Identifying global research and conservation priorities for Columbidae: a quantitative approach using random forest models.

International audienceThe family of Columbidae, including pigeons and doves, remains understudied despite their patrimonial value and high ecological and conservation relevance. Currently, 353 extant columbid species are listed in the IUCN red list, with about 20% of them being threatened with extinction. However, there has been little effort so far to synthetize the available information on factors influencing extinction risk and the allocation of research effort among columbid species. In this context, using random forest models, the present study aims at quantitatively assessing to what extent environmental, life history and socio-political factors may drive the extinction risk of pigeons and doves and explain differences in scientific attention among species. We found that high risk of extinction in columbids is associated with small historical range, exposure to invasive alien mammals and living in isolated islands and/or at low altitudes, while the probability of population decline is associated with species body size, surrounding human density and narrow habitat breadth. We also evidenced a large disparity between species or population extinction risk and scientific interest. Indeed, most of the studies on columbids have been conducted by scientists from North America and Western Europe on their local species, whereas species from biodiversity hotspots, which are more at risk of extinction, have comparatively received little attention. This unequal acquisition of knowledge creates gaps that deserve to be filled in order to have a good appreciation of extinction risk in columbids and associated threats, through fair transnational cooperation, academic training and regional coordination in conservation-oriented research on columbids