The PHYTOBS-MARCOBOLO dataset comprises long-term time series on marine microphytoplankton, from 2003 to 2021, along the entire French metropolitan coastline. Microphytoplankton data cover microscopic taxonomic identifications and counts. The PHYTOBS-MARCOBOLO dataset corresponds to a dataset extracted from the PHYTOBS network (DOI:10.17882/85178). The PHYTOBS network provides the scientific community and stakeholders with validated and qualified data on the biomass, abundance and composition of marine microphytoplankton in coastal and lagoon waters, with the aim of supporting scientific research. PHYTOBS-MARCOBOLO is a dataset used as part of the Horizon Europe MARCO-BOLO project (https://marcobolo-project.eu/), in which we are currently working to understand and analyze multi-decadal trends in coastal and marine biodiversity on a European scale. The PHYTOBS-MARCOBOLO dataset gathers data from 18 sampling sites, selected from the PHYTOBS-Network dataset according to requirements of time series quality and geographical location of sampling sites established as part of the MARCO-COLO project. This dataset was also formatted according to a template imposed for the European project.

Understanding divergence in the highly dispersive and seemingly homogeneous pelagic environment for organisms living as free drifters in the water column remains a challenge. Here, we analyzed the transcriptome wide mRNA sequences, as well as the morphology of 18 strains of Alexandrium minutum, a dinoflagellate responsible for Harmful Algal Blooms worldwide, to investigate the functional bases of a divergence event. Analysis of the joint site frequency spectrum (JSFS) pointed toward an ancestral divergence in complete isolations followed by a secondary contact resulting in gene flow between the two diverging groups, but heterogeneous across sites. The sites displaying fixed SNPs were associated with a highly restricted gene flow and a strong over-representation of non-synonymous polymorphism, suggesting the importance of selective pressures as drivers of the divergence. The most divergent transcripts were homologs to genes involved in calcium/potassium fluxes across the membrane, calcium transduction signal and saxitoxin production. The implication of these results in terms of ecological divergence and build-up of reproductive isolation are discussed. Dinoflagellates are especially difficult to study in the field at the ecological level due to their small size and the dynamic nature of their natural environment, but also at the genomic level due to their huge and complex genome and the absence of closely related model organism. This study illustrates the possibility to identify traits of primary importance in ecology and evolution starting from high throughput sequencing data, even for such organisms.