4.1 Standardized eDNA sampling at a large scale
For the first time here, we have implemented eDNA surveys at a large scale, with standardized sampling protocols and analyses. Our results are therefore comparable, both in space and time: it is now possible to reproduce our sampling in the future to monitor the evolution of bivalve populations. This is particularly relevant for monitoring invasive species: eDNA analyses are usually used to detect a given species’ occurrence. But it can also be used to argue for absence. Our large-scale eDNA sampling revealed the occurrence of invasive species in their previously known range and also in locations they were expected to colonize considering the potential colonization routes (i. e. mainly artificial channels connecting main watersheds, eg. Prié & Fruget, 2017). Contrarily, eDNA samples confirm the absence of the invasive species where they were not (yet) expected. Re-sampling these places regularly could allow documenting colonization. Our results also confirm that some introduced species are not invasive. For example, the Long Fingernail Clam Sphaerium transversum (Say, 1829) has been recorded only twice in France with traditional methods, and eDNA only added one isolated data.
On the other hand, large-scale eDNA sampling was valuable to detect rare and endangered species. The False Orb Pea Mussel was detected in only two sites in France, supporting its critical conservation status. Only one new location could be added by our eDNA samples for the very well-known Giant Freshwater Pearl Mussel (Prié et al., 2018). Furthermore, some species such as the Depressed River Mussel, the Nut Orb Mussel, and above all the Witham Orb Mussel were rarely detected despite their supposed large distribution range in France, which flashes a red light on their conservation status.