Researchers from the MARBEC Unit in Montpellier, France, led by Nicolas Loiseau and Nicolas Mouquet, published a study in PLOS Biology predicting that 12.7% of marine teleost fish species are at risk of extinction. This is a significant increase from the International Union for Conservation of Nature’s prior estimate of 2.5%. The researchers included nearly 5,000 species that had not received an IUCN conservation status due to insufficient data, highlighting the need for more research and conservation efforts.
The IUCN Red List of Threatened Species currently tracks over 150,000 species to guide global conservation efforts for the most threatened species. However, 38% of marine fish species, or 4,992 species, are considered Data-Deficient and do not receive an official conservation status or associated protections. To address this issue, Loiseau and colleagues used a machine learning model and artificial neural network to predict the extinction risks of Data-Deficient species based on occurrence data, biological traits, taxonomy, and human uses from 13,195 species.
The researchers categorized 78.5% of the 4,992 species as either Non-Threatened or Threatened, which includes Critically Endangered, Endangered, and Vulnerable categories. Predicted Threatened species increased fivefold, from 334 to 1,671, while predicted Non-Threatened species increased by a third, from 7,869 to 10,451. The characteristics of predicted Threatened species included small geographic range, large body size, low growth rate, and correlation with shallow habitats. Hotspots for predicted Threatened species included the South China Sea, Philippine and Celebes Seas, and the west coasts of Australia and North America, suggesting a need for increased research and conservation efforts in these areas.
The study also noted a change in conservation priority ranking after predicting extinction risks for species. The researchers recommend prioritizing the Pacific Islands and Southern Hemisphere’s polar and subpolar regions to account for emerging at-risk species. Many Data-Deficient species were found in the Coral Triangle, indicating the need for additional research in that region. While models cannot replace direct evaluations of at-risk species, AI provides a unique opportunity for rapid, extensive, and cost-effective evaluation of extinction risk for species.
Nicolas Loiseau emphasizes the importance of AI in assessing extinction risks for species that have not yet been evaluated by the IUCN. The analysis of 13,195 marine fish species revealed a significantly higher extinction risk than initially estimated by the IUCN, rising from 2.5% to 12.7%. The researchers propose incorporating recent advancements in forecasting species extinction risks into a new synthetic index called ‘predicted IUCN status,’ which can complement the current ‘measured IUCN status.’ This approach could help guide conservation efforts more effectively by identifying species at risk and prioritizing areas for research and protection.