Tuesday , December 1 2020

With artificial intelligence, you can predict which plants will disappear

Y International Union for Nature Conservation (IUCN, for its acronym in English) Work continuously when making a denomination Red List, which analyze the species and distribute them that vary from minor anxiety & # 39; I disappear, without passing through intermediate categories.

These classes are used to define conservation actions, so they are of "huge utility", but "for the species to be included they must be individually evaluated, ask for a prescribed protocol, available funds and the presence of experts who carry out the evaluation, making it a slow process, "he said. Telam Anahí Espindola, co-author of the study of Argentina.

"The basic approach we used was a random forest, known about its ability to distribute and predict data," he says, explaining "in this case, we are trying to predict the The likelihood of a species being endangered, using data related to the characteristics of its range of distribution, from climatic and preferred preferences and some morphological features ".

Espíndola, an entomologist professor at the University of Maryland in the United States, explained that "this method allows us to use all the species that have already been evaluated by IUCN to train and create our forest distribution randomly, using the species's features as predecessor variables. "

"Once a sufficiently detailed distribution model has been obtained, we can use that same model on species that we know the features used in the model (range of distribution, climatic conditions and morphology and favored) but we do not know risk level disappears. "

In this sense, co-author of the article published in the specialist magazine PNAS He said that using and distributing these data will allow "Calculate the likelihood that those species that have not yet been assessed by the IUCN Red List are at risk."

This "very useful" system is that it is "quite accurate, and can also be analyzed without accessing important computer resources", it also has the "advantage" of being based solely on "public data "(open access), it says that anyone can perform these analyzes and use their results.

"In addition, this method can be modified to any geographical or taxonomic interest, as it can also be used at national, regional or local levels, and allows species identification that should be evaluated with priority by IUCN, "he said. Spend it and described this tool as "useful and complementary to these evaluations".

That expert noted Of the 150,000 species analyzed, "about 10% (15,000) have a high similarity of conservation categories with the exception of minor concern.

"From a global perspective, we identify regions with higher probability of threatened species, such as some of the Andean regions of northern North America, or the Brazilian Atlantic forest." These regions are characterized by a high level of endemism, and of the presence of many rare species, "it ended.

Source link