Machine learning shows more reptile species may be at risk of extinction than previously thought

Potamites montanicola, classified as “Critically Endangered” by the automated assessment method and as “Data Deficient” by the IUCN Red List of Threatened Species. Credit: Germán Chávez, Wikimedia Commons (CC-BY 3.0)

The machine learning tool estimates the risk of extinction of species whose conservation was previously not a priority.

Species threatened with extinction are identified in the iconic Red List of Threatened Species, published by the International Union for Conservation of Nature (IUCN). A new study presents a new machine-learning tool to assess extinction risk, then uses that tool to show that reptile species that go unlisted due to a lack of assessment or data are more likely to be threatened than the assessed species. The study, by Gabriel Henrique de Oliveira Caetano of Ben-Gurion University of the Negev, Israel, and colleagues, was published on May 26.e in the review PLOS Biology.

The IUCN Red List of Threatened Species is the most comprehensive assessment of species extinction risk and informs conservation policy and practice worldwide. However, the process of species categorization is time-consuming, laborious, and subject to bias, relying heavily on manual curation by human experts. Consequently, many animal species have not been assessed or lack sufficient data, which creates gaps in protection measures.

To assess 4,369 reptile species that previously could not be prioritized for conservation and develop accurate methods to assess the risk of extinction of obscure species, these scientists created a machine-learning computer model. The model assigned IUCN extinction risk categories to the 40% of the world’s reptiles that lacked published assessments or were classified as “DD” (“Data Deficient”) at the time of the study. The researchers validated the model precisionby comparing it to the Red List risk categorizations.

The authors found that the number of threatened species is much higher than that indicated in the IUCN Red List and that the non-assessed (“not assessed” or “NE”) reptiles and those with insufficient data were more likely to be threatened than assessed species. Future studies are needed to better understand the specific factors underlying the extinction risk of threatened reptile taxa, to obtain better data on obscure reptile taxa, and to create conservation plans that include newly identified threatened species. .

According to the authors, “Overall, our models predict that the conservation status of reptiles is much worse than currently estimated, and that immediate action is needed to avoid loss of reptile biodiversity. Regions and taxa that we have identified as likely to be at greater risk should receive increased attention in further assessments and conservation planning. Finally, the method we present here can be easily implemented to help fill the assessment gap on other lesser-known taxa”.

Co-author Shai Meiri adds, “It is important to note that the additional reptile species identified as threatened by our models are not randomly distributed across the world or the reptilian evolutionary tree. Our supplemental information highlights that there are more reptile species at risk – particularly in Australia, Madagascar and the Amazon Basin – all of which have high reptile diversity and should be targeted for additional conservation efforts. Additionally, species-rich groups, such as geckos and elapids (cobras, mambas, coral snakes, and others), are likely to be at greater risk than currently highlighted in the Global Reptile Assessment, these groups should also be considered. subject to greater conservation attention.

Co-author Uri Roll adds: “Our work could be very important in helping global efforts to prioritize the conservation of species at risk – for example using the IUCN Red List mechanism. Our world is facing a biodiversity crisis and severe human-caused changes to ecosystems and species, but funds for conservation are very limited. Therefore, it is essential that we use these limited funds where they could bring the most benefit. Advanced tools, such as the ones we used here, along with the accumulation of data, could significantly reduce the time and cost needed to assess extinction risk, and thus pave the way for more informed decision-making. in conservation.

Reference: “Automated assessment reveals that reptile extinction risk is vastly underestimated across space and phylogeny” by Gabriel Henrique de Oliveira Caetano, David G. Chapple, Richard Grenyer, Tal Raz, Jonathan Rosenblatt, Reid Tingley, Monika Böhm, Shai Meiri and Uri Roll. May 26, 2022, PLOS Biology.
DOI: 10.1371/journal.pbio.3001544

Sherry J. Basler