Phytoplankton assemblage structure in and around a massive under-ice bloom in the Chukchi Sea

Deep-Sea Research II ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Contents lists available at ScienceDirect

Deep-Sea Research II journal homepage: www.elsevier.com/locate/dsr2

Phytoplankton assemblage structure in and around a massive under-ice bloom in the Chukchi Sea Samuel R. Laney n, Heidi M. Sosik Biology Department, Woods Hole Oceanographic Institution, Woods Hole, MA 02543, USA

art ic l e i nf o

Keywords: Phytoplankton Algal blooms Sea ice Community composition

a b s t r a c t Standard and imaging flow cytometry were used to examine the composition of phytoplankton assemblages in and around a massive under-ice bloom in the Chukchi Sea in 2011. In the core of this bloom, roughly 100 km northwest of Hanna Shoal, diatoms represented roughly 87% of the water column carbon-specific biomass of phytoplankton, while nanophytoplankton contributed
9%. Picoeukaryotes were also observed in this bloom, as were phycoerythrin-containing cells consistent with Synechococcus spp., but picophytoplankton, dinoflagellates, and prymnesiophytes each represented only
1% of the bloom's phytoplankton biomass. More broadly along this part of the Chukchi shelf, nanophytoplankton typically comprised a larger fraction of phytoplankton biomass in the water column, 22% on average but up to 82% at certain locations. Dinoflagellates and prymnesiophytes contributed at most 2% of water column biomass at any location and were most abundant in the deeper slope stations northeast of Hanna Shoal, east of the bloom. Picophytoplankton were most abundant in these deeper slope stations as well, and also in recently ice-free areas to the south around Hanna Shoal. These cellderived estimates of phytoplankton carbon biomass, which were computed from imaging and standard cytometric observations of phytoplankton cell sizes and from published carbon:volume relationships, agree well with independent measurements of particulate organic carbon concentration from traditional biochemical assays. & 2014 Elsevier Ltd. All rights reserved.

1. Introduction Several large-scale field programs have been conducted in the Chukchi Sea over the past decade, motivated in large part by an interest in understanding better the effects that changing climate will have on Arctic marine ecosystems (Bluhm et al., 2010; Grebmeier and Harvey, 2005; Grebmeier et al., 2009). These field programs have generated considerable new insight into the distribution and seasonality of phytoplankton assemblages in this region of the Arctic Ocean (e.g., Hill et al., 2005; Joo et al., 2012; Min Joo et al., 2012; Sukhanova et al., 2009), but our understanding of phytoplankton assemblages in the Chukchi Sea still contains significant gaps. A notable example is the very high phytoplankton biomass levels that were observed under consolidated pack ice northwest of Hanna Shoal during the ICESCAPE field program (Impacts of Climate on EcoSystems and Chemistry of the Arctic Pacific Environment) in July 2011 (Arrigo et al., 2012). n Correspondence to: Biology Department, Woods Hole Oceanographic Institution, MS#34 Redfield 1-36, Woods Hole, MA 02543, USA. Tel.: þ 1 508 289 3647; fax: þ1 508 457 2134. E-mail address: [email protected] (S.R. Laney).

High phytoplankton biomass in ice-covered waters has been documented previously in the Arctic: both directly from measurements of chlorophyll and/or production (Fortier et al., 2002; Mundy et al., 2009; Sherr et al., 2003; Strass and Nö thig, 1996) and also indirectly from observed depletion of water column nutrients early in the growing season (e.g., Cota et al., 1996). The 2011 Chukchi event we studied was exceptional in terms of the magnitude of water column biomass, which exceeded 32 g C m  2 in the core of this bloom (Arrigo et al., this issue). Its coincidence with drawdown of nitrate in the upper water column and with very high photosynthetic rates and efficiencies signify an active bloom. The surface melt ponds that were observed over the bloom region likely played a role in creating the high level of production, by increasing the coupling of incident sunlight into the water column below (Christensen and Melling, 2010; Perovich et al., 1998). Optical characteristics of melt ponds previously observed in this region of the Arctic (Frey et al., 2011) appear appropriate for providing the light intensities necessary to form an under-ice bloom of this magnitude. Initial taxonomic assessments of this under-ice bloom focused on the question of its genesis, i.e., whether the bloom arose from phytoplankton living in situ in the water column or if instead it

http://dx.doi.org/10.1016/j.dsr2.2014.03.012 0967-0645/& 2014 Elsevier Ltd. All rights reserved.

Please cite this article as: Laney, S.R., Sosik, H.M., Phytoplankton assemblage structure in and around a massive under-ice bloom in the Chukchi Sea. Deep-Sea Res. II (2014), http://dx.doi.org/10.1016/j.dsr2.2014.03.012i

S.R. Laney, H.M. Sosik / Deep-Sea Research II ∎ (∎∎∎∎) ∎∎∎–∎∎∎

2

was seeded by algae living in the overlying sea ice. Images of individual nanophytoplankton (2–20 μm) and microphytoplankton ( 420 μm), collected with an Imaging FlowCytobot (IFCB, Olson and Sosik, 2007), revealed that the water column assemblages in the immediate core of the bloom were dominated by diatoms of the genera Chaetoceros, Thalassiosira, and Fragilariopsis, taxonomically distinct from the assemblages found in samples taken from cores in the overlying ice (Arrigo et al., 2012). This preliminary assessment of the region around the core of the bloom examined only a subset of the over 50,000 algal images collected in this region during the 2011 ICESCAPE field season. Moreover, standard flow cytometric (FCM) analyses of cells too small to identify from IFCB images indicate that the smaller size fractions may have played a larger role than initially assumed. This broader dataset suggests a much greater degree of complexity in phytoplankton assemblages in this region of the Chukchi Sea than would be inferred from the core of the bloom alone. Here we synthesize these two complementary methods (standard and imaging flow cytometry) to examine in more detail the distribution and taxonomic structure of phytoplankton assemblages in and around this under-ice bloom, as well as the relative contribution of different taxa and size classes to total algal biomass.

2. Methods 2.1. Stations and sampling This analysis considers 35 stations occupied in the Chukchi Sea in July 2011, along three hydrographic sections extending from recently ice-free waters near Hanna Shoal northward into the ice pack (Fig. 1, Table 1). These stations and transects cover the immediate area of the under-ice bloom reported by Arrigo et al. (2012) northwest of Hanna Shoal, as well as adjoining areas outside the bloom to the east. At each station water samples were collected typically from six depths, beginning at around 1.5 m and including the deep chlorophyll maximum if one was observed. A core suite of measurements was performed on these samples including chlorophyll concentration, particulate organic carbon concentration, and macronutrient concentrations. Materials and methods for these measurements are documented in Arrigo et al. (this issue). 2.2. Standard flow cytometry A portable flow cytometer (Accuri C6, Becton–Dickson) was used during this field study to determine the abundance of picophytoplankton (o2 μm) and nanophytoplankton (2–20 μm). Seawater samples were prepared by prefiltering through Nitex screening of nominal mesh size 200 μm to remove larger cells, chains, and colonies that might clog the instrument's flow cell. For each sample approximately 200 μl of seawater was counted. Distilled water blanks were measured after each set of samples at every station, and standard beads (PeakFlow P14827, 2.5 mm, 515 nm emission) were also run periodically during the cruise to track instrument behavior, to determine the instrument's level of sensitivity in detecting cells or other particles, and to provide reference data for normalizing the phytoplankton cell scattering measurements into bead units. Custom software was used to interpret the measured cell scattering and fluorescence data and determine the abundances of pico- and nanoeukaryotes in these samples. Phytoplankton cells were discriminated from other particles by their relative combinations of chlorophyll fluorescence and side scattering. Phycoerythrin fluorescence served as an additional discriminant for

identifying very small (ca. 1 μm) particles consistent with cells of Synechococcus spp. In order to assess the relationship between forward scattering and cell size, a laboratory size calibration study was performed with twelve phytoplankton cultures with cell sizes between 1 and
14 μm. These cultures included two coccolithophores (Emiliania huxleyi, Syracosphaera elongata), one other prymnesiophyte (Isochyrsis sp.), a cyanobacterium (Synechococcus sp.), a diatom (Cylindrotheca fusiformis), two dinoflagellates (Amphidinium carterae, Gymnodinium galatheanum), two chlorophytes (Dunaliella tertiolecta and Nannochloris sp.), a pelagophyte (Pelagomonas sp.), a prasinophyte (Micromonas pusilla), and an unidentified small (2.2 μm diameter) eukaryote. Mode cell volume for each culture was estimated by analysis on a Coulter Multisizer II outfitted with a 30 μm orifice and calibrated with NIST traceable 10 μm latex beads (Coulter size standard L10). The mode values of integrated forward light scattering (FSC-A, measured on the same cultures with the Accuri C6 flow cytometer) were normalized to the same 2.5 μm beads measured during the cruise in order to provide results in bead units. A second-order polynomial was fit to these log transformed data, generating an empirical calibration for converting field measurements of Accuri FSC-A into estimates of individual cell volume (Fig. 2). While the fit represents variation across a range of cell sizes and types, cell volume of Cylindrotheca fusiformis is underestimated by this approach, which is consistent with previous work showing that relatively low forward light scattering is expected due to the elongated shape of many pennate diatoms (Olson et al., 1989). This laboratory-derived relationship between Accuri C6 forward scattering and cell size suggests that this flow cytometer does not have the sensitivity necessary to resolve scattering of very small cells (size
1 μm), because when measured with the Accuri C6 the Synechococcus sp. cells had FSC-A values indistinguishable from noise levels. Thus, this data point can be considered an estimate of the detection level intercept for Accuri C6 observations. Since Synechococcus is also difficult to size accurately with a Multisizer configured with a 30 μm aperture, its volume was approximated as a 1 μm diameter sphere, as were all cells in the FCM field measurements that were identified as Synechococcus by virtue of their phycoerythrin fluorescence. It is important to emphasize that the lack of sufficient light scattering sensitivity to characterize Synechococcus does not impact the ability to detect and enumerate these cells with the Accuri C6, because their phycoerythrin and chlorophyll fluorescence signals are readily distinguishable from background. 2.3. Imaging flow cytometry Digital micrographs of phytoplankton in the nano- and microsize fraction were collected with a variant of the Imaging FlowCytobot based on the design described by Olson and Sosik (2007). Briefly, volumes of seawater (typically 5 ml) were injected through an 860  180 μm flow cell through which a 635 nm laser beam was focused. Chlorophyll-containing particles that passed through this beam emitted a fluorescence signal (4650 nm) that was detected by a photomultiplier tube. Fluorescence events triggered a digital camera that captured a micrograph of that particular cell, chain, or colony. A Nitex screen with nominal mesh size of 130 μm was placed on the instrument's sample intake during the ICESCAPE study to prevent larger particles from clogging the flow cell. This screen, combined with the camera field of view, set the effective upper size limit of cells, chains, or colonies seen in these images to those of length ca. 300 μm or less. The lower size limit of cells seen in these IFCB images is a function of the minimum fluorescence intensity needed to trigger the camera. For samples collected during the ICESCAPE study, cells typically above 8 μm

Please cite this article as: Laney, S.R., Sosik, H.M., Phytoplankton assemblage structure in and around a massive under-ice bloom in the Chukchi Sea. Deep-Sea Res. II (2014), http://dx.doi.org/10.1016/j.dsr2.2014.03.012i

S.R. Laney, H.M. Sosik / Deep-Sea Research II ∎ (∎∎∎∎) ∎∎∎–∎∎∎

3

74 oN

101 57 58 59 60 61 62 63 64 65 66 67

30’

73 oN

76 77 78 68 79 69 80 84 70 71 81 72 73 82 Chukchi Sea Hanna Shoal

30’

Latitud o e ( N)

99 97 95 93 91 90 88 87 86

72 oN

30’

71 oN

C en tra

Herald Shoal

lC ha

30’

nn el

70 oN 170 W

168oW

166oW

o

164 W

o

162 W

o 160 W

158 W

Longitude (oW) Fig. 1. Station locations and numbering, relevant bathymetry, and ice extent (stippling) during the time surrounding station sampling. Ice coverage was computed from 6.25 km AMSR-E (Advanced Microwave Scanning Radiometer) data, using the ARTIST Sea Ice (ASI) algorithm (Spreen et al., 2008). Stippling represents locations where this algorithm predicts ice cover in excess of 20%.

Table 1 Stations and hydrographic sections presented in this analysis. Stations 58 to 63 in the Chukchi Slope West section roughly coincide with the under-ice bloom described in Arrigo et al. (2012). Section

Abbreviation

End stations

Depths at end stations

Chukchi Slope West Hanna Shoal North Chukchi Slope Center

CSW HSN CSC

57–73 76–82 101–84

150–39 m 74–27 m 3285–41 m

have strong enough fluorescence intensities to trigger images that were visually informative. The location and morphometrics of phytoplankton and other particles in these micrographs were determined with custom software that performed automated image processing approaches described previously (Sosik and Olson, 2007). Initial classification of images was done with a supervised machine learning strategy similar to that described in Sosik and Olson (2007), except with a random forest algorithm (Breiman, 2001) in place of the original support vector machine. The classifier included 47 separate operational classes, 26 of which were observed in these ICESCAPE samples (Table 2) and 22 of which represented phytoplankton identified to the level of genus or better. Several non-algal classes were also included to account for images that contained detritus, empty diatom frustules, calibration beads, or instances where multiple taxonomic classes were seen in a single image. Results of this automated classification process were manually verified for all samples assessed for this study, to correct for any mistakes

Fig. 2. Relationship between mode cell volume as measured by a Coulter Multisizer (ordinate) and mode integrated forward light scattering measured on an Accuri C6 (parameter FSC-A, normalized to 2.5 μm beads; abscissa), for 12 phytoplankton cultures with cell sizes ranging between
1 and
14 μm. The solid curve represents the best fit second order polynomial. Dashed lines indicate 95% confidence intervals.

made by the automated classifier, to ensure the highest possible accuracy in taxonomic identifications, and to include additional taxonomic classes not included in the automated classifier.

Please cite this article as: Laney, S.R., Sosik, H.M., Phytoplankton assemblage structure in and around a massive under-ice bloom in the Chukchi Sea. Deep-Sea Res. II (2014), http://dx.doi.org/10.1016/j.dsr2.2014.03.012i

S.R. Laney, H.M. Sosik / Deep-Sea Research II ∎ (∎∎∎∎) ∎∎∎–∎∎∎

4

Table 2 The 26 classes used in this analysis of IFCB images, ranked in order of the number of images observed. Diamond symbols denote diatom classes. Crosses indicate classes for which the ‘large diatom’ relationship was applied when estimating cell-specific carbon (i.e., if the computed target image volume exceeded 3000 μm3, see text). Taxonomic class ⋄

Chaetoceros Fragilariopsis⋄ þ Thalassiosira⋄ þ Pseudonitzschia⋄ Unclassified nanophytoplankton Unclassified dinoflagellates Bacterosira⋄ Cylindrotheca⋄ Unclassified pennates⋄ þ Navicula⋄ þ Phaeocystis Phaeocystis_with_diatoms Detonula⋄ Pleurosigma⋄ þ Eucampia⋄ Unclassified diatom fragments⋄ Nitzschia frigida⋄ Euglena Dictyocha Odontella⋄ Coscinodiscus⋄ Leptocylindrus⋄ Pyramimonas Ephemera⋄ þ Dinobryon Rhizosolenia⋄ þ Total

Images (#) 38,797 1019 1565 3142 3220 376 68 2198 524 77 119 28 61 53 31 55 13 11 5 1 10 2 10 1 1 1 51,418

The absolute abundances of cells (or chains, or colonies) in these samples was determined from their apparent abundance by correcting the nominally 5 ml volume sampled to reflect the actual volume of seawater observed by the IFCB. The actual observed volume is always less than the nominal 5 ml volume due to a short ‘dead time’ following each trigger during which time each image is processed and stored. The morphometrics of each cell, chain, or colony were also used to determine phytoplankton biovolumes from the two-dimensional outlines in each image, using a distance map method (Moberg and Sosik, 2012). Images of fluorescent microspheres of a known diameter (9 μm, Duke Scientific Inc.) were collected to provide the pixel-to-micron factor necessary to calibrate these biovolumes in μm3.

2.4. Phytoplankton carbon estimates These measurements of phytoplankton volume, provided separately by standard or imaging flow cytometry, were then each transformed into units of carbon (pg C particle  1) with one of two volume-to-carbon relationships described by Menden-Deuer and Lessard (2000): one for diatom taxa with biovolumes typically larger than 3000 μm3 and a second one for all other cells. This approach is justified by the meta-analysis of Menden-Deuer and Lessard (2000) which showed no significant differences in carbonto-volume relationship among taxa, except in the case of large diatoms, presumably due to the presence of intracellular vacuoles. Of the classes of phytoplankton identified in these ICESCAPE samples (Table 2), those for which this this second ‘large diatom’ category was applied (if the target exceeded 3000 μm3) were Ephemera, Pleurosigma, Rhizosolenia, Thalassiosira, Fragilariopsis, Navicula, and unclassified pennates. Note that genera such as Chaetoceros, whose long setae give them large external dimensions, are not included in this group because their actual cell volumes are relatively small.

The phytoplankton-specific carbon estimates from standard flow cytometry and from the IFCB analyses were then merged to generate carbon estimates for the entire phytoplankton assemblage in the region around the Chukchi under-ice bloom, covering cells of
1 μm in size up to chains of 4300 μm in length. Carbon estimates for the smallest size fraction (picophytoplankton: cells o2 μm) were derived from the Accuri analyses alone because the smallest size cells reliably captured in IFCB images were
8 μm. Similarly, estimates for the largest size fraction (microphytoplankton: cells, colonies, or chains 420 μm) were derived solely from the IFCB analyses because the largest cells included in the Accuri size calibration study were only
14 μm, and extrapolating these results outside the observed size range would introduce unacceptable uncertainties. For the intermediate size fraction (nanophytoplankton: cells 2–20 μm) it was necessary to merge the results of the separate FCM and IFCB analyses. A size threshold of 10 μm was chosen such that nanophytoplankton abundances and carbonspecific biomass estimates reflect data from standard flow cytometry for cells between 2 and 10 μm, and data from the IFCB for cells, chains, and colonies between 10 and 20 μm. The final result of this merging of FCM and IFCB data are estimates of phytoplankton carbon biomass per unit volume of seawater, separately for the picophytoplankton, nanophytoplankton, and microphytoplankton size fractions, and also independently for Synechococcus spp. or for any of the individual phytoplankton classes that could be identified from IFCB cell imagery (Table 2). In these size class designations, all specifications of cell length represent equivalent spherical diameter (ESD) computed from the cell volume, and for practical reasons associated with the automated image processing chains and colonies are represented as a single large target (not separately counted smaller cells). Particularly for the microphytoplankton, these length-based size estimates are only a idealized representation of the actual, complex morphologies of these cells, chains, or colonies (e.g., see Fig. 5 below), but ESDs provide a means for consistent representation of size spectra and intercomparison broadly across samples.

3. Results and discussion 3.1. Photosynthetic prokaryotes and picoeukaryotes: abundance and biomass Standard FCM analyses of seawater samples indicated that phycoerythrin-containing cells consistent with Synechococcus spp. were present in this region of the Chukchi shelf, albeit at very low abundances, typically o100 ml  1 and sometimes undetectable (Fig. 3, top left panels). These low typical abundances of Synechococcus are in line with previous observations from the same region in both summer and winter reported by Cottrell and Kirchman (2009), who used a more sensitive flow cytometer. At a few locations Synechococcus was evident with much higher abundances, e.g., in near-surface depths within the bloom at station 62 (exceeding 18,000 ml  1) and southward closer to the ice edge at station 64, and throughout the full water column along the 40-m isobath around Hanna Shoal at stations 73, 80, and 81. Although Synechococcus is typically thought to be excluded from polar waters due to its poor tolerance to colder temperatures (Murphy and Haugen, 1985; Olson et al., 1990a; Olson et al., 1990b), recent genomic analyses suggest that certain Synechococcus strains may be better cold-adapted than others (Huang et al., 2012). With these FCM observations we cannot conclusively say whether these few high-abundance observations for Synechococcus reflect in situ growth of such cold-tolerant strains, or instead represent transport of cells from the North Pacific source waters that enter the Chukchi Sea through the Bering Strait (Liu et al.,

Please cite this article as: Laney, S.R., Sosik, H.M., Phytoplankton assemblage structure in and around a massive under-ice bloom in the Chukchi Sea. Deep-Sea Res. II (2014), http://dx.doi.org/10.1016/j.dsr2.2014.03.012i

S.R. Laney, H.M. Sosik / Deep-Sea Research II ∎ (∎∎∎∎) ∎∎∎–∎∎∎

5

Fig. 3. Top three rows: abundances (left column, log10 ml  1) and computed carbon-specific biomass (right column, log10 μg l  1) of Synechococcus spp. along the three sections (rows), derived from standard flow cytometry. Bottom three rows: similarly structured plots for picoeukaryote phytoplankton ( o2 μm) along the three sections (rows), also derived from standard flow cytometry. Different color bar scales are used to enhance features along these three sections. The dashed boxes in the CSW sections indicate the region defined here as the under-ice bloom.

2002). Regardless, where Synechococcus was most abundant it represented only around 1 μg l  1 of the total phytoplankton carbon biomass; elsewhere it represented two orders of magnitude less on average (Fig. 3, top right panels). Photosynthetic picoeukaryotes were widespread throughout the study region, observed at over 30,000 ml  1 in some locations but typically found at much lower abundances (Fig. 3, bottom left panels). These cells tended to inhabit the upper water column throughout the study region. The highest abundances were observed in near-surface samples from recently ice-free areas just south of the bloom (e.g., Chukchi Shelf West stations 65 to 72) and also at stations just north of Hanna Shoal between the 40 m and 50 m isobaths (stations 78, 79, and 84). The distribution of estimated carbon-specific biomass in these small cells generally followed their distribution in abundance (Fig. 3, bottom right panels), with values ranging between 6.7 ng l  1 to almost 400 μg l  1 where most abundant. Given their relatively low abundances and the small size of these cells, Synechococcus and picoeukaryotes contributed only negligibly to the biomass of Chukchi Sea assemblages in and

around the massive 2011 under-ice bloom. Only at five stations (70–72, 99, and 101) did picophytoplankton contribute more than 10% of the total phytoplankton carbon biomass in the water column (Table 3). The maximum contribution of picophytoplankton to the total water column biomass was 22%, seen at the furthest offshore station in deepest waters (station 101). 3.2. Nanophytoplankton and microphytoplankton: abundance and biomass Nanophytoplankton were most abundant in recently ice-free waters northwest of Hanna Shoal (e.g., stations 67–73; Fig. 4, upper left panels) with the highest absolute abundance observed on the northeastern edge of the shoal at mid-depths at station 84. The nanophytoplankton contribution to total water column carbon biomass in this part of the Chukchi Sea ranged between 17% and 83% (Table 3). The lowest relative contribution occurred at station 60, within the core of the under-ice bloom, with nearby stations 59 and 61 also low. Nanophytoplankton contribution was also low at stations 80 and 84, on the 40 m isobath on the north side of

Please cite this article as: Laney, S.R., Sosik, H.M., Phytoplankton assemblage structure in and around a massive under-ice bloom in the Chukchi Sea. Deep-Sea Res. II (2014), http://dx.doi.org/10.1016/j.dsr2.2014.03.012i

6

S.R. Laney, H.M. Sosik / Deep-Sea Research II ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Fig. 4. Top three rows: abundances (left column) and computed carbon-specific biomass (right column) of nanophytoplankton (2–20 μm) along the three sections (rows), derived from merged flow cytometry and imaging flow cytometry data. Bottom three rows: similarly structured plots for microphytoplankton ( 420 μm) along the three sections (rows), derived from imaging flow cytometry alone. Different scales are used with the color bars to enhance features along these three sections. The dashed boxes in the CSW sections indicate the region defined here as the under-ice bloom.

Hanna Shoal, whereas the highest contribution was seen at station 72, on the same isobaths but on the northwest side of the shoal. IFCB analyses of the 181 discrete seawater samples taken from these 35 stations produced 51,418 images that could be positively identified as containing phytoplankton cells, chains, or colonies (Fig. 5). Over 92% of these images contained solitary diatoms or diatom chains, and 6% were of nanophytoplankton that could not be classified further by visual inspection. The smallest cells reliably seen in the IFCB images were found in this nanophytoplankton class, with ESD
8 μm. The largest phytoplankton seen were colonies of Chaetoceros spp., most likely C. socialis, with length scales exceeding 300 μm. It is likely that larger chains or colonies were present in these seawater samples but were not captured in these images because of the 130 μm Nitex screen placed on the instrument's sample intake. Within the microphytoplankton taxa that could be quantified with our method, an even larger variability was found compared to nanophytoplankton. Relative contributions of microphytoplankton to total phytoplankton C biomass ranged from 2% (e.g., at

station 72 in ice-free waters on the northwest edge of Hanna Shoal, dominated by nanophytoplankton) to 90% (at station 60 in the core of the bloom). Neukermans et al. (this issue) observed similar ranges in pico-, nano-, and microphytoplankton contributions estimated from optical- and pigment-based methods. We cannot make a direct comparison because their work focused on detailed analysis of two stations just to the east of our study area. They report, however, surface waters with 29% microplankton at an open water station contrasting with 98% at an under-ice bloom site. Across the study region as a whole we observed over 4 orders of magnitude difference in the contribution of different phytoplankton taxa to water column carbon biomass. The largest contributors to phytoplankton carbon were diatoms of genus Chaetoceros, followed by unclassified nanophytoplankton and diatoms of genera Fragilariopsis and Thalassiosira (Fig. 6). Picophytoplankton as a group were the fifth most important contributor to phytoplankton carbon in this region of the Chukchi Sea. In terms of percentages Chaetoceros alone contributed 53% of the total

Please cite this article as: Laney, S.R., Sosik, H.M., Phytoplankton assemblage structure in and around a massive under-ice bloom in the Chukchi Sea. Deep-Sea Res. II (2014), http://dx.doi.org/10.1016/j.dsr2.2014.03.012i

S.R. Laney, H.M. Sosik / Deep-Sea Research II ∎ (∎∎∎∎) ∎∎∎–∎∎∎

7

Phaeocystis sp.

Chaetoceros sp.

Navicula sp. Navicula sp.

Bacterosira sp.

Thalassiosira sp. Pleurosigma sp. Detonula sp. Unclassified pennates Pseudo-nitzschia sp.

(A)

(B)

(D)

(C)

10 µm

Fragilariopsis sp. Fig. 5. Example Imaging FlowCytobot images of phytoplankton cells, chains, and colonies for those classes in Table 2 where 50 or more images were seen in imagery from the study region. Frames too small to accommodate labels are (A) unclassified diatom fragments, (B) unclassified dinoflagellates, (C) Cylindrotheca spp., and (D) unclassified nanophytoplankton. A 10 μm scale bar is shown in the bottom right frame.

phytoplankton carbon biomass encountered at these 35 stations, and these top three diatom taxa together accounted for 68% of total phytoplankton carbon. The remaining top ten contributors to phytoplankton carbon include the diatoms Bacteriosira and Pseudonitzschia, dinoflagellates that could not be classified further from imagery alone, similarly unclassified species of pennate diatoms, and the diatom Detonula. Detailed taxonomic analyses of Arctic phytoplankton bloom assemblages are rare, but it is notable that the two dominant diatom taxa in this region of the Chukchi Sea (Chaetoceros and Thalassiosira) were also found to be the major contributors to primary production in a well-studied bloom in a recurrent polynya in northern Baffin Bay (Booth et al., 2002). Our approach for combining IFCB-based and FCM-based estimates of cell carbon introduces unavoidable artifacts due to the

arbitrary decision for the size threshold below which IFCB-based, taxon-specific estimates of cell carbon are replaced by FCM-based, ataxonomic estimates. For several of the taxa we were able to identify in IFCB imagery (most notably for Cylindrotheca and for unclassified diatom fragments), a substantial fraction of IFCBbased carbon was observed in cells with characteristic sizes of 10 μm or less. To avoid double counting, this carbon was deducted from these categories, as it should already be included in the pooled carbon data by FCM-based estimates assigned to nanophytoplankton (Fig. 6, unshaded bar areas). For Cylindrotheca for example, around 89% of IFCB-based estimates of carbon occurs in cells less than 10 μm and is instead represented in the ‘unclassified nanophytoplankton’ category by FCM-derived carbon estimates. Overall this type of artifact does not affect our conclusions

Please cite this article as: Laney, S.R., Sosik, H.M., Phytoplankton assemblage structure in and around a massive under-ice bloom in the Chukchi Sea. Deep-Sea Res. II (2014), http://dx.doi.org/10.1016/j.dsr2.2014.03.012i

S.R. Laney, H.M. Sosik / Deep-Sea Research II ∎ (∎∎∎∎) ∎∎∎–∎∎∎

8

Table 3 The fractional contribution of different size classes (micro-, nano-, and picophytoplankton) and broad taxonomic groups (diatoms, dinoflagellates, and prymnesiophytes) to total water column phytoplankton biomass, from the merged carbon estimates (FCM-derived plus IFCB-derived) across the 35 stations. Station

Environment

Micro

Nano

Pico

Diato

Dino

Prym

57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 76 77 78 79 80 81 82 84 86 87 88 90 91 93 95 97 99 101

Shelf break Under-ice bloom Under-ice bloom Under-ice bloom Under-ice bloom Under-ice bloom Under-ice bloom Ice cover, on shelf Ice edge Recently ice free Recently ice free Recently ice free Recently ice free Recently ice free Recently ice free Recently ice free Recently ice free Ice cover, on shelf Ice cover, on shelf Ice edge Recently ice free Recently ice free Recently ice free Recently ice free Recently ice free Ice cover, on shelf Ice cover, on shelf Ice cover, on shelf Ice cover, on shelf Ice cover, on shelf Ice cover, 4200 m Ice cover, 4200 m Ice cover, 4200 m Ice cover, 4200 m Ice cover, 4200 m

0.69 0.81 0.87 0.90 0.83 0.76 0.63 0.64 0.71 0.66 0.70 0.45 0.77 0.03 0.16 0.01 0.82 0.79 0.77 0.70 0.69 0.86 0.77 0.77 0.88 0.78 0.74 0.66 0.66 0.38 0.58 0.45 0.56 0.26 0.10

0.29 0.18 0.12 0.09 0.16 0.20 0.33 0.32 0.21 0.24 0.26 0.49 0.20 0.78 0.73 0.83 0.17 0.21 0.22 0.29 0.29 0.12 0.22 0.22 0.12 0.21 0.26 0.33 0.34 0.61 0.41 0.52 0.41 0.59 0.68

0.02 0.01 0.01 0.01 0.01 0.04 0.04 0.04 0.08 0.09 0.05 0.06 0.03 0.19 0.11 0.16 0.01 0.01 0.01 0.01 0.02 0.02 0.01 0.01 0.00 0.00 0.00 0.01 0.00 0.01 0.01 0.03 0.03 0.15 0.22

0.68 0.80 0.80 0.87 0.80 0.74 0.61 0.61 0.69 0.65 0.69 0.44 0.75 0.02 0.15 0.00 0.81 0.78 0.77 0.69 0.68 0.85 0.77 0.76 0.87 0.78 0.74 0.66 0.65 0.38 0.57 0.43 0.54 0.20 0.04

0.00 0.00 0.02 0.01 0.00 0.01 0.01 0.01 0.01 0.01 0.01 0.00 0.02 0.01 0.01 0.01 0.00 0.01 0.00 0.00 0.00 0.01 0.00 0.01 0.00 0.01 0.00 0.00 0.00 0.00 0.01 0.02 0.02 0.06 0.06

0.00 0.00 0.03 0.01 0.01 0.01 0.01 0.02 0.00 0.00 0.00 0.01 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

significantly, given the relatively low ranking of Cylindrotheca among contributors to phytoplankton carbon in the study area. However, in other regions or ecosystems this may not be the case.

As one metric for considering the overall effects of these issues, we systematically compared particle concentration estimates for the 8–15 μm overlap range between the FCM and IFCB methods (Fig. 7). Despite the several challenges we identified above, we observe a wide range of samples where particle concentrations are not significantly different. We do observe a tendency for FCM to estimate higher concentrations than IFCB for a subset of samples where the overall concentration of particles is low (considering all observations, Model II regression: IFCB ¼ 0.92FCM 71.3, units ml  1 for the 8–15 μm range). Without more detailed studies comparing these methods for a comprehensive range of particle types, we cannot be sure of the source of these discrepancies at low concentration. A possible explanation that should be explored further is associated with our use of a constant chlorophyll fluorescence trigger for IFCB image acquisition, which could lead to 8–15 μm cells being missed if they contained especially low pigment. Nonetheless, considering the magnitude of the challenges discussed above, we have taken the level of agreement (correlation coefficient r ¼0.96) between methods to be adequate enough to consider some details of these merged FCM and IFCB size distributions. The cytometry-derived phytoplankton size distributions in this region of the Chukchi Sea reflect substantial differences in community structure beyond what the relative biomasses of pico-, nano-, and microphytoplankton would indicate (Fig. 8). Interestingly, under-ice bloom samples (Fig. 8, station 60) have relatively featureless spectra with an indication of slope decreasing with depth. Upper water column spectra from areas with high relative contributions of pico- and nanophytoplankton (e.g., station 70 in Fig. 8) exhibit more distinct features including peaks just under 2 μm ESD, shoulders near 3–5 μm, and a sharp truncation above
10 μm. In areas where phytoplankton biomass is dominated by the diatom Chaetoceros sp., subsurface spectra are notably flattened throughout the pico- and nano-size range with a distinct slope increase just above 10 μm ESD (Fig. 8, station 86). While the features of these size distributions are related to taxonomic composition in complex ways, future analyses coupling them with details about spectral absorption and scattering may provide a means to quantify links between community structure and bulk optical properties under different conditions in the Chukchi Sea. 3.4. Taxonomic richness and carbon-specific biomass among taxa

3.3. Phytoplankton size distributions For many applications where information beyond total biomass is important, broad pico-, nano-, and micro-size classes (o2 μm, 2–20 μm, 420 μm, respectively) are often invoked to characterize phytoplankton assemblages. Our cytometry-based approach provides a means to assess an even more detailed level of structure in phytoplankton size distributions, which might be important, for instance, in influencing upper ocean optical properties. Comparing and quantifying overlapping size distributions inferred from different kinds of measurements is challenging, however, and requires careful evaluation. Even if each method is well characterized and measurement uncertainty is low, there is no guarantee that different size estimates will be directly comparable across a range of particle morphologies and types. Factors that introduce complexity in our case include the ways that size, shape, composition, and refractive index influence electrical resistance in the Coulter Multisizer, interact with illumination conditions and optical geometry to determine light scattering measured by the Accuri C6, and impact edge contrast and detection in IFCB images. Furthermore, differences in sample handling (e.g., possibility that large chains or colonies are disrupted under flow conditions in the Accuri C6) could also be relevant, as are simplifications associated with conversion to ESD for intercomparison.

Images taken by the IFCB cannot always provide taxonomic resolution to the species level, but the number of operational classes seen at any given station can nonetheless provide an index of taxonomic richness within the microphytoplankton. Of the 39 classes found within the entire 2011 ICESCAPE IFCB image dataset, 26 were observed in and around this 2011 under-ice bloom (Table 2). The highest number of co-occurring classes was observed deepest into pack ice at the northwest end of the CSW section (station 57) where 22 of the 26 classes were seen (Fig. 9, top panel). Along this same section heading southeast toward Hanna Shoal, only 6 to 8 classes were found at stations near the southern ice-free end, where unclassified nanophytoplankton comprised a larger fraction of total water column phytoplankton biomass. An exception was seen at the southernmost end station directly on Hanna Shoal (station 73) which had an anomalously high number of observed classes, comparable to that seen at nearby stations in other sections also on the shoal (e.g., 80–82, 84). To the east (e.g., stations 76–82), the spatial trend in richness was similar to that seen to the west: the northernmost, ice-covered stations at the shelf break exhibited higher richness than did southern, ice-free stations well onto the shelf. Further to the east (stations 84–101), apparent richness was generally lower than in the western two sections although a slight trend was still evident,

Please cite this article as: Laney, S.R., Sosik, H.M., Phytoplankton assemblage structure in and around a massive under-ice bloom in the Chukchi Sea. Deep-Sea Res. II (2014), http://dx.doi.org/10.1016/j.dsr2.2014.03.012i

S.R. Laney, H.M. Sosik / Deep-Sea Research II ∎ (∎∎∎∎) ∎∎∎–∎∎∎

9

Fig. 6. Pooled results showing how total water column phytoplankton carbon at these 35 stations is distributed among various taxa. The values shown represent the average depth-integrated water column carbon biomass (mg m  2) in each taxon, computed from all 35 stations. Outlined bars indicate the amount of carbon in IFCB classes corresponding to cells smaller than the 10 μm threshold below which FCM-based cell carbon estimates were used. The fraction of IFCB-based carbon biomass deducted from each category for cells smaller than this threshold is also noted for the 8 instances where this fraction exceeded 0.01.

4

8−15µm

log10 IFCB (ml-1)

3

2

1

0

−1 −1

0

1

2

3

4

log10 FCM (ml-1) Fig. 7. Concentration of phytoplankton in the size range 8–15 μm ESD, comparing estimates from IFCB and FCM analysis of water from the same samples (n¼ 359, plus an additional 31 not visible on this log–log plot due to one or more 0 count; 16 of those have confidence intervals that overlap between the two methods). Error bars represent 95% confidence intervals computed assuming counts are distributed according to a Poisson distribution. A 1:1 line is indicated for reference.

with richness increasing inshore toward Hanna Shoal. The lowest richness seen in this region of the Chukchi Sea occurred at stations identified as having anomalously low levels of diatom biomass and anomalously high contributions by nanophytoplankton and picophytoplankton (e.g., stations 70–73, 99, and 100).

This more detailed analysis expands upon the preliminary assessment reported in Arrigo et al. (2012) which indicated that Chaetoceros, Thalassiosira, and Fragilariopsis dominated phytoplankton carbon in the core of the bloom. Our subsequent analysis of standard flow cytometry samples reveals that smaller nanophytoplankton also contributed materially to biomass at stations in the core of the bloom, and much more strongly in the ice-free region to the south. Throughout the study region Chaetoceros was the dominant contributor to biomass at the depths of chlorophyll maxima, except for the deeper stations along the Chukchi Shelf West section (stations deeper than 200 m, i.e., 93–101) where it ranked second behind nanophytoplankton. This more detailed analysis also indicates that Pseudonitzschia and Bacteriosira comprised unusually high fractions of microphytoplankton biomass, particularly at specific locations. The ecological significance of these ‘hotspots’ of specific taxa remains unclear, but it is possible that these may serve as indicators for specific hydrographic conditions or particular phases in the progression of under-ice blooms in this region. Additional research will be necessary to explore this aspect of taxonomic diversity in Chukchi Sea assemblages further. Our cytometry-derived estimates of depth-integrated phytoplankton carbon biomass at each station generally correspond with measured depth-integrated particulate organic carbon (POC, Fig. 9, bottom panel). The closest correspondence was seen in the southernmost stations on the Chukchi Shelf Center section (e.g., stations 87–91). On a per-sample basis these cytometry-derived estimates are highly correlated with the POC measured on water taken from the same bottles (Fig. 10). Our estimates of phytoplankton C are on average only
85% of the measured filtered POC, with no apparent bias or trend across this hundred-fold range of POC. Our cytometry-derived estimates are lower possibly

Please cite this article as: Laney, S.R., Sosik, H.M., Phytoplankton assemblage structure in and around a massive under-ice bloom in the Chukchi Sea. Deep-Sea Res. II (2014), http://dx.doi.org/10.1016/j.dsr2.2014.03.012i

S.R. Laney, H.M. Sosik / Deep-Sea Research II ∎ (∎∎∎∎) ∎∎∎–∎∎∎

10

Station 60 4

FCM IFCB

10

Station 70

Station 86

9.9 m

10.4 m

10.3 m

24.2 m

32.7 m

25.3 m

49.5 m

44.1 m

43.4 m

2

10

0

10

Merged 95% CIs −2

Number concentration (ml-1 µm-1)

10

4

10

2

10

0

10

−2

10

4

10

2

10

0

10

−2

10

10

0

1

10

10

2

0

10

1

10 Diameter (µm)

2

10 10

0

1

10

2

10

Fig. 8. Phytoplankton size distribution from selected depths at three stations, representative of the under-ice bloom (station 60), low diatom waters (station 70), and Chaetoceros-dominated waters near Hanna Shoal (station 86). Merged size distributions are shown along with components derived separately from FCM and IFCB observations including the 8–15 μm overlap region. 95% confidence intervals were computed assuming counts in each size bin are independent and distributed according to a Poisson distribution. Relatively larger confidence intervals for FCM compared to IFCB around the overlap region occur at low cell concentrations due to smaller sample volumes (
200 μl vs. several ml).

due to the screen used on the IFCB intake to prevent clogging. This would necessarily exclude any very large cells, chains, and colonies that would be otherwise included in bulk POC measurements. This potential bias would be expected to be greater when very large chains or colonies are more common in the water column, but the relative contribution of large chain-forming diatoms such as Chaetoceros to total phytoplankton carbon is not well studied, especially in Arctic systems. Published reports are in line with what we observed here in the Chukchi Sea, e.g., the report of Booth et al. (2002) that found Chaetoceros to comprise 49% of total phytoplankton carbon export from a polynya bloom in Baffin Bay, compared to the 53% we observed in this part of the Chukchi Sea. Additionally, an instrument bias may also be involved: with large chains or colonies it is typically the case that a single IFCB image does not capture the entire chain or colony, leaving an unknown amount of biomass outside the image frame, leading to underestimates of cell-specific carbon based on images. Even if such instrument and filtering biases are negligible, total POC measurements should always exceed phytoplankton C to the extent that detritus and heterotrophs also contribute to particulate C. More detailed analysis of particle properties would be required to quantify the importance of these three factors in contributing to

the patterns observed between phytoplankton C and POC, spatially and with depth across the study area.

4. Conclusions The distribution and taxonomic composition of phytoplankton assemblages are poorly characterized in most ice-covered ocean ecosystems in the Arctic, largely due to the challenges of surveying and sampling in perennial or seasonal ice cover. Marginal Arctic seas such as the Chukchi are no exception: ice cover persists for much of the year and few research ships are capable of safely operating in this remote region. Our current understanding of phytoplankton dynamics in the Arctic Ocean contains substantial gaps, and the massive under-ice bloom in 2011 is a good example of a significant but previously unobserved large-scale event in Chukchi phytoplankton ecology. This bloom likely contributed a substantial fraction of the region's yearly primary production and represented a material shift in the composition of phytoplankton assemblages in the region. Our cytometry-based approach has provided new insights into the taxonomic composition of phytoplankton assemblages in the Chukchi Sea and their relative

Please cite this article as: Laney, S.R., Sosik, H.M., Phytoplankton assemblage structure in and around a massive under-ice bloom in the Chukchi Sea. Deep-Sea Res. II (2014), http://dx.doi.org/10.1016/j.dsr2.2014.03.012i

S.R. Laney, H.M. Sosik / Deep-Sea Research II ∎ (∎∎∎∎) ∎∎∎–∎∎∎

11

Fig. 9. Top: fraction of total phytoplankton biomass found in the top 7 contributing taxonomic classes, in depth-integrated assemblages at each station. The number of different classes seen at each station (from list in Table 2, i.e., not counting miscellaneous pico- and nanophytoplankton) is noted at the top of each column. Bottom: fraction of total, pooled depth-integrated carbon biomass at each station (merged estimates of FCM-derived and IFCB-derived carbon), as well as the contribution by nano- and picophytoplankton. The general trend follows that seen in depth-integrated POC estimates, as measured on filtered samples (crosses). The gray shaded box encompasses stations within the region of the under-ice bloom.

Fig. 10. The relationship between bottle-derived POC measurements (abscissa) and pooled (FCM-derivedþ IFCB-derived) estimates of algal carbon-specific biomass (ordinate). The line indicates a 1:1 relationship.

contributions to phytoplankton carbon biomass. Our detailed analysis of Chukchi Sea phytoplankton reveals a more complex picture of assemblage structure in and around this 2011 under-ice bloom than was described in the initial assessment presented by Arrigo et al. (2012). One key new insight is the contribution of picophytoplankton and small nanophytoplankton to total assemblage biomass. A second new insight is the level of richness seen in the microphytoplankton fraction, compared to the few but dominant taxa described in the earlier 2012 report. This analysis focuses on samples taken from depths typically starting at
1.5 m, and as a result cannot resolve information on

assemblages in stratified thin layers that may have existed immediately underneath the overlying ice. Such layers can contain considerable phytoplankton biomass, such as the dense blooms of Pyramimonas spp. that were described by Gradinger (1996) and which were periodically observed during the ICESCAPE field program. Although these thin under-ice layers do not account for the production that was observed deeper in the water column where the under-ice bloom was most pronounced, phytoplankton in these thin layers may have important ecological roles in the under-ice assemblages. Future analyses into the finer-scale distributions with depth, especially in the upper part of the water column directly under the surface ice layer, may generate valuable information about the vertical segregation of important taxa not seen in abundance in the depths examined here in our study. Similarly, assemblages close to the bottom of the water column may provide key insight into the history of taxa recently inhabiting the euphotic zone and their role in carbon transport from the water column to the benthos.

Acknowledgments We thank Rob Olson for help with fabricating the Imaging FlowCytobot used in this study, Emily Peacock and Emily Brownlee for their help with manually correcting the classified cell images, and Emily Brownlee for at-sea sampling and image collection

Please cite this article as: Laney, S.R., Sosik, H.M., Phytoplankton assemblage structure in and around a massive under-ice bloom in the Chukchi Sea. Deep-Sea Res. II (2014), http://dx.doi.org/10.1016/j.dsr2.2014.03.012i

12

S.R. Laney, H.M. Sosik / Deep-Sea Research II ∎ (∎∎∎∎) ∎∎∎–∎∎∎

during ICESCAPE 2011. Alexi Shalapyonok provided assistance with Coulter Multisizer measurements. Particulate organic carbon data were provided by the Arrigo laboratory (Stanford University) as part of the ICESCAPE core data set. Quincy Allison and Susan Tolley (NASA ESPO) provided valuable logistical support for field operations. We thank the ICESCAPE research team for their efforts in developing the core ICESCAPE data set, and also the Captain, officers, and crew of USCGC Healy for their dedication and effort. Comments from two anonymous reviewers were greatly valued. This work was funded by NASA grant NNX10AG07G to SRL and HMS. References Arrigo, K.R., Perovich, D.K., Pickart, R.S., Brown, Z.W., van Dijken, G.L., Lowry, K.E., Mills, M.M., Palmer, M.A., Balch, W.M., Bahr, F., Bates, N.R., Benitez-Nelson, C., Bowler, B., Brownlee, E., Ehn, J.K., Frey, K.E., Garley, R., Laney, S.R., Lubelczyk, L., Mathis, J., Matsuoka, A., Mitchell, B.G., Moore, G.W.K., Ortega-Retuerta, E., Pal, S., Polashenski, C.M., Reynolds, R.A., Schieber, B., Sosik, H.M., Stephens, M., Swift, J.H., 2012. Massive phytoplankton blooms under Arctic sea ice. Science 336, 1408. Arrigo, K.R., Perovich, D.K., Pickart, R.S., Brown, Z.W., van Dijken, G.L., Lowry, K.E., Mills, M.M., Palmer, M.A., Balch, W.M., Bates, N.R., Benitez-Nelson, C.R., Brownlee, E., Frey, K.E., Laney, S.R., Mathis, J., Matsuoka, A., Mitchell, B.G., Moore, G.W.K., Reynolds, R.A., Sosik, H.M., Swift, J.H., 2014. Phytoplankton blooms beneath the sea ice in the Chukchi Sea. Deep-Sea Res. II, this issue [doi: 10.1016/j.dsr2.xxxx.xx.xxx]. Bluhm, B.A., Iken, K., Hopcroft, R.R., 2010. Observations and exploration of the Arctic's Canada Basin and the Chukchi Sea: The Hidden Ocean and RUSALCA expeditions. Deep-Sea Res. II 57, 1–4. Booth, B.C., Larouche, P., Bélanger, S., Klein, B., Amiel, D., Mei, Z.P., 2002. Dynamics of Chaetoceros socialis blooms in the North Water. Deep-Sea Res. Part II 49, 5003–5025. Breiman, L., 2001. Random forests. Mach. Learn. J. 45, 5–32. Christensen, J.P., Melling, H., 2010. Mesoscale distribution of springtime waters and nutrients near the continental shelf break, Beaufort Sea, Alaska. Open Oceanogr. J. 4, 115–136. Cota, G., Pomeroy, L., Harrison, W., Jones, E., Peters, F., Sheldon, W.J., Weingartner, T., 1996. Nutrients, primary production and microbial heterotrophy in the southeastern Chukchi Sea: Arctic summer nutrient depletion and heterotrophy. Mar. Ecol. Prog. Ser. 135, 247–258. Cottrell, M.T., Kirchman, D.L., 2009. Photoheterotrophic Microbes in the Arctic Ocean in Summer and Winter. Appl. Environ. Microbiol. 75, 4958–4966. Fortier, M., Fortier, L., Michel, C., Legendre, L., 2002. Climatic and biological forcing of the vertical flux of biogenic particles under seasonal Arctic sea ice. Mar. Ecol. Prog. Ser. 225, 1–16. Frey, K.E., Perovich, D.K., Light, B., 2011. The spatial distribution of solar radiation under a melting Arctic sea ice cover. Geophys. Res. Lett. 38, L22501. Gradinger, R., 1996. Occurrence of an algal bloom under Arctic pack ice. Mar. Ecol. Prog. Ser. 131, 301–305. Grebmeier, J.M., Harvey, H.R., 2005. The Western Arctic Shelf–Basin Interactions (SBI) project: an overview. Deep-Sea Res. II 52, 3109–3115. Grebmeier, J.M., Harvey, H.R., Stockwell, D.A., 2009. The Western Arctic Shelf–Basin Interactions (SBI) project, volume II: an overview. Deep-Sea Res. II 56, 1137–1143.

Hill, V., Cota, G., Stockwell, D., 2005. Spring and summer phytoplankton communities in the Chukchi and Eastern Beaufort Seas. Deep-Sea Res. II 52, 3369–3385. Huang, S., Wilhelm, S.W., Harvey, H.R., Taylor, K., Jiao, N., Chen, F., 2012. Novel lineages of Prochlorococcus and Synechococcus in the global oceans. ISME J. 6, 285–297. Joo, H.M., Lee, S.H., Jung, S.W., Dahms, H.-U., Lee, J.H., 2012. Latitudinal variation of phytoplankton communities in the western Arctic Ocean. Deep-Sea Res. II 81-84, 3–17. Liu, H., Suzuki, K., Minami, C., Saino, T., Watanabe, M., 2002. Picoplankton community structure in the subarctic Pacific Ocean and the Bering Sea during summer 1999. Mar. Ecol. Prog. Ser. 237, 1–14. Menden-Deuer, S., Lessard, E.J., 2000. Carbon to volume relationships for dinoflagellates, diatoms, and other protist plankton. Limnol. Oceanogr. 45, 569–579. Min Joo, H., Lee, S.H., Won Jung, S., Dahms, H.-U., Hwan Lee, J., 2012. Latitudinal variation of phytoplankton communities in the western Arctic Ocean. Deep-Sea Res. II 81–84, 3–17. Moberg, E.A., Sosik, H.M., 2012. Distance maps to estimate cell volume from twodimensional plankton images. Limnol. Oceanogr. -Methods 10, 278–288. Mundy, C.J., Gosselin, M., Ehn, J., Gratton, Y., Rossnagel, A., Barber, D.G., Martin, J., Tremblay, J.-É., Palmer, M., Arrigo, K.R., Darnis, G., Fortier, L., Else, B., Papakyriakou, T., 2009. Contribution of under-ice primary production to an ice-edge upwelling phytoplankton bloom in the Canadian Beaufort Sea. Geophys. Res. Lett. 36, L17601. Murphy, L.S., Haugen, E.M., 1985. The distribution and abundance of phototrophic ultraplankton in the North Atlantic. Limnol. Oceanogr. 30, 47–58. Neukermans, G., Rick A. Reynolds, Stramski, D., 2014. Contrasting inherent optical properties and particle characteristics between an under-ice phytoplankton bloom and open water in the Chukchi Sea. Deep-Sea Res. II, this issue [doi: 10.1016/j.dsr2.xxxx.xx.xxx]. Olson, R.J., Chisholm, S.W., Zettler, E.R., Altabet, M.A., Dusenberry, J.A., 1990a. Spatial and temporal distributions of prochlorophyte picoplankton in the North Atlantic Ocean. Deep-Sea Res. 37, 1033–1051. Olson, R.J., Sosik, H.M., 2007. A submersible imaging-in-flow instrument to analyze nano- and microplankton: Imaging FlowCytobot. Limnol. Oceanogr.: Methods 5, 195–203. Olson, R.J., Zettler, E.R., Anderson, O.K., 1989. Discrimination of eukaryotic phytoplankton cell types from light scatter and autofluorescence properties measured by flow cytometry. Cytometry, 10. Olson, R.J., Zettler, E.R., Armbrust, E.V., Chisholm, S.W., 1990b. Pigment, size and distribution of Synechococcus in the North Atlantic and Pacific oceans. Limnol. Oceanogr. 35, 45–58. Perovich, D.K., Roesler, C.S., Pegau, W.S., 1998. Variability in Arctic sea ice optical properties. J. Geophys. Res. 103, 1193–1208. Sherr, E.B., Sherr, B.F., Wheeler, P.A., Thompson, K., 2003. Temporal and spatial variation in stocks of autotrophic and heterotrophic microbes in the upper water column of the central Arctic Ocean. Deep-Sea Res. Part I 50, 557–571. Sosik, H.M., Olson, R.J., 2007. Automated taxonomic classification of phytoplankton sampled with imaging in-flow cytometry. Limnol. Oceanogr.: Methods 5, 204–216. Spreen, G., Kaleschke, L., Heygster, G., 2008. Sea ice remote sensing using AMSR-E 89-GHz channels. J. Geophys. Res.-Ocean. 113 (C02S03). Strass, V.H., Nö thig, E.M., 1996. Seasonal shifts in ice edge phytoplankton blooms in the Barents Sea related to the water column stability. Polar Biol. 16, 409–422. Sukhanova, I.N., Flint, M.V., Pautova, L.A., Stockwell, D.A., Grebmeier, J.M., Sergeeva, V.M., 2009. Phytoplankton of the western Arctic in the spring and summer of 2002: structure and seasonal changes. Deep-Sea Res. II 56, 1223–1236.

Please cite this article as: Laney, S.R., Sosik, H.M., Phytoplankton assemblage structure in and around a massive under-ice bloom in the Chukchi Sea. Deep-Sea Res. II (2014), http://dx.doi.org/10.1016/j.dsr2.2014.03.012i