Skip to main content

Improving biodiversity protection through artificial intelligence

Over a million species face extinction, highlighting the urgent need for conservation policies that maximize the protection of biodiversity to sustain its manifold contributions to people’s lives. 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.

Conservation of birds in fragmented landscapes requires protected areas

For successful conservation of biodiversity, it is vital to know whether protected areas in increasingly fragmented landscapes effectively safeguard species. However, how large habitat fragments must be, and what level of protection is required to sustain species, remains poorly known. We compiled a global dataset on almost 2000 bird species in 741 forest fragments varying in size and protection status, and show that protection is associated with higher bird occurrence, especially for threatened species. Protection becomes increasingly effective with increasing size of forest fragments.

Safeguarding Seafood Security, Marine Biodiversity and Threatened Species: Can We Have Our Fish and Eat It too?

The ocean contains an abundance of biodiversity that is vital to global food security. However, marine biodiversity is declining. Marine protected areas and marine reserves have been used to protect biodiversity, conserve threatened species and rebuild exploited species, but are perceived as restrictive to fishing, which has slowed progress towards ocean protection targets. Here, we perform a spatial prioritisation of the ocean to protect biodiversity, threatened species and food security.

Predicting the optimal amount of time to spend learning before designating protected habitat for threatened species

Deciding when to protect threatened species habitat when complete knowledge about the habitat extent is uncertain is a common problem in conservation. More accurate habitat mapping improves conservation outcomes once that habitat is protected. However, delaying protection to improve accuracy can lead to species decline or, at worst, local extinction when threats to that habitat continue unabated before protection is implemented. Hence, there is a trade-off between gaining knowledge and taking conservation action.

Scheduling incremental actions to build a comprehensive national protected area network for Papua New Guinea

Systematic conservation planning identifies priority areas to cost-effectively meet conservation targets. Yet, these tools rarely guide wholesale declaration of reserve systems in a single time step due to financial and implementation constraints. Rather, incremental scheduling of actions to progressively build reserve networks is required. To ensure this incremental action is guided by the original plan, and thus builds a reserve network that meets all conservation targets, strategic scheduling, and iterative planning is needed.

Minimizing cross-realm threats from land-use change: A national-scale conservation framework connecting land, freshwater and marine systems

There is a growing recognition that conservation strategies should be designed accounting for cross-realm connections, such as freshwater connections to land and sea, to ensure effectiveness of marine spatial protection and minimize perverse outcomes of changing land-use. Yet, examples of integration across realms are relatively scarce, with most targeting priorities in a single realm, such as marine or freshwater, while minimizing threats originating in terrestrial ecosystems.