Recent research indicates that 12.7% of marine teleost fish species are in danger of extinction, which is a dramatic increase from the previous estimate of 2.5% by the International Union for Conservation of Nature (IUCN). This study, led by Nicolas Loiseau and Nicolas Mouquet from the MARBEC Unit in Montpellier, France, published its findings on August 29th in the open-access journal PLOS Biology. The study examined nearly 5,000 species that were not assigned an IUCN conservation status due to a lack of data.
The IUCN’s Red List of Threatened Species monitors over 150,000 species to inform global conservation efforts for those in greatest peril. However, a significant 38% of marine fish species (around 4,992 species) fall into the Data-Deficient category, meaning they lack an official conservation status and the protections that come with it.
To enhance conservation strategies for these vulnerable species, Loiseau and his team utilized a combination of machine learning and artificial neural networks to assess the extinction risks of Data-Deficient species. They employed data on occurrences, biological characteristics, taxonomy, and human interactions from a total of 13,195 species for their models.
The researchers identified 78.5% of the 4,992 species as either Non-Threatened or Threatened, which includes the IUCN categories of Critically Endangered, Endangered, and Vulnerable. The number of species predicted to be Threatened surged from 334 to 1,671, while the number of predicted Non-Threatened species rose by a third, from 7,869 to 10,451.
Species categorized as Threatened often exhibited characteristics such as a limited geographic range, larger body sizes, and slower growth rates. The risk of extinction was also linked to species inhabiting shallow waters. Notable regions that emerged as hotspots for these predicted Threatened species included the South China Sea, the Philippine and Celebes Seas, as well as the western coasts of Australia and North America. The researchers advocate for heightened research and conservation initiatives in these critical areas.
There was a clear shift in conservation priority following the predictions based on species IUCN status, prompting the researchers to suggest that the Pacific Islands and the polar and subpolar regions of the Southern Hemisphere should receive heightened attention for emerging at-risk species. Additionally, many Data-Deficient species are located in the Coral Triangle, underscoring the need for more research in that region.
While the researchers acknowledge that these models cannot replace direct assessments of at-risk species, they emphasize that AI provides a valuable tool for conducting rapid, comprehensive, and cost-effective evaluations of extinction risks.
Loiseau remarked, “Artificial Intelligence (AI) allows for reliable assessments of extinction risks for species that have not been evaluated by the International Union for Conservation of Nature (IUCN). Our analysis of 13,195 marine fish species shows that the risk of extinction is significantly greater than previously estimated by the IUCN, shifting from 2.5% to 12.7%. We recommend implementing advancements in predicting species extinction risks into a new comprehensive index called ‘predicted IUCN status,’ which could effectively complement the existing ‘measured IUCN status.’
