Multi-species approaches are necessary to optimize conservation: integrating a genetic algorithm with stochastic models in the Murray–Darling Basin (Australia)

Summary Environmental water delivery is implemented globally to redress the ecological impacts of river regulation and consumptive water use. Allocation must balance the competing (and often conflicting) needs of humans, a diversity of water-dependent species and abiotic processes in the managed system. To optimize water allocation for conservation we developed a novel optimization framework that integrated a genetic algorithm with an integrated systems model to refine near-optimal allocations that maximize outcomes for a shrub ( Duma florulenta ), tree ( Eucalyptus camaldulensis ) and tree frog ( Ranoidea raniformis ) in one of the world’s most regulated catchments: the Murray–Darling Basin (Australia). Our algorithm weighted 16 wetland attributes and optimized those weights to prioritize allocation over a 25-year forecast window. The best-performing solutions required no additional resources but improved outcomes for the two priority plants and doubled the surviving frog populations. Modifying the multi-user model to only consider R. raniformis improved outcomes for that species at the expense of the two plant species, highlighting the risks of allocating a shared resource through a single-species lens. We demonstrate the value of optimization and multi-species approaches as decision-support tools for environmental water allocation and similar resource allocation problems in conservation.