Mitigating ecological connectivity loss: A comprehensive "service-habitat-network" approach to ecological network design in the Qilian Mountains

Intensifying climate change and persistent human disturbance have triggered severe habitat fragmentation and biodiversity loss worldwide, among which the decline of ecological connectivity acts as the core driving force. Traditional ecological network construction predominantly depends on Morphological Spatial Pattern Analysis (MSPA) or subjective/equal-weighted superposition of ecosystem services, which cannot capture the nonlinear synergies and trade-offs among multiple services nor integrate the habitat requirements of key species. Meanwhile, single-objective corridor optimization fails to balance ecological integrity, biodiversity conservation and practical feasibility. To address these limitations, this study takes the Qilian Mountains as the research area and puts forward a Service-Habitat-Network framework. This research integrated InVEST, scSE and MaxEnt models to identify ecological sources through dual-space coupling, extracted ecological corridors via circuit theory, and conducted multi-objective optimization using the NSGA-Ⅱ algorithm. Results indicate that ecological sources show an obvious east-clustered and west-scattered pattern, covering 36.08% of the total study area. The optimized network comprises 763 corridors, with 224 ecological barriers and 160 pinch points identified, and is classified into optimal and sub-optimal corridors. This framework promotes the transformation of ecological governance from static pattern mapping to adaptive management, offering scientific support for the conservation and restoration of fragile alpine mountain ecosystems.