PhD thesis. Asterisk: automated assessment of species extinction risk

Application summary

Biological diversity is eroding at an increasing pace and probably more species fall into extinction than are described every single day. Useful knowledge for the conservation of most species is extremely scarce and the current population trends for almost all of them are unknown. Threat assessment of species allows prioritizing conservation efforts. Yet, this process requires intensive human labour and less than 5% of described taxa are currently assessed by the International Union for the Conservation of Nature (IUCN). Advanced computational methods and the cooperation between ecologists and computer scientists are now growing fast and this project intends to take advantage of them. The revolutionary aim of this project is to create a computational workflow and tool - Asterisk – based on the latest data-mining, machine learning and artificial intelligence methods to automate the assessment of species extinction risk, a novel initiative with a global impact. It will include all steps of an assessment process, from (1) gathering distribution, trait and threat data from disparate sources, to (2) standardizing such data, (3) modelling the extinction risk of species, (4) outputting new data to an open-access database and (5) publishing assessments in journals; all in a single, supervised but largely automated, process.