Collaborative platforms for biodiversity observation, coupled with taxonomic databases, now allow to automatically generate species checklists and/or distribution atlases at various temporal and geographical scales. We assess here the reliability of such automatically generated species checklist using non-marine molluscs from the Grand-Est region as a model. The assessed primary list was generated from the extraction filter of platform of the SINP (Système d’Information sur la Nature et les Paysages) and its associated taxonomic framework TAXREF. This primary list was compared with the taxonomical list independently elaborated by the authors of the present article. We show that the recovery rate between the primary list, citing 233 terminal taxa, and our list is about 90%. The differences correspond to 60.7% of false-absences, 32.1% of false-presences and 7.2% of distinct taxonomic treatments. Finally, we document the presence of 241 terminal taxa including 156 terrestrial gastropods, 46 aquatic gastropods and 39 bivalves. We show that reliable identification requires a combined morpho-anatomical approach for at least 40% of molluscs species. This functional ckecklist has, for some taxa, focused on the specific criteria considered as stable and on the level of identification difficulty, in order to homogenize the species occurrence data and/or their a posteriori evaluation. We recommend that primary data recorded in collaborative databases refer to specimens deposited in public collections, the specific characters to be observed, and the transmission of good quality photograph(s).
Collaborative data, Validation of primary biodiversity data, SINP-Information System on Nature and Landscapes.