Objective River ecosystems are increasingly threatened by habitat degradation and heightened flood risks due to rapid urbanization and the intensification of extreme climate events. Restoring river floodplains, as a nature-based solution (NBS), offers a promising approach to address these challenges by providing multiple benefits, including flood control, ecological restoration, and recreational opportunities. However, identifying floodplains with high restoration suitability remains a critical scientific challenge in basin management. This research aims to develop a comprehensive framework for evaluating the suitability of river floodplains restoration, with the goal of maximizing social and ecological benefits at the basin scale. By focusing on the integration of hydrological, ecological, and socio-economic factors, this research seeks to provide a scientifically robust method for prioritizing floodplain restoration efforts.

Methods This research proposes a multi-dimensional floodplain restoration suitability evaluation framework, integrating Geographic Information Systems (GIS) and the HEC-RAS (Hydrologic Engineering Center’s River Analysis System) hydrodynamic model. The framework incorporates multi-source data, including digital elevation models (DEM), land use classifications, vegetation indices (NDVI), soil types, and socio-economic factors, to evaluate the restoration potential of floodplains. This research uses a multi-level hierarchical screening method to select indicators at three scales: basin, sub-basin and site. Through multi-scale and multi-dimensional synergistic evaluation, the research avoids not being able to conduct global analysis due to insufficient data, and also prevents ignoring the overall restoration effect by limiting to the local area. Key indicators such as hydrological conditions, ecological habitats, water quality, vegetation cover, and socio-economic factors are identified and quantified. The entropy weight method is employed to calculate the restoration suitability index, ensuring an objective weighting of each indicator. Finally, region-specific socio-economic factors are incorporated into the evaluation system for the final selection of restorable floodplains. Hydrodynamic simulations are conducted using HEC-RAS to model baseline and post-restoration scenarios. These simulations help evaluate changes in flood behavior and habitat conditions under various flow events. A case study is conducted in the middle and lower reaches of the Liuxi River Basin in Guangzhou, China, to validate the framework. The research area is selected due to its susceptibility to frequent flooding, rich biodiversity, and strategic importance in regional greenway planning.

Results The results indicate that the middle and lower reaches of the Liuxi River Basin exhibit high floodplain restoration suitability. Nine floodplains are identified as highly suitable for restoration, located predominantly in low-slope riparian zones with minimal human development pressure. The post-restoration simulations demonstrate significant improvements in hydrological regulation, with a 7.7% reduction in flood peak flow, and smoother flow change curves. Additionally, the restoration efforts have led to a substantial increase in suitable habitats for indicator species (e.g., Spinibarbus hollandi, a local fish species), with an expansion of 56.52 hm² under one-year flood conditions. The research also highlights the importance of socio-economic factors, such as proximity to greenways and tourist attractions, for determining the feasibility and sustainability of restoration projects. The integration of these factors ensure that the selected floodplains not only provide ecological benefits but also align with local development goals.

Conclusion This research presents a comprehensive framework for evaluating floodplain restoration suitability that integrates hydrological modeling, ecological evaluation, and socio-economic analysis. The case study in the Liuxi River Basin demonstrates that the restored floodplains can significantly enhance flood defense capabilities and ecological functions, while also supporting recreational and socio-economic needs at multiple spatial scales. The integration of HEC-RAS simulations with GIS-based spatial data enables precise identification of high-priority areas, while the entropy weight method ensures that indicator weighting remains data-driven and objective. The framework not only enhances the scientific basis for restoration decision-making but also narrows the gap between ecological science and land-use planning. Importantly, it emphasizes the need to balance ecological objectives with socio-economic considerations, especially in densely populated and ecologically sensitive regions. As such, the methodology developed in this research may offer valuable insights for practitioners and policymakers seeking to implement nature-based solutions for river basin management. The framework is adaptable to other river systems facing similar challenges, and contributes to the broader discourse on sustainable urban water management, climate adaptation, and ecosystem resilience. Ultimately, the research underscores that successful floodplain restoration requires a holistic, interdisciplinary approach rooted in both science and local context.