Rebuilding ecological networks: A multi-scale approach to global restoration

Species interactions are fundamental to biodiversity maintenance and ecosystem functioning, yet their restoration remains a complex and evolving challenge. While integration of interaction networks into restoration ecology is gaining momentum, practical frameworks to address this complexity at scale are still needed. We present a multi-scale, interaction-based framework to support restoration of ecological networks and associated functions amid urgent biodiversity and ecosystem crises. Building on meta-networks as a foundation, the framework addresses three core dimensions: (1) Where to restore—by identifying spatial priorities through projected network attributes under current or future conditions; (2) What to restore—by assessing the potential for ecosystem function and resilience through trait and network attributes; and (3) How to restore—by selecting species based on their topological and functional roles within interaction networks. The proposed approach builds on the importance of bridging scales ––from upscaling sparse interaction data to regional contexts, to refining site-specific priorities. We highlight key considerations for integrating interactions into restoration planning, including interaction and functional homogenization risk and the expected cascading effects from interconnected networks, which must be accounted for to avoid unintended trade-offs. This framework advances restoration science by aligning ecological theory, data, and tools to support the recovery of diverse, resilient ecosystems and the services they provide to people.