A study dealing with the status assigned to species by the International Union for Conservation of Nature (IUCN) was published in the open-access archive arXiv with research input from MAMBO partners. The co-authors in question are Pierre Bonnet from The French Agricultural Research Centre for International Development (CIRAD) and Alexis Joly from the National Institute for Research in Digital Science and Technology (INRIA).

The paper explores a new approach to classify species' IUCN status by using the broad predictive capabilities of deep learning-based distribution models. This complements the Union’s efforts to measure the extinction risk via new automated techniques offering data in cases of underevaluated taxa status. Furthermore, the article points to a new method which uses flexible features from species distribution models that highlight environmental preferences.

As climate change modifies the spread of populations, it is crucial to evaluate the increased extinction risk. In order to do so, scientists use Species Distribution Models (SDMs) to estimate future outcomes. The projections are converted into inputs for the aforementioned new IUCN classification method. 

The results show that the number of threatened species is increasing worldwide, especially in Africa, Asia and South America. Moreover, the highest proportion of threatened species is projected to occur near the Tropics, around the Equator, in lowland areas and at elevations between 800 and 1,500 meters.

Modelling plant species distribution is crucial to MAMBO’s exploration into innovative approaches to biodiversity monitoring. MAMBO’s online library allows for access to more insights from this article and the project’s numerous other outputs, as well as relevant publications beyond the project.