Grant Agreement: 101082021Project Acronym: MARCO-BOLOProject Title: MARine COastal BiOdiversity Long-term ObservationsDeliverable Number: D2.3Work Package Number: WP2Deliverable Title: Report on the congruence between traditional and eDNA-based biodiversity observations, and their robustness across bioinformatics issues and diversity metricsDue Date: 30.11.2025Submission Date: 28.11.2025 Here, we present the outcomes of Task 2.3: “Comparison of spatial and temporal eDNA-based vs traditional observations”. The report summarises the level of agreement between traditional and eDNA-based biodiversity observations and evaluates how stable these comparisons remain when different analytical decisions are applied. Performance (or congruence) refers to the extent to which eDNA observations reproduce the ecological patterns detected by traditional methods, including seasonal cycles, interannual trends and diversity gradients. Robustness refers to the consistency of these eDNA-derived patterns when analytical choices vary, for example the choice of bioinformatic workflow or the use of different diversity metrics. The work was conducted in conjunction with Tasks 2.1 and 2.2, building on the meta-analysis of eDNA-based approaches (D2.1) and the long-term time-series datasets curated under D2.2. Three main analyses were performed: (A) comparison of complementary plankton datasets spanning 20 years (2001–2021) at coastal station ‘L4’ (Western English Channel) between eDNA (18S-V9 metabarcoding) and microscopy counts; (B) assessment of phytoplankton diversity values from different metabarcoding methods and microscopy in the global Tara Oceans dataset; and (C) a ‘Data Analysis Challenge’ inviting the global eDNA community to run their preferred pipelines on a shared SOMLIT-Astan dataset, allowing assessment of pipeline-driven variation in diversity outputs. Across the three analyses, the degree of congruence between eDNA (metabarcoding) and traditional (microscopy) datasets was most dependent on the diversity metric applied; both the methods employed in data production and processing, and the nature of the community being observed, further impacted similarity. Long-term seasonal diversity patterns were more congruent between data types than interannual changes. Estimators that accounted for relative abundance (e.g. Shannon; Hill numbers q = 1.5) produced more congruent patterns than richness-only metrics. Designation of taxa into major plankton/phytoplankton groups revealed group-dependent differences in diversity patterns between data types. Pipeline choice was more important than season or year to the community structure of eDNA samples in the Data Analysis Challenge results. These findings indicate that, despite innate differences between data types, there is partial congruence in seasonal diversity patterns and total diversity values obtained from eDNA and traditional microscopy data — supporting integration of eDNA-based approaches into established marine plankton monitoring frameworks, particularly when using evenness-aware diversity metrics.
Warwick-Dugdale et al. (Fri,) studied this question.