Environmental DNA Approach for Lake Phytoplankton Monitoring

The European Water Framework Directive (WFD) requires monitoring to achieve or maintain good ecological status of lakes, which support major socio-economic services. Traditionally, a morphology-based identification of individuals is used for the biomonitoring. Environmental DNA (eDNA) has significantly enhanced the evaluation of biodiversity in aquatic ecosystems. In addition, the eDNA metabarcoding approach without taxonomic-assignment (the so-called taxonomy-free approach) seems to be more promising for biomonitoring than the taxonomy-based approach. For the moment, there is a lack of empirical evidence showing the relevance of the taxonomy-free approach for assessing the ecological status of lakes based on phytoplankton, a key biological group used under the WFD. In this study, we coupled an eDNA metabarcoding approach with a niche model and an unsupervised machine learning method to assess the ecological status of 239 samples collected in 98 lakes across France. In practice, we have developed a biotic index using the abundance profiles of Amplicon Sequence Variants (ASVs) along a specific environmental gradient to assess the latter. We showed that this index, based on the taxonomy-free approach, had a strong ability to predict the ecological status of samples along an anthropogenic gradient, representing eutrophication levels.