Extensive misidentification of European ivy species (Hedera L.): How taxonomically reliable are online biodiversity databases?

Abstract

The global digitization effort and the popularization of citizen science is providing unprecedented amount of biodiversity online data. However, these databases can be severely affected by species misidentifications, whose validation may require access to specimens. This is often difficult, as most online records lack a physical voucher, and time-consuming in all cases. For this reason, the use of online databases to address biological questions has long been questioned by taxonomists. Ivies (Hedera L.) are an excellent case to evaluate this issue as they have tons of online records, whose use at the species level is questioned due to a high expected taxonomic uncertainty. Here we analyze the sources and extent of taxonomic errors in ivy species in Europe, the area of the world with the highest number of species. We estimate a high misidentification rate (18% on average), varying between species (maximized in H. iberica: 38%; H. azorica: 48%, H. hibernica: 55%) and regions (maximized in Spain: 27%, UK: 38%). We identify a systematic misidentification of all European ivies with H. helix and analyze the associated cascade error effects. We compile a spatial database to overcome the large discrepancies we observed in species distributions between online and morphologically reviewed records.