Description: The Arctic Ocean is home to a unique fauna that is disproportionately affected by global warming but that remains under-studied. Due to their high mobility and responsiveness to global warming, cephalopods and fishes are good indicators of the reshuffling of Arctic communities. Here, we established a nekton biodiversity baseline for the Fram Strait, the only deep connection between the North Atlantic and Arctic Ocean. Using universal primers for fishes (12S) and cephalopods (18S), we amplified environmental DNA (eDNA) from seawater (50–2700 m) and deep-sea sediment samples collected at the LTER HAUSGARTEN observatory. We detected 12 cephalopod and 31 fish taxa in the seawater and seven cephalopod and 28 fish taxa in the sediment, including the elusive Greenland shark (Somniosus microcephalus). Our data suggest three fish (Mallotus villosus, Thunnus sp., and Micromesistius poutassou) and one squid (Histioteuthis sp.) range expansions. The detection of eDNA of pelagic origin in the sediment also suggests that M. villosus, Arctozenus risso, and M. poutassou as well as gonatid squids are potential contributors to the carbon flux. Continuous nekton monitoring is needed to understand the ecosystem impacts of rapid warming in the Arctic and eDNA proves to be a suitable tool for this endeavor.

Description: Critical questions exist regarding the abundance and, especially, the export of picophytoplankton (≤2 µm diameter) in the Arctic. These organisms can dominate chlorophyll concentrations in Arctic regions, which are subject to rapid change. The picoeukaryotic prasinophyte Micromonas grows in polar environments and appears to constitute a large, but variable, proportion of the phytoplankton in these waters. Here, we analyze 81 samples from the upper 100 m of the water column from the Fram Strait collected over multiple years (2009–2015). We also analyze sediment trap samples to examine picophytoplankton contributions to export, using both 18S rRNA gene qPCR and V1-V2 16S rRNA Illumina amplicon sequencing to assess the Micromonas abundance within the broader diversity of photosynthetic eukaryotes based on the phylogenetic placement of plastid-derived 16S amplicons. The material sequenced from the sediment traps in July and September 2010 showed that 11.2 ± 12.4% of plastid-derived amplicons are from picoplanktonic prasinophyte algae and other green lineage (Viridiplantae) members. In the traps, Micromonas dominated (83.6% ± 21.3%) in terms of the overall relative abundance of Viridiplantae amplicons, specifically the species Micromonas polaris. Temporal variations in Micromonas abundances quantified by qPCR were also observed, with higher abundances in the late-July traps and deeper traps. In the photic zone samples, four prasinophyte classes were detected in the amplicon data, with Micromonas again being the dominant prasinophyte, based on the relative abundance (89.4% ± 8.0%), but with two species (M. polaris and M. commoda-like) present. The quantitative PCR assessments showed that the photic zone samples with higher Micromonas abundances (〉1000 gene copies per mL) had significantly lower standing stocks of phosphate and nitrate, and a shallower average depth (20 m) than those with fewer Micromonas. This study shows that despite their size, prasinophyte picophytoplankton are exported to the deep sea, and that Micromonas is particularly important within this size fraction in Arctic marine ecosystems.

Description: The long-term dynamics of microbial communities across geographic, hydrographic, and biogeochemical gradients in the Arctic Ocean are largely unknown. To address this, we annually sampled polar, mixed, and Atlantic water masses of the Fram Strait (2015–2019; 5–100 m depth) to assess microbiome composition, substrate concentrations, and oceanographic parameters. Longitude and water depth were the major determinants (~30%) of microbial community variability. Bacterial alpha diversity was highest in lower-photic polar waters. Community composition shifted from west to east, with the prevalence of, for example, Dadabacteriales and Thiotrichales in Arctic- and Atlantic-influenced waters, respectively. Concentrations of dissolved organic carbon peaked in the western, compared to carbohydrates in the chlorophyll-maximum of eastern Fram Strait. Interannual differences due to the time of sampling, which varied between early (June 2016/2018) and late (September 2019) phytoplankton bloom stages, illustrated that phytoplankton composition and resulting availability of labile substrates influence bacterial dynamics. We identified 10 species clusters with stable environmental correlations, representing signature populations of distinct ecosystem states. In context with published metagenomic evidence, our microbial-biogeochemical inventory of a key Arctic region establishes a benchmark to assess ecosystem dynamics and the imprint of climate change.

Description: Raw data acquired by two thermosalinographs (SBE21, SeaBird GmbH) on board RV Polarstern were processed to yield a calibrated and validated data set of temperature, conductivity and salinity during expedition PS121. Both sensors were equipped with a more accurate external temperature sensor (SBE38, Sea-Bird GmbH). Data were downloaded from the DAVIS SHIP data base (https://dship.awi.de) with a resolution of 1 sec. The raw hex data were converted to temperature and conductivity while a sensor drift correction was applied using calibration coefficients from before and after the expedition. Salinity was calculated according to the instructions from the Practical Salinity Scale PSS-78, using the obtained (internal) temperature and conductivity data and a pressure of 11 dbar which represents the water depth of the inlet of the TSG system on Polarstern. A speed filter of 0.5 knots minimum speed was applied. Processed data are provided as 10min means of seawater salinity, conductivity and temperature, aligned with position data taken from the master track of PS121 (doi:10.1594/PANGAEA.908157). Further details and evaluation of the data is outlined in the data processing report found at the EPIC repository under URL (https://hdl.handle.net/10013/epic.7fffb528-06bd-48ae-8489-cea0444c4eab).

Description: Water samples were collected from the long-term ecological research (LTER) site at Helgoland and the eukaryotic microbial community was assessed. 18S V4 region was amplified using the primer set 528iF /964iR and amplicon sequencing was performed on an Illumina MiSeq™ sequencer in a 2 × 300 bp paired-end run. Sequence data have been deposited in the European Nucleotide Archive (ENA) at the European Bioinformatics Institute (EMBL-EBI) under accession number PRJEB37135 using the data brokerage service of the German Federation for Biological Data (GFBio). 21 million sequences remained after bioinformatic processing and were clustered into Operational Taxonomic Units (OTUs). The Protist Ribosomal Reference database (PR2), version 4.11.1 was used as reference database.

Description: Water sampling was carried out weekly between April and October 2016. Water samples were taken from two shallow sites (less than 5 km apart) in the Orkneys. Sampling was done using a bucket at both sampling locations. Water for filtration was transported to the Orkney Lobster Hatchery where 500–1000 mL of water was filtered over 0.4 µm polycarbonate (PC) filter (Whatman, 47mm). Filters were stored at -20 °C until DNA isolation in the laboratory. The samples were processed at the AWI to produce a metabarcoding data set with the goal of analysing the general diversity at the two sites but also to investigate the dynamics of harmful algal bloom species (HABs). As the data also contained a wealth of information on protistan parasites they were used to produce an additional manuscript on oomycete infections in the planktonic foodweb at the two sites. The bioinformatic processing of the raw sequence files was conducted as follows. The low-quality 3'-ends of the reads were trimmed by Trimmomatic (version 0.38) and the paired-ends were merged by VSEARCH (version 2.3.0). Cutadapt (version 1.19, was used to adjust the sequence orientation and to remove the forward and reverse primer matching sequence segments. Sequences were only kept in the sequence pool if both primer matching segments could be detected. To ensure a high sequence quality the remaining sequences were filtered by VSEARCH and only kept if the expected base error (sum of all base error probabilities) of a sequence was above 0.25. Chimeric sequences were predicted sample-wise by VSEARCH in de novo mode with default settings and removed from the sequence pool as well. Only samples which consisted of at least 10000 sequences after filtering were considered for further analyses. The remaining sequences were clustered into OTUs by the tool swarm (version 2.2.2) with default settings. For each OTU the most abundant amplicon was selected as representative and taxonomically annotated with the default classifier implemented in mothur (version 1.38.1). As reference the Protist Ribosomal Reference database (PR2), version 4.10 was chosen and the minimum confidence cut-off for annotation was set to a value of 80. Further details of the pipeline are provided in (Sprong, et al. 2020; doi:10.1016/j.seares.2020.101914).