Publications
Following publications have been announced by our department Biological Carbon Pump. For further information please contact Dr. Klas Ove Möller, co-author of the publications:
Orenstein, E.C., Ayata, S.-D., Maps, F., Becker, É.C., Benedetti, F., Biard, T., de Garidel-Thoron, T., Ellen, J.S., Ferrario, F., Giering, S.L.C., Guy-Haim, T., Hoebeke, L., Iversen, M.H., Kiørboe, T., Lalonde, J.-F., Lana, A., Laviale, M., Lombard, F., Lorimer, T., Martini, S., Meyer, A., Möller, K.O., Niehoff, B., Ohman, M.D., Pradalier, C., Romagnan, J.-B., Schröder, S.-M., Sonnet, V., Sosik, H.M., Stemmann, L.S., Stock, M., Terbiyik-Kurt, T., Valcárcel-Pérez, N., Vilgrain, L., Wacquet, G., Waite, A.M., & Irisson, J.-O. (2022): Machine learning techniques to characterize functional traits of plankton from image data. Limnol Oceanogr., doi:10.1002/lno.12101
Abstract:
Plankton imaging systems supported by automated classification and analysis have improved ecologists’ ability to observe aquatic ecosystems. Today, we are on the cusp of reliably tracking plankton populations with a suite of lab-based and in situ tools, collecting imaging data at unprecedentedly fine spatial and temporal scales. But these data have potential well beyond examining the abundances of different taxa; the individual images themselves contain a wealth of information on functional traits. Here, we outline traits that could be measured from image data, suggest machine learning and computer vision approaches to extract functional trait information from the images, and discuss promising avenues for novel studies. The approaches we discuss are data agnostic and are broadly applicable to imagery of other aquatic or terrestrial organisms.
Martin-Cabrera, P., Perez Perez, R., Irisson, J.-O., Lombard, F., Möller, K.O., Rühl, S., Creach, V., Lindh, M., Stemmann, L., & Schepers, L. (2022): Best practices and recommendations for plankton imaging data management: Ensuring effective data flow towards European data infrastructures. Ostend, Belgium, Flanders Marine Institute, 31pp, doi:10.25607/OBP-1742
Abstract:
The best practices and recommendations for plankton imaging data management enable users to report a detailed taxonomic characterisation of plankton observations as well as quantitative information that is useful for ecological studies. This format allows biodiversity data portals to extend their scope beyond species occurrence data. Furthermore, proposing the use of more Darwin Core fields in this format, users now have the possibility to publish manually validated datasets, but also datasets produced by fully automated plankton identification workflows. The proposed data and file formats are simple and both human- and machine-readable to automatise workflows. This format will allow data generators to submit enriched plankton imaging datasets to the international biodiversity data portals, (Eur)OBIS and EMODnet Biology. We encourage plankton imaging data generators to implement these workflows into their pipelines, to share their data with the international data portals easily, enriching these databases with this valuable data.
