Software for prioritizing conservation actions based on probabilistic information

Improving biodiversity protection through artificial intelligence

Over a million species face extinction, urging the need for conservation policies that maximize the protection of biodiversity to sustain its manifold contributions to people. Here we present a novel framework for spatial conservation prioritization based on reinforcement learning that consistently outperforms available state-of-the-art software using simulated and empirical data. Our methodology, CAPTAIN (Conservation Area Prioritization Through Artificial INtelligence), quantifies the trade-off between the costs and benefits of area and biodiversity protection, allowing the exploration of multiple biodiversity metrics. Under a limited budget, our model protects substantially more species from extinction than areas selected randomly or naively (such as based on species richness). CAPTAIN achieves substantially better solutions with empirical data than alternative software, meeting conservation targets more reliably and generating more interpretable prioritization maps. Regular biodiversity monitoring, even with a degree of inaccuracy characteristic of citizen science surveys, substantially improves biodiversity outcomes. Artificial intelligence holds great promise for improving the conservation and sustainable use of biological and ecosystem values in a rapidly changing and resourcelimited world.

Improving biodiversity protection through artificial intelligence

Over a million species face extinction, urging the need for conservation policies that maximize the protection of biodiversity to sustain its manifold contributions to people. Here we present a novel framework for spatial conservation prioritization based on reinforcement learning that consistently outperforms available state-of-the-art software using simulated and empirical data. Our methodology, CAPTAIN (Conservation Area Prioritization Through Artificial INtelligence), quantifies the trade-off between the costs and benefits of area and biodiversity protection, allowing the exploration of multiple biodiversity metrics. Under a limited budget, our model protects substantially more species from extinction than areas selected randomly or naively (such as based on species richness). CAPTAIN achieves substantially better solutions with empirical data than alternative software, meeting conservation targets more reliably and generating more interpretable prioritization maps. Regular biodiversity monitoring, even with a degree of inaccuracy characteristic of citizen science surveys, substantially improves biodiversity outcomes. Artificial intelligence holds great promise for improving the conservation and sustainable use of biological and ecosystem values in a rapidly changing and resourcelimited world.

Phylogeny-based conservation priorities for Australian freshwater fishes

Conservation scientists are increasingly interested in the question of how extinction prunes the tree of life. This question is particularly important for Australian freshwater fishes because there is a broad mix of ∼300 old and young species, many of which are severely threatened. We used a complete species-level phylogeny of Australian freshwater fishes to examine phylogenetic nonrandomness of extinction risk. We computed the potential loss of phylogenetic diversity by simulating extinction across the tree under a pattern weighted based on International Union for Conservation of Nature extinction risk category and compared this loss to projected diversity loss under a random null model of extinction. Finally, we calculated EDGE (evolutionary distinctiveness, global endangerment) scores for 251 freshwater and 60 brackish species and compiled a list of high-priority species for conservation actions based on their extinction risk and evolutionary uniqueness. Extinction risk was not random and was clustered in both diversity cradles (recently diversifying, species-rich clades, such as Galaxiidae and Percichthyidae) and museums (older, species-poor groups, such as freshwater chondrichthyans). Clustered extinction made little difference to the average expected loss of phylogenetic diversity. However, the upper bound of loss was higher under a selective model of extinction, particularly when the counts of species lost were low. Thus, the loss of highly threatened species would diminish the tree of life more than a null model of randomly distributed extinction.  High priority species included both widely recognized and charismatic ones, such as the Queensland lungfish (Neoceratodus forsteri), river sharks, and freshwater sawfishes, and lesser-known species that receive less public attention, including the salamanderfish (Lepidogalaxias salamandroides), cave gudgeons, and many galaxiids, rainbowfishes, and pygmy perches.