Multi-tiered objective ecological restoration planning in the Yangtze River Basin

Abstract

Biodiversity conservation and climate change mitigation are critical global environmental goals that drive large-scale ecological restoration efforts. However, the strong emphasis on these global targets in ecological restoration planning (ERP) may inadvertently diminish attention to local needs. To this end, we developed a multi-tiered objective ERP framework centered around hierarchical optimization model, designed to identify restoration priority areas that maximize cost-effectiveness across multiple benefits. Taking the Yangtze River Basin (YRB) as a case study, we identified two-tiered objectives: global (biodiversity conservation and climate change mitigation) and local (water yield and soil conservation). First, potential restorable areas and restoration reference ecosystems were identified using a land use/land cover transition matrix and historically-based reference strategies. Next, taking catchment basins as planning units, we quantified and mapped the potential benefits and costs of restoration. Subsequently, a hierarchical optimization model was employed to prioritize catchment basins for restoration across four scenarios, designed by all possible priority orderings within a fixed two-tiered structure in which global objectives were always prioritized over local ones. Finally, shared priority areas across scenarios were identified as optimal restoration areas. Approximately 1.65 million hectares of degraded land were identified as optimal restoration areas, distributed across 48.9 % (N = 6739) catchment basins. This study emphasizes the importance of reconciling multi-objectives and demonstrates how hierarchically organizing them in regional-scale EPR can assist in identifying cost-effective restoration priority areas to enhance both global and local benefits. Our planning framework provides a practical reference for resource allocation aimed at optimizing cost-effectiveness across multiple environmental challenges.