Key Expoitable Results (KERs)

Ocean warming can affect plankton both directly, through altered metabolic activities, and indirectly, modifying the physical–chemical properties of the water column, with possible effects on ecosystem functioning. To evaluate the combined action of warming-related physiological responses and environmental changes on plankton functioning, we carried out a long-term analysis (from 1994 to 2019) of the Bermuda Atlantic Time-series Study (BATS) dataset where ocean warming and stratification have driven a decrease in the net primary production over the last decade. Using the time series of plankton observations, we assembled 1000 replicates of a food web model for each year. We observed that the total flow of matter through the model remained constant over time, despite the increased oligotrophication, due to global warming, after 2014. In fact, the plankton food web remained robust through re-modulated trophic interactions with an increased detritivory to herbivory ratio of the food web over time. However, it was problematic to re-establish the trophic connections of the food web broken by ocean warming, as remarked by the increased relative internal ascendency. Thanks to trophic plasticity, the reduced zooplankton dependence on herbivory was compensated by a significant increase in the reliance on carnivory and detritivores, highlighting the crucial role of trophic interactions in buffering significant environmental short-term changes.

Marine plankton are key drivers of ocean productivity and global carbon cycling, yet their quantitative biogeography remains poorly characterized. Environmental genomic datasets are inherently compositional, restricting analyses to relative abundances and limiting their integration into ecological and biogeochemical models. Here we combine DNA mass measurements, filtered seawater volumes, and metagenomic relative abundances to generate absolute estimates of cell concentrations and carbon biomass of plankton across the global ocean. Leveraging thousands of samples, this approach provides quantitative estimates for hundreds of eukaryotic and thousands of prokaryotic environmental genomes, unveiling ecological associations not captured by compositional data. Using the psbO marker gene, we reconstruct quantitative biogeographies of photosynthetic lineages and, through global-scale modeling, project their distributions to generate quantitative maps of phytoplankton communities across the world’s ocean. By bridging genomic data with biogeochemical metrics, this study provides novel resources for integrating plankton at genomic resolution into next-generation biogeochemical models.