Abstract
As the world strives to meet the ambitious targets of the Kunming-Montreal Global Biodiversity Framework, our research highlights a critical gap: while climate policy is driven by robust, forward-looking projections, biodiversity action remains largely reactive, relying on retrospective indicators. We advocate for the widespread adoption of Biodiversity Model Intercomparison Projects (BMIPs) to provide standardized, transparent, and defensible projections needed to support global policy. By applying lessons from decades of climate modeling, we outline how BMIPs can move beyond simple correlations to embrace mechanistic approaches that capture essential eco-evolutionary processes like dispersal and adaptation. Our paper describes how BMIPs can empower countries to test “what-if” scenarios, ensuring that national conservation actions are truly sufficient to halt and reverse biodiversity loss. We propose best practices for the next generation of biodiversity science, emphasizing open-access data infrastructure, historical benchmarking, and inclusive global governance. This work serves as a roadmap for the scientific community and policymakers to collaborate on a more predictive and effective approach to nature conservation.
