Results of the 6th Nordic Network
The results of the 6th NorBAF (Nordic Network – Benthic Algae in Freshwater) diatom exercise are now published (Kahlert et al. 2021). With our actions, we NorBAF participants aim to harmonize and improve sample treatment and diatom identification, to make people aware of identification problems, to share knowledge and experience, and also to spread the fun of identifying diatoms! We are by no means perfect, and our work and agreements, taxa tables and results of diatom names, should be seen in the light of improving the harmonization of diatom identification of the Nordic countries with the focus on environmental assessment. The 6th NorBAF exercise in 2020 had 27 participants from eight countries (Estonia, Finland, Germany, Iceland, Lithuania, Sweden, The Netherlands, United Kingdom) representing 18 institutions, companies, universities and agencies. Three samples were prepared and counted, and the results were evaluated in terms of taxonomic similarity and index comparability. A digital workshop was held to discuss the results. All participants were encouraged to contribute with discussions, images, questions and disagreements.
We made five important findings on how to improve the process, both for ourselves and as recommendations for others. First, not all dissimilarities in diversity and indices were caused by disagreements about species identification, but instead by not following the instructions and the standard list of taxa. This was a reminder that it is necessary to harmonize taxa identification of both beginners and experts at least if the results will be used together with the results of other analysts, otherwise taxa lists will not be comparable. Second, there were also some suspicious results probably generated by incorrect calibration of scale bars. Third, we found for some participants a suspicious high share of the quite large taxon Tabellaria flocculosa, and a low share of smaller taxa such as small Naviculoides. We discussed that maybe some beginners had counted girdle bands of T. flocculosa inflating those counts, and decreasing the share of smaller taxa. Fourth, we also stated that it is necessary to work with sufficient optical technique, and much patience, when going through the slide to count the smaller and often delicate taxa appropriately. Finally, underestimation of the small Naviculoides led to an overestimation of the index values, i.e. seemingly better water quality, because many of those taxa have low index values indicating pollution and eutrophication.
In summary, a clear cost-effective recommendation of our exercises is to always double-check the identification of the most abundant taxa in a slide. Those are making the difference in diversity and index calculation (when using indices which include abundance). Another recommendation is to identify rare taxa, especially singletons, to genus level, or at least indicate some form of uncertainty, e.g. add a “c.f.” to avoid the risk of force fitting. The detailed reasons for differing taxa lists and index values were manifold. Identification differences were mainly devoted to Gomphonema, small Naviculoides, and Fragilaria. Additionally, participants showed differences in the number of counted valves of certain taxa. The differences in deformation counts were partly devoted to participants who were using this tool for the first time, and we hope to improve harmonization here in the future.