Machine learning and underwater cameras to predict the global distribution of zooplankton

As part of an international collaboration, a team of researchers from the Villefranche sur Mer Oceanography Laboratory (LOV, Sorbonne University/CNRS) has gathered a huge dataset on zooplankton acquired by underwater cameras at the global scale, the analysis of which made it possible to model the oceanic composition and biomass of zooplankton. The results of this study entitled “Global Distribution of Zooplankton Biomass Estimated by In Situ Imaging and Machine Learning” have been published in Marine Science Frontiers in August.

Plankton is made up of a wide variety of organisms, ranging from bacteria to vertebrates, including protists and algae, living adrift in the aquatic and marine compartments of the terrestrial globe and playing a central role in the trophic chains. marine and geochemical cycles. These organisms, with a great taxonomic diversity and whose size ranges from a few micrometers to several meters, produce 50% of carbon dioxide, allow carbon sequestration and are at the base of the marine food chain.

Quickly quantifying their diversity and their biomass in the oceans is essential, satellites, such as Seasat launched in 1978, can observe autotrophic plankton on a global scale, but they can only do so in the first few meters of the surface. Expedition vessels, on the other hand, can obtain data in depth but locally. In addition, the processing of samples brought on board is often long and tedious. The overall biomass of zooplankton is thus very poorly known.

The study

For this study, the team gathered a huge dataset of zooplankton acquired by underwater cameras on a global scale. Its analysis, carried out as part of Laetitia Drago’s ongoing thesis at LOV, has made it possible to model the composition and oceanic biomass of zooplankton.

The study predicted, for the first time, the global biomass distribution of 19 zooplankton taxa with a spherical diameter ranging from 1 to 50 mm, using observations made with the Underwater Vision Profiler 5, a quantitative imaging instrument on the site. After classifying 466,872 organisms from more than 3,549 profiles (0-500 m) obtained between 2008 and 2019 around the world, the researchers estimated their individual biovolumes and converted them to biomass using conversion factors taxon specific.

They then associated these biomass estimates with climatologies of environmental variables (temperature, salinity, oxygen, etc.), in order to build habitat models using boosted regression trees.

Study results

The results reveal maximum values ​​of zooplankton biomass around 60° N and 55° S as well as minimum values ​​around the large rotating marine currents, the oceanic gyres, in particular in the southern hemisphere.

The lowest biomass values ​​were predicted north of 80°N and in the Weddell Sea, researchers also observed an increase in predicted biomass around the equator. The highest biomass values ​​were predicted between 50 and 80°N, in the coastal waters of the Labrador Sea and Baffin Bay, and in the Greenland Sea. Relatively high biomass was predicted around these locations as well as in the Gulf of Alaska, Bering Sea and Sea of ​​Okhotsk. A band of high biomass was predicted between 40 and 50°S, a region associated with the Arctic Polar Front.

The overall integrated biomass (0-500 m) was estimated at 0.403 PgC. It was largely dominated by Copepoda (35.7%, mainly in the polar regions), followed by Eumalacostraca (26.6%) Rhizaria (16.4%, mainly in the intertropical convergence zone).

The machine learning approach used here is sensitive to the size of the training set and generates reliable predictions for abundant groups such as Copepoda (R2 ≈ 20-66%) but not for rare groups (Ctenophora, Cnidaria, R2 < 5%). Nonetheless, this study offers a first protocol for estimating spatially-resolved global zooplankton biomass and community composition from in situ imagery observations of individual organisms.

The underlying dataset covers a 10-year period, while approaches that rely on net samples use datasets collected since the 1960s. Obtain basin-scale zooplankton biomass distribution estimates on a global scale in shorter timeframes in the future.

Sources of the article:

Global distribution of zooplankton biomass estimated by in situ imagery and machine learning »


  • Laetitia Drago, Thelma Panaïotis, Jean-Olivier Irisson, Lionel Guidi, Marc Picheral, Lars Stemman and Rainer Kiko from Sorbonne University, Villefranche-sur-mer Oceanography Laboratory;
  • Marcel Babin of the Takuvik International Research Laboratory, Océan Québec, Université Laval (Canada) – National Center for Scientific Research (CNRS), Department of Biology and Québec-Océan, Université Laval, QC, Canada;
  • Tristan Biard from the Laboratory of Oceanology and Geosciences (LOG), Univ. Littoral Côte d’Opale, Univ. Lille, National Center for Scientific Research (CNRS), UMR 8187, Wimereux, France;
  • François Carlotti from the Marine Ecology and BIOdiversity Department (EMBIO), MIO Mediterranean Institute of Oceanology Mediterranean Building, Marseille, and the Physical and Biological Oceanography Laboratory (LOPB), case 901 13288, Marseille;
  • Laurent Coppola of Sorbonne University, Oceanography Laboratory of Villefranche-sur-mer and Sorbonne University, National Center for Scientific Research (CNRS), OSU STAMAR, Paris;
  • Leep Karp-Boss of the School of Marine Sciences, University of Maine, Orono, ME, USA;
  • Fabien Lombard of Sorbonne University, Oceanography Laboratory of Villefranche-sur-mer, and the Institut Universitaire de France (UITA), Paris;
  • Andrew MP McDonnell of the Department of Oceanography, University of Alaska Fairbanks, Fairbanks, AK, USA;
  • Andreas Rogge of the Benthopelagic Processes Section, Alfred Wegener Institute Helmholtz Center for Polar and Marine Research, Bremerhaven, Germany;
  • Anya Mary Waite from the Ocean Frontier Institute and Oceanography Department, Dalhousie University, Halifax, NS, Canada;
  • Helena Haus you Department of Ocean Ecosystem Biology, GEOMAR Helmholtz Center for Ocean Research Kiel, Kiel, Germany;

Marine Science Frontiers, 09 Aug 2022 Sect. Ocean watching
DOI: 10.3389/fmars.2022.894372

Similar Posts

Leave a Reply

Your email address will not be published. Required fields are marked *