Changing phytoplankton ecotypes in Fram Strait
After elucidating the limitations, advantages, and the synergistic potential of the primer sets we tested, we then combined the information gathered for each taxonomic group from all primer sets to provide a more holistic overview of how exported eukaryotic microbial communities changed in response to a warm anomaly in Fram Strait (2005-2007).
The Arctic Ocean is one of the most vulnerable regions where effects of the changing climate are very much apparent, mainly demonstrated by the shrinking annual summer ice extent associated with increasing sea surface temperature and an increase of warmer water masses entering from both the Pacific and Atlantic Oceans (Woodgate, Weingartner, & Lindsay, 2010; Beszczynska-Moller et al., 2012). The deep Fram Strait particularly serves as the Arctic gateway to the Arctic Ocean, as its eastern side is influenced by the incoming warmer Northern Warm Atlantic current (NWAC) continuing as the West Spitsbergen Current (WSC) and by the colder East Greenland Current in the west (EGC). Since the physico-chemical conditions of the water masses differ, they have also been reported to harbor distinct plankton communities, including microbial eukaryotes (Kilias, Wolf, Nothig, & Peeken, 2013). The LTER observatory HAUSGARTEN, where the sediments traps used in this study were deployed is situated 120 km west of Spitsbergen (Svalbard) and has been continuously monitored since 1999 (Soltwedel et al., 2016), providing unique long-term data in this particular part of the Arctic Ocean. One remarkable observation was the detection of a warming event of the Atlantic Water in the WSC starting in late 2004, peaking in 2006 and lasting until early 2008, usually referred to as the warm anomaly of 2005-2007 (Beszczynska-Möller et al., 2012). This period was not only accompanied by temperature anomalies but also by decreased ice extent (Lalande et al., 2013). A great number of studies have already been published complimenting the described event including changes in the composition of the community based on microscopy (Kraft et al., 2013; Bauerfeind et al., 2015; Kubiszyn, Piwosz, Wiktor, & Wiktor, 2014), decreased fluxes in biogenic matter (Lalande, Bauerfreind, Nothig, Beszcynska-Moller, 2013), primary productivity (Nöthig et al., 2015), shift in dominant cell size (Vernet, Richardson, Metfies, Nöthig, & Peeken, 2017), and protistan community composition based on gene surveys (Metfies et al., 2017). Although these studies revealed the shift in cell sizes (from large to small) and diversity of dominant taxonomic groups, the methods used in these previous investigations were not powerful enough to reveal potential changes at the level of the ecotypes.
Here, we observed that sequence assemblages of eukaryotic microbial communities based on sequencing were mainly dominated by chlorophytes even before the warm anomaly. However, in addition, we further distinguished that the years before and after the warm anomaly were mainly dominated by OTUs of the cold-ecotype of Mamiellophyceae, particularly closest to CCMP2099 or formerly Clade Ea (Lovejoy et al., 2007) but now known as Micromonas polaris (Simon et al., 2017). This ecotype has only been reported in the cold waters of the Arctic Ocean and tend to be the most abundant phytoplankton in the region (Lovejoy et al., 2007; Simon et al., 2017). M. polaris thrive even in nutrient-limited conditions, allowing them to outcompete larger phytoplankton due to their surface area to volume ratio advantage (Lovejoy et al., 2007; Li, Mclaughlin, Lovejoy, & Carmack, 2009). In comparison, the warm anomaly years were characterized by the increased abundance of the warm ecotype Clade C (CCMP1195), now known asMicromonas commoda commonly found in temperate and tropical regions (Simon et al., 2017). Recently, Hoppe, Flintrop, and Rost (2018) showed through laboratory experiments that the Micromonas pusillaisolated from Kongsfjorden, Svalbard benefited from warming and acidification with increased growth rate and biomass buildup. The strain they tested however might actually be more associated with the warm ecotype of M. commoda (Clade C), which we found abundant in this study, rather than the Arctic resident M. polaris (Clade Ea). This could partially explain why chlorophytes (Clade C) were also abundant during the warm anomaly period. The alternating patterns between M. commoda and M. polaris have significantly contributed to the sustained overall high abundance of chlorophytes despite changes in the conditions, which might have implications to ecosystem resilience in the study area. To our knowledge, the same alternating pattern in ecotypes has not been reported elsewhere in the Arctic and could be unique to Fram Strait and specifically to Atlantic-influenced waters of the Arctic Ocean. The mechanisms of sinking of the small Micromonas however that allowed them to be transported to around ~300 m in this study remain unclear. Nevertheless, the sensitivity and patterns shown byMicromonas species make them sentinels of the changing environment (Demory et al., 2019).
We also observed alternating abundances in the dominant OTUs of the coccolithophore Emiliania huxleyi and colony-formingPhaeocystis pouchetii , with the former being higher during the warm anomaly period. Although the seasonal succession ofEmiliania and Phaeocystis from May until July has been reported in one short-term mooring in the Fram Strait in 2003 (Lalande, Bauerfeind, & Nöthig, 2011), their inter-annual variability has not yet been fully explored. Phaeocystis is a major bloom former in the Eurasian side of the Arctic, making up much of the blooms in Svalbard fjords (Hegseth and Tverberg, 2013; Marquadt, Vader, Stubner, Reigstad, & Gabrielsen, 2016), Fram Strait (Smith, Baumann, Wilson, & Aletsee, 1987; Gradinger and Baumann, 1991; Fadeev et al., 2018), Barents Sea (Hovland et al., 2014), Dutch waters (Veldhuist et al., 1986), Greenland (Arendt, Nielsen, Rysgaard, & Tönnesson, 2010; Waniek et al., 2005), Norwegian coasts, Barents Sea (Degerlund and Eilertsen 2010), Southern Ocean (Arrigo et al., 1999) and even under ice blooms (Assmy et al., 2017), and has been implicated in carbon transport in most regions of the central Arctic Ocean (Lalande et al., 2014, Wollenburg et al. 2018). In comparison, E. huxleyi was also reported in Fram Strait and North of Svalbard but in lower abundance and biomass (Hegseth and Sundfjord, 2008; Lalande et al., 2014). The high sequence abundance of the coccolithophore during the warm anomaly, known to contain carbonaceous cell walls, could partially explain the relatively sustained CaCO3 transport for the same period (Bauerfeind et al., 2009). This is consistent with the report of Neukermans, Oziel, & Babin (2018), where using satellite-derived data showed that blooms of E. huxleyi has been increasing and further moving northward in the Arctic. Interestingly, despite the decreased abundance of diatoms, POC flux was observed to peak in some years in Fram Strait, coinciding with peak abundance of Phaeocystis , suggesting their potential role in the transport of particulate organic carbon. Intriguingly however, some studies argued thatPhaeocystis colonies do not readily sink (Passow and Wassmann, 1994; Wolf, Kilias, & Metfies, 2015). Recent evidence however suggests that their lysis and disintegration either through grazing, apoptosis, and viral infections could induce the formation of transparent exopolymer (TEP) that allows production of Phaeocystis -derived aggregates (Engel et al., 2017), which then efficiently sink (Schoemann et al., 2005; Verity et al., 2007). Further, mineral ballasting particularly that of gypsum among P. pouchetii cells in the Arctic has been shown to enhance the vertical transport of carbon under the ice (Wollenburg et al., 2018), emphasizing the increasing role of the haptophytes in the biogeochemical cycles in the polar regions.
Compared to haptophytes and chlorophytes, which are small and difficult to identify using conventional microscopy techniques, identity and abundance of diatoms are easier to track and detect due to their larger size and distinct rigid silicate tests. Previous studies based on microscopy reported that the diatoms abundant before the warm anomaly were mostly Thallasiosira spp., Chaetoceros spp.,Fragilariopsis spp., Navicula spp., Achnanthesspp., and Fossula arctica (Kubiszyn et al., 2014; Nöthig et al., 2015; Vernet et al., 2017). In contrast, findings in this study revealed that diatom reads in all libraries generated by all primer sets were dominated by Chaetoceros , either indicating biases towards this group or inefficiency in amplifying other species. In addition, the RAxML-EPA approach was also able to identify the sequences down to the species level with some belonging to the C. brevis/debilis and majority from C. gelidus/socialis complexes, which are almost not distinguishable based on light microscopy. The more abundant ‘C. socialis ’ species complex is described to be cosmopolitan, occurring in the colder polar waters to the warmer Mediterranean and Asian waters (Hasle & Syvertsen, 1997; Degerlund, Huseby, Zingone, Sarno, & Landfald, 2010; Kooistra et al., 2010). Careful morphological and phylogenetic re-examination of the representatives of the C. socialis complex however revealed the presence of a new cold-adapted clade C. gelidus sp. nov. and the warm-associated clade C. socialis (Chamnansinp, Li, Lundholm, & Moestrup, 2013). Interestingly, the most abundant OTU in the Stoeck dataset (OTU2) belonged to C. gelidus (Figure 4C). To our knowledge, this would be the first time that this species will be reported abundant in this region. OTUs belonging to this species have also been reported in the Arctic waters on the Canadian side of Baffin Bay (Joli et al., 2018) and abundance of cells in Beaufort Sea (Balzano et al., 2017), indicating their potential widespread occurrence and distribution but underappreciated role in the Arctic Ocean. In addition, C. brevis / debilis species have also been found in northern temperate areas and Arctic waters but seemed to be a more important diatom in the Antarctic (Trimborn et al., 2017). Nevertheless, the presence of these cold-adapted species signifies favorable conditions for their proliferation and in turn, their absence during the warm anomaly period indicates significant changes that filtered them out from the environment. Most of these diatoms are bloom- and colony-forming species, making them important drivers of carbon and silica fluxes, and their decreased abundance significantly affected the vertical transport of silicate in Fram Strait (Lalande et al., 2014; 2016). Even after the warm anomaly, the relative abundances of these species did not return to their before-state, indicating that long-term changes in the community occurred, which would have significant implications to the pelagic-benthic coupling of Arctic ecosystems.