, the research area is divided into three types of construction management and control zones, for which 12 specific measures for differentiated construction management and control are adopted. The specifics of construction management and control correspond to the forestry unit, which defines the spatial scope of the implementation of ecological corridors. Ultimately, the natural succession area covers 633 units with an area of 116.35 km², accounting for 37.45% of the total ecological corridor area; the artificially promoted natural succession area involves 431 units with an area of 70.15 km², accounting for 22.58% of the total ecological corridor area; and the artificially repaired area covers 1,203 units with an area of 124.21 km², accounting for 39.98% of the total ecological corridor area. According to statistics, more than half of the ecological corridor area entails the introduction of manual intervention into construction management and control, indicating the importance of manual intervention for ecological corridors.

Conclusion This research proposes a framework for constructing a multi-level ecological corridor system for national parks, and establishes such a system for Nanling National Park. The research simulates the distribution of ecological source areas under different future scenarios, delineates construction management and control zones, and formulates specific implementation measures. Based on this, the research concludes that establishing a multi-level ecological corridor system can enhance the ecological connectivity between national parks and their surrounding areas, improve the adaptability of ecological corridors to future environmental changes, effectively support the construction and implementation of ecological corridors, fully leverage the ecological radiation role of national parks, and further strengthen the central role of national parks in the protected area system.