Résumé
Global biodiversity in freshwater and the oceans is declining at high
rates. Reliable tools for assessing and monitoring aquatic biodiversity,
especially for rare and secretive species, are important for efficient and
timely management. Recent advances in DNA sequencing have provided a new
tool for species detection from DNA present into the environment. In this
study, we tested if an environmental DNA (eDNA) metabarcoding approach,
using water samples, can be used for addressing significant questions in
ecology and conservation. Two key aquatic vertebrate groups were targeted:
amphibians and bony fish. The reliability of this method was cautiously
validated in silico, in vitro, and in situ. When compared with traditional
surveys or historical data, eDNA metabarcoding showed a much better
detection probability overall. For amphibians, the detection probability
with eDNA metabarcoding was 0.97 (CI = 0.90-0.99) versus 0.58 (CI =
0.50-0.63) for traditional surveys. For fish, in 89% of the studied sites,
the number of taxa detected using the eDNA metabarcoding approach was
higher or identical to the number detected using traditional methods. We
argue that the proposed DNA-based approach has the potential to become the
next-generation tool for ecological studies and standardized biodiversity
monitoring in a wide range of aquatic ecosystems.