Introduction
Marine protists encompass a large and heterogeneous community representing the majority of the eukaryotic diversity in the oceans (Worden et al. 2015). They include primary producers (autotrophs), heterotrophs (phagotrophs and parasitic), and a substantial collection of lineages exhibiting varying degrees of mixotrophic strategies (Mitra et al. 2016). These groups occupy distinct niches in the marine food web, playing essential roles in biogeochemical cycles (Mitra et al. 2014; Worden et al.2015). They are responsible for ~50% of annual planktonic photosynthetic primary productivity (PP), of which they consume ~66%, plus an additional 10% of bacterial PP (Calbet & Landry 2004; Steinberg & Landry 2017). Recent applications of molecular tools such as metabarcoding (Choi et al. 2020; de Vargas et al. 2015), metagenomics and single cell genomics (Latorre et al. 2021), have significantly sharpened our understanding of protist diversity, distributions, and functionality, from basic trophic modes to complex metabolic pathways, and emphasized their importance in channeling marine productivity to upper trophic levels. The available observational data is rich in horizontal spatial and temporal coverage, yet lacks vertical resolution, particularly below the photic zone (Ollison et al. 2021). There, a large and diverse community of heterotrophic protists thrives on sinking particulate matter, preys upon the prokaryotic populations (Ollison et al.2021; Rocke et al. 2015) and removes a similar percentage of the prokaryotic standing stock compared to the epipelagic realm (Rockeet al. 2015). This trophic transfer may represent a critical mechanism sustaining the upper levels of mesopelagic trophic webs.
The warm oligotrophic subtropical gyres of the major ocean basins are the largest biome in the planet, and these nutrient-poor regions are expanding in size (Irwin & Oliver 2009). To predict future conditions, it is essential to characterize their present state (Agusti et al. 2019). In these regions, PP is dominated by the prokaryotic and eukaryotic picophytoplankton (Agusti et al. 2019; Cotti-Rauschet al. 2020; Orsi et al. 2018; Riemann et al.2011). Much of the PP of the larger size fractions depends on mixotrophic strategies, combining autotrophy with phagotrophy on the smaller primary producers (Mitra et al. 2016). The Bermuda Atlantic Time-series Study (BATS) is located in the western limits of the North Atlantic subtropical gyre (Lomas et al. 2013). The hydrography at BATS responds to a locally large seasonal cycle in atmospheric forcing, reflected in winter mixing that extends below the photic layer and contrasts sharply with a stratified summertime photic zone that is progressively mixed away during the fall. These characteristics drive nutrient availability, primary production and community composition (Church et al. 2013). Particle fluxes (Conte et al. 2001) and zooplankton community composition (Blanco-Bercial 2020) show a seasonal signal. Likewise, metazooplankton exhibit signals that reflect seasonal timescales, and it is expected that the protist community also exhibit seasonality, at least in the epipelagic layers.
A significant hurdle to studying the marine protist community is the complex and tedious procedure needed to characterize these organisms. Historically, this was achieved by microscopy, and required expert personnel for sample preparation, analyses and taxonomic identification. Microscopy or image-based methods have the advantage of being quantitative, providing an estimate of cell sizes, and enabling biomass proxies. These methods are limited, as any other morphology-based analyses, by the presence of cryptic species, and the natural complexity of the diversity of protists. In the oligotrophic ocean, a large proportion of the protist community falls within the 2-20 µm size class, many of them naked, which makes the identification very complicated when using automated optical instruments (e.g. FlowCam, IFCB), or even with classical methods after fixation. Molecular techniques as metabarcoding (Ollison et al. 2021) and metagenomics (Sunagawa et al.2015) allow for direct characterization of planktonic communities (Notet al. 2007; Treusch et al. 2012; Treusch et al.2009) and its potential functionality. These methods provide a better understanding of the diversity of the protist community in the oceans (Caron & Hu 2019) and the associated trophic relationships, including the large presence of symbionts (Ollison et al. 2021).
The present study applies DNA metabarcoding to samples collected over three years in conjunction with the BATS time-series to assess marine protist communities in the epipelagic and mesopelagic zones. Its purpose is to describe the vertical and temporal distributions of community composition, their evolution and transitions over the annual cycle, and relationships to hydrographic structure. We aimed to characterize the distinct communities with depth and detail how their boundaries evolve and transition, as a response to the hydrographic features throughout the year. In lieu of traditional approaches that bin data in depth layers, and seasons by calendar months, we employ a physical framework defined by four seasons and 11 vertical layers that delineate dynamical zones shaped by locally varying ocean mixing, stratification, and light penetration in the Sargasso Sea (Table 1; Curry et al., in prep)