The advective origin of an under-ice spring bloom in the Arctic Ocean using multiple observational platforms

The advective origin of an under-ice spring bloom in the Arctic Ocean using multiple... Under-ice blooms of phytoplankton in the Chukchi Sea have been observed, with strong implications for our understand- ing of the production regimes in the Arctic Ocean. Using a combination of satellite remote sensing of phytoplankton bio- mass, in situ observations under sea ice from an autonomous underwater vehicle (AUV), and in vivo photophysiology, we examined the composition, magnitude and origin of a bloom detected beneath the sea ice Northwest of Svalbard (Southern −3 Yermak Plateau) in May 2010. In situ concentration of up to 20 mg chlorophyll a [Chl a] m , were dominated by the north- ern planktonic spring species of diatoms, Thalassiosira nordenskioeldii, T. antarctica var. borealis, Chaetoceros socialis species complex and Fragilariopsis oceanica. These species were also found south of the marginal ice zone (MIZ). Cells in the water column under the sea ice were typically high-light acclimated, with a mean light saturation index (E ) of 138 μmol −2 −1 photons m  s and a ratio between photoprotective carotenoids (PPC) and Chl a (w:w) of 0.2. Remotely sensed data of [Chl a] showed a 32,000 km bloom developing south of the MIZ. In effect, our data suggest that the observed under-ice bloom was in fact a bloom developed in open waters south of the ice edge, and that a combination of northward-flowing water masses and southward drifting sea ice effectively positioned the bloom under the sea ice. This have implications for our general understanding of under-ice blooms, suggesting that their origin and connection with open water may be differ - ent in different regions of the Arctic. Keywords Sea-ice algae · Under-ice bloom · Phytoplankton · Photoacclimation · Advection of cells · Arctic · Satellite · Autonomous underwater vehicle · Chlorophyll a · Pulse amplitude modulated fluorescence (PAM) * Geir Johnsen School of Marine Science and Policy, University geir.johnsen@ntnu.no of Delaware, Newark, DE 19958, USA California Polytechnic State University, San Luis Obispo, Centre for Autonomous Marine Operations and Systems, CA 93407, USA Department of Biology, Norwegian University of Technology and Science (NTNU), 7491 Trondheim, Department of Environment and Mapping, Norwegian Polar Norway Institute (NP), 9296 Tromsø, Norway 2 8 Department of Arctic Biology, University Centre on Svalbard Scottish Association for Marine Science (SAMS), Oban, (UNIS), 9171 Longyearbyen, Norway Argyll PA37 1QA, UK Department of Arctic and Marine Biology, UiT The Arctic University of Norway, 9037 Tromsø, Norway Section of Marine Biogeochemistry and Oceanography, Norwegian Institute of Water Research (NIVA), 0349 Oslo, Norway Vol.:(0123456789) 1 3 1198 Polar Biology (2018) 41:1197–1216 physiological status of phytoplankton (Leu et al. 2011; Introduction Pettersen et al. 2011; Hovland et al. 2013; Hancke et al. 2014; Hovland et al. 2014). The key properties of the water Sea ice plays a dual role in the control of high-latitude masses and ice conditions in this region are well docu- phytoplankton blooms by influencing both light shading mented in terms of current speed and direction, tempera- and stratification (Arrigo et al. 2012; Assmy et al. 2017; ture, salinity and density (Rudels et al. 2005; Ingvaldsen Kauko et  al. 2017). Importantly, primary production and Loeng 2009; Lind and Ingvaldsen 2012; Assmy et al. (PP) in regions with an annual ice cover comes from two 2017). main sources, i.e. ice-associated and open water blooms Pelagic phytoplankton blooms in the MIZ are typically (Søreide et al. 2010). These two main sources of PP have initiated from mid-April to June in the waters west of Spits- traditionally been considered as temporally and spatially bergen and in the Barents Sea, and are associated with both distinct. It has been suggested that withdrawal of ice will Arctic cold water species and/or temperate species (McMinn generally lead to an increase in PP (Drinkwater 2011; and Hegseth 2007; Sakshaug et al. 2009a, b). In contrast, Slagstad et al. 2015), and that the nutrient-dependent new blooms in the perennial ice zone (PIZ) are generally thought production may not increase proportionally with increas- of as mainly being associated with ice algae, with irradiance ing light intensity due to insufficient access to essential in the water column below ice too low to support a pelagic nutrients. Recently, new reports of a “third” source of PP bloom (Sakshaug et al. 2009b; Berge et al. 2015). Recently, were presented, under-ice phytoplankton blooms (see e.g. Arrigo et al. (2012) provided new and important insights Arrigo et al. 2012). A phytoplankton bloom beneath the into the occurrence and development of pelagic blooms sea ice is a phenomenon that is poorly understood, they are under Arctic sea ice. A recent report from the same area as hard to detect and effectively invisible from satellites, and this study (NW of Svalbard), shows the dynamics between they occur in habitats that are logistically hard to sample. sea ice, the local current systems and a phytoplankton bloom Yet, there are several reports documenting their existence dominated by Phaeocystis pouchetii (Assmy et al. 2017). (Gradinger 1996; Mundy et al. 2009; Boetius et al. 2013; The light regime (light climate) is a function of three Assmy et al. 2017), but few that estimate their role in the variables, the irradiance (E) in the PAR region, E in total productivity and carbon budget of the Arctic. In the PAR −2 −1 μmol quanta m s (photosynthetic active radiation, Chukchi Sea, Arrigo et  al. (2012, 2014) estimated that 400–700 nm), the spectral irradiance, E (λ) in μmol quanta 90% of the total PP occurred under sea ice. Importantly, as −2 −1 −1 m s nm , and day length in hours (Falkowski and most current models rely on remote sensing for input and Raven 1997; Sakshaug et al. 1997, 2009b). Photoacclima- validation of phytoplankton biomass, occurring beneath an tion (short-term physiological adjustment of growth rate to ice cover, phytoplankton biomass in an under-ice bloom light climate) is divided in short- (seconds–minutes) and is at best poorly represented in contemporary modelling long-term (hours–days) responses (MacIntyre et al. 2000; efforts (Vancoppenolle et al. 2013). Under-ice blooms are Brunet et al. 2011; Valle et al. 2014). Short-term photoac- therefore important to map and understand, as they may climation includes the xanthophyll cycle for photoprotec- represent a significant component that is missing in the tion, an important part of non-photochemical quenching total production regime of the Arctic Ocean. Also, the (NPQ). Long-term photoacclimation is defined as change sea-ice zone has been identified as the region with largest in ratio between light-harvesting pigments (LHP) and pho- model uncertainty regarding Arctic PP and biophysical toprotective carotenoids (PPC), photosynthetic parameters, interactions (Assmy et al. 2017; Kauko et al. 2017). Know- enzymatic activities involved in photosynthesis and respira- ing that under-ice blooms have been reported from across tion, and changes in cell volume and chemical composition the central Arctic Ocean and surrounding shelf seas, yet (Brunet et al. 2011). not taken into account in contemporary modelling efforts, Qualitative and quantitative pigment analyses of chloro- there is a clear and important gap in knowledge that needs phylls and carotenoids by high performance liquid chroma- to be filled: A quantitative and mechanistic understanding tography (HPLC) can be used for chemotaxonomical and of the processes that control PP in ice-covered waters. photoacclimation information. Different pigment groups The concentration of Chlorophyll a, [Chl a], is the main can be identified (Johnsen and Sakshaug 2007; Roy et al. source of information to detect, map and monitor phyto- 2011) such as the carotenoid fucoxanthin (spring blooms plankton biomass (review in Johnsen et al. 2011a, b). The of diatoms), Chlorophyll c (prymnesiophytes such as differences in water mass-related effects on phytoplankton Emiliania huxleyi and Phaeocystis pouchetii) and small dynamics in the Svalbard area, often seen as the balance prasinoxanthin-containing prasinophytes, typical for this between the warm and saline water masses of Atlantic area (Pettersen et al. 2011). Likewise, HPLC-isolated pig- origin versus the colder and fresher water masses of Arctic ments comprise information of light-harvesting pigments origin, is reflected in the differences in biodiversity and (LHP) and photoprotective carotenoids (PPC), (Johnsen 1 3 Polar Biology (2018) 41:1197–1216 1199 and Sakshaug 2007; Brunet et  al. 2011; Johnsen et  al. vertically and horizontally below 1 m depth and under sea 2011a). The degradation state of chlorophylls can be also ice, in contrast to satellite-derived data. used to indicate the physiological state of the phytoplank- Based on this, we report on an under-ice bloom consisting ton cells, i.e. pre-bloom, bloom and post-bloom state of HL-acclimated cells in the Eurasian sector of the Arctic (Johnsen and Sakshaug 2000; Roy et al. 2011). Ocean (southern part of the Yermak Plateau, Fig. 1)resulting Through photoacclimation, the cells are optimizing the from northward advection of water masses. This study shows light harvesting properties, the corresponding photosyn- that by combining information from three observational plat- thetic electron transport rates (ETR, see Materials and forms (satellite, AUV and ship) we obtain a “picture” of Methods) and pigment functions (light harvesting ver- overall biomass in both open waters (satellite) and under sus photoprotective) to provide maximum growth rates sea ice (AUV and ship). The ship-based CTD equipped with (Falkowski and Raven 1997; Sakshaug et al. 1997; Brunet an in situ Chl a fluorometer and Niskin bottles for water et al. 2011). Chlorophyll a fluorescence-based photosyn- sampling (phytoplankton) made it possible to study the thetic relative ETR rates (rETR), are typically measured by taxa diversity and photoacclimation status (photosynthetic means of PAM (Genty et al. 1989; Kromkamp and Forster 2003; Hancke et al. 2008a, b) or by fast repetition rate fluorometer, FRRF (Kolber and Falkowski 1993; Suggett et al. 2009). The photoacclimation status of phytoplankton can be identified by looking at the photosynthetic param- eters termed maximum light utilization coefficient (α , ETR similar to α for C-based P vs. E curves), Chl a fluo- rescence-based maximum rETR rate, rETR (similar to max maximum photosynthetic rate, P for C-based P vs. E max curves), and the corresponding light saturation parameter −2 −1 (E , = rETR /α in μmol quanta m s ) (Behrenfeld k max ETR et al. 2004; Hancke et al. 2008b; Schuback et al. 2016). Using E we can divide cells into high-light- (HL, cells growing in “surface open waters”) and low-light- (LL, cells grown “under sea ice” or deep waters) acclimated cells (Sakshaug et al. 2009b). The in situ E (μmol photons −2 −1 m s ) of spring bloom forming phytoplankton are typi- −2 −1 cally 50–60 μmol photons m s at 0–10 m and 30–50 m depth, respectively in May–June (numbers from the Bar- ents Sea using C-incubated cells, at latitudes comparable to the west of Spitsbergen at 78°N (reviewed by Sakshaug et al. 2009b). To address our key scientific question “are the observed under-ice phytoplankton blooms a result of local production, or are most of the algal cells transported from open and well- lit waters south of the Marginal Ice Zone (MIZ)”, we com- bined information of phytoplankton biomass [Chl a] data from: Satellite (open water bloom south of ice edge covering 32,000  km ), an AUV deployed under sea ice [providing areal coverage of under-ice light transmission patchiness, (Chl a), temperature and salinity; horizontally 80,000 m and vertically 16,500 m ], vessel-based in situ profiles and corresponding work on cells in the laboratory (in vivo and Fig. 1 Remotely sensed sea surface phytoplankton bloom covering 2 −3 32,000  km (green–yellow area), defined as [Chl a]  >  4  mg  m , in in vitro). The major “pro” of using polar orbiting satellites May 2010 (mean values) W and N of the Svalbard archipelago (black equipped with ocean colour multispectral imagers for [Chl −3 area). The phytoplankton biomass (colour bar in mg Chl a m ), sea- a] are the covering of large areas while the “cons” are that ice distribution (white areas) with ice stations (red circles, northern- the information is restricted to water surface (upper 1–2 m), most stations at 31 km north of MIZ). The ice edge at 16 and 17 May indicated a southward ice movement during the campaign. Note that dependent on illumination from the sun and absence of the southernmost station was in open waters during the actual tran- clouds (review in Johnsen et al. 2011b). Autonomous under- sect. Image based on Envisat satellite with MERIS multispectral water vehicles (propelled AUV’s) may cover the [Chl a] imager. For details, see Materials and methods 1 3 1200 Polar Biology (2018) 41:1197–1216 parameters) in addition to pigment speciation and function of the living cells sampled under the sea ice. Materials and methods The measurements were made to the north and south of the Marginal Ice Zone (MIZ) north of Svalbard 15–18 May 2010 (Fig. 1, Table 1). During sampling, the MIZ was at station (St) 6, St 7 in open waters (Fig. 1–2, Table 1) and the sea-ice coverage at St 1–5 was close to 100% with an ice thickness of 1.1–1.9 m and with some leads present. At station 1–3 the ice cores were ranging from 122 cm ± 12 (n = 11, SD) to 146 cm ± 40 (n = 8, SD). In addition, the corresponding snow depth on top of sea ice were ranging from 23 to 30 cm (n = 16). Ice cores were melted for ice algal analyses in cool- ing lab at 4°C in research vessel under dim light. Observational platforms and sensors Envisat satellite data from the MEdium resolution imaging spectrometer (MERIS) were used to identify and map the −3 extent of the phytoplankton bloom (Chl a, mg m ) south of the MIZ (Fig. 1–2) through a monthly mean [Chl a] map of the area. All available reduced resolution data (MERIS-RR) from May 2010 with a pixel size of 1 km from the MERIS archive at Brockmann Consults were used. The Algal-1 Chl a product algorithm from the 3rd reprocessing was pro- cessed using the software BEAM version 5.0.1 to produce the L3 binning product. A REMUS 100 AUV from Hydroid Inc. (Moline et al. 2005) was used under the sea ice at St 3 the 16 May 2010 at 11:00–12:15 local time (Figs. 3, 4, 5, Tables 1, 2, 3). Fig. 2 Ship-based under-ice 0–200  m vertical distribution of salin- The REMUS 100 was equipped with both ultra short base ity (upper panel), temperature (mid panel) and [Chl a] (lower panel) line (USBL) and long base line (LBL) transducers for through a transect line of 31  km north of the MIZ. Sea-ice stations positioning and navigation under the ice. With a moving 1–6, station 3 was the site for AUV survey, station 1–2 northernmost ice field (typically at ≈ 0.5 knots, data from ship log), it stations (right side), MIZ is the marginal ice zone and open water (station 7, southernmost) Table 1 Overview of ice station 1–7 with date, time of day (local depth of water samples for pigment (all pigments by HPLC and cor- time), latitude, longitude, CTD and Chl a fluorometer vertical tran- responding in vitro Chl a analyses) and in vivo photosynthesis sect from 0 to 200  m depth (all stations) from RV Helmer Hanssen, Station Date May Time Latitude Longitude Water sample Photosynthesis sam- Pigments sample AUV sam- depths (m) ple depths (m) depths (m) ple depths (m) 1 15 10–13 80.665°N 5.048°E 2, 10 2, 10 2, 10 2 15 22–01 80.673°N 5.100°E 2, 10, 40 2, 10, 40 2, 10, 40 3 16 10–18 80.640°N 5.024°E 10, 45 10, 45 10, 45 5–55 4 16 23–24 80.607°N 4.841°E 5 17–18 24–01 80.566°N 4.918°E 6 17–18 24–01 80.528°N 5.097°E 7 18 03–04 80.416°N 5.307°E Downwelling irradiance E and vertical net hauls of phytoplankton (both from surface to 50 m depths) were carried immediately after CTD profiles PAR 1 3 Polar Biology (2018) 41:1197–1216 1201 Fig. 3 AUV-based under sea-ice photomosaic images at 10  m depth images along track line shown on right side. White frame indicates (horizontal survey, covering 80,000  m ) under the sea ice the 16th area in b and c. b Complete photomosaics along grid line at 10  m May 2010 at  11–12:15 local time at ice station 3. Deployment at depth. c Details (from b, white frame) of geo-tagged photomosaic −1 80°640′ N, 5°024′ E at average speed of 1.5  m  s . At noon, under- images of under-ice light climate and ice morphology as a function of −2 −1 ice E was between 1.2 and 2.1 μmol photons m s at 10 m depth thick sea ice (dark picture frames), leads and thin ice (bright frames), (St 1–3, Fig. 4). At 22:00, E was typically about 30% of noon values blue areas indicating melt ponds (blue frames) and areas with high (see results). a Schematic image of underwater AUV photomosaic [Chl a] (green frames, indicating ice algae and phytoplankton). Cor- images of underside of sea ice with camera facing upwards at 10  m responding horizontal and vertical AUV transect lines for Chl a, T depth during horizontal transect grid line. Indications on photomosaic and S in Fig. 5 Fig. 4 Downwelling irradiance (E ) as a function of depth at ice station 1–3. The profiles indi- cate maximum irradiance values (measured from leads) was important to ensure that the AUV was navigating rela- Progress of the AUV mission was monitored in real time tive to the ice features. The average velocity of AUV was via acoustic modem (hydroid ranger). Two mission plans −1 1.5 m s (2.9 knots) and the total mission time was 1.15 h were run; the first was an under-ice zig-zag pattern gener - with total mission length of 6663 m. Three transponders ating three vertical transects spanning depths of 5–50 m, were placed under the sea ice; two were used for LBL the second was an under-ice grid pattern (20 m spacing) navigation for both the AUV data collection providing ver- at 10 m fixed depth to ensure clearance of pressure ridges. tical profiles and horizontal area coverage at 10 m depth The REMUS 100 was equipped with a Neil Brown CTD for under sea ice. A third transponder (homing transponder) temperature and salinity and depth measurements. In situ was used for USBL navigation to ensure return of the vehi- [Chl a] was measured with a BBFL2 Eco-Puck, with excita- cle to a particular opening in the ice for retrieval. tion wavelength maximum at 470 nm and corresponding Chl 1 3 1202 Polar Biology (2018) 41:1197–1216 Fig. 5 AUV-based vertical (left panels, black lines) and horizontal (right panels, black lines) mapping of phytoplank- ton biomass ([Chl a], lower panel), temperature (mid panel) and salinity (upper panel) of water masses under sea ice at ice station 3. Spatial coverage of −3 Chl a (mg m ), salinity (PSU) and temperature (°C) using Remus 100 AUV in vertical zig-zag mode transect covering 16,500 m and a horizontal grid area at 10 m depth covering 80,000 m (400 × 200 m) a emission measured at 695 nm (Wet Labs, Oregon, USA, 200 m depth along a transect from stations 1–7 using a calibrated by producer prior to cruise). In addition to these Sea-bird 911 CTD (Sea-Bird Electronics, Inc., USA). In sensors, an upward-looking DC1400 SeaLife digital under- addition, the Sea-bird 911 was equipped with a calibrated water camera was mounted in the nose of AUV to provide (prior to cruise) in situ Seapoint (Seapoint Inc, USA) Chl −3 geo-tagged (through time-synchronization with AUV data) a fluorometer (Chl a, mg  m ), and Niskin water sam- qualitative pictures of the underside of the sea ice provid- pler rosette (5 L) for collecting water samples for in vivo ing information of light field, ice morphology, water colour analyses of phytoplankton taxa, photosynthetic parameters (blue, green and white) related to inherent optical proper- (onboard research vessel) and for harvesting cells on fil - ties (IOP) of water and its constituents [phytoplankton, col- ters for HPLC pigment characterization (light-harvesting oured dissolved organic matter and total suspended matter, pigments, photoprotective carotenoids and degraded pig- (Johnsen et al. 2009)]. Vehicle position, time and the lens ments). At each station, immediately after the CTD pro- geometry were used in the mosaic reconstruction of the grid files, vertical net hauls of phytoplankton samples (from portion of the mission (Fig. 3). below Chl a maximum to the surface) were carried out The research vessel “RV Helmer Hanssen” (UiT, the using a phytoplankton net with 20 μm mesh size. Water Arctic University of Norway) was used for water col- sample depths are indicated in Tables 1, 2, 3. Phytoplank- umn profiles of salinity, temperature and depth from 0 to ton samples were kept alive in polyethylene containers at 1 3 Polar Biology (2018) 41:1197–1216 1203 4 °C. Cell counts and identification were performed on live cells 1–4 h after sampling. No fixatives were added to ensure that fragile naked cells were included in the analy- sis. Detailed investigation of algal eco-physiology was undertaken at sea-ice sites (St 1–3); note St 3 was also the AUV survey location. This was followed by a transect line southward (St 4–7) with ship-based in situ CTD and Chl a fluorescence profiles from surface to 200 m depth (Fig. 1, 2, 3, Table 1). −2 Downwelling irradiance (E , μmol photons m d-PAR −1 s ), was measured using two different light sensors and data loggers. Irradiance depth profiles (0–40 m at station 1–3) were measured using a calibrated Li-Cor UW1800 2π irradiance (cosine corrected) light collector connected to a Li-Cor 1100 data logger with an underwater cable to 50 m depth (1-m depth increment from surface to 10 m, 5-m depth increments for depths > 10 m, Fig. 4). E d-PAR in open deck incubators (measuring photosynthesis versus irradiance in kelp, data not shown) from a DIVING-PAM irradiance sensor (planar), measuring diurnal time series (15 min between measurements) at 0.5 m depth. Identification of phytoplankton groups Light microscopic (LM) identification of living cells from phytoplankton nets was performed to identify major taxa (class to species level). Cell identification by microscopy (200 and 400 × magnification, using at least 4 subsam- ples) was further used to separate species into “northern cold water species” and “cosmopolitan species” (Sakshaug et al. 2009b). Ice algae were sampled from the lower part of ice cores and compared with species composition from the water column in order to be able to determine pos- sible origin of the species in the water column according to (von Quillfeldt et al. 2009). In addition, HPLC-derived pigment chemotaxonomy gives information of major pig- ment groups and pigment-based biomass (including small and fragile cells that are hard to detect using LM), such as sub-groups of Chromophytes (Chl c-containing algae), Chlorophytes (Chl b-containing classes) and phycobilipro- tein-containing cyanobacteria and eukaryotic algae present (Johnsen et al. 1994; Johnsen and Sakshaug 2007; Johnsen et al. 2011a, b; Pettersen et al. 2011; Roy et al. 2011). No traces of prasinoxanthin (marker for small prasinophytes, pigment group 6 in Johnsen and Sakshaug (2007)), phy- cobiliprotein-containing cryptophytes (alloxanthin) and cyanobacteria (zeaxanthin and b-carotene), affecting pho - tosynthetic parameters (  ) were found (Table  2, see PSII photosynthetic parameters below). 1 3 −3 Table 2 Phytoplankton pigment data (HPLC, mg pigment m ) at ice station (st) 1 and 3 Station Depth (m) Chl a + Phaeo Chl a Chlc Chl b Fuco Diadino Diato LHP/Chl a + Phaeo LHP/Chl a PPC/Chl a + Phaeo PPC/Chl a 1+2 1 2 10.8 ± 0.5 9.8 ± 0.5 3.6 ± 0.17 6.0 ± 0.3 1.9 ± 0.1 0.89 ± 0.01 0.98 ± 0.01 0.17 ± 0.01 0.19 ± 0.01 10 10.7 ± 0.8 9.6 ± 0.6 3.4 ± 0.25 5.6 ± 0.4 1.6 ± 0.1 0.84 ± 0.02 0.94 ± 0.02 0.15 ± 0.01 0.17 ± 0.00 3 10 9.9 ± 0.3 7.9 ± 0.4 2.0 ± 0.43 0.2 ± 0.01 4.0 ± 0.1 1.2 ± 0.07 0.5 ± 0.01 0.63 ± 0.07 0.79 ± 0.1 0.17 ± 0.01 0.22 ± 0.02 45 7.2 ± 0.6 6.2 ± 0.4 1.9 ± 0.09 3.4 ± 0.1 1.1 ± 0.03 0.75 ± 0.06 0.87 ± 0.06 0.15 ± 0.01 0.17 ± 0.01 Total Chl a + Phaeo includes phaeophorbide a, phaeophytin a and chlorophyllide a. Standard error of average (SE, n = 3) indicated as ± after [pigment]. Light-harvesting pigments (LHP): Chl a, Chl c , Fucoxanthin (Fuco). Photoprotective carotenoids (PPC): Sum of diadinoxanthin (Diadino) and diatoxanthin (Diato). Pigment ratios in w:w. Station 3 is the site of AUV survey. Only 1+2 −3 traces (< 0.02 mg pigment m ) of Chl c and acyloxy-fucoxanthins were found under ice (indicating low biomass of prymnesiophytes such as Phaeocystis pouchetii, data not shown) 3 1204 Polar Biology (2018) 41:1197–1216 −2 −1 Table 3 Photosynthetic parameters from ice station 1–3: rETR (  × E ),  ,  , E (μmol  m s ) and non-photosynthetic max PAR ETR k PSII PSII - max quenching (NPQ) of phytoplankton Station Depth (m) rETR  (± SE) R  E NPQ E  = 180 NPQ E  = 500 max ETR k PAR PAR PSII - max (± SE) 1 2 105 ± 7.1 0.69 ± 0.10 0.88 0.60 ± 0.01 152 ± 24.3 0.35 ± 0.03 1.47 ± 0.20 10 99 ± 7.6 0.64 ± 0.10 0.85 0.62 ± 0.01 155 ± 27.0 0.23 ± 0.07 0.81 ± 0.08 2 2 144 ± 20.8 0.59 ± 0.11 0.79 0.61 ± 0.02 244 ± 57.7 0.23 ± 0.10 1.41 ± 0.30 10 96 ± 7.7 0.76 ± 0.15 0.80 0.62 ± 0.02 127 ± 27.0 0.89 ± 0.37 1.36 ± 0.30 40 81 ± 5.4 0.77 ± 0.14 0.84 0.63 ± 0.02 106 ± 20.4 0.58 ± 0.16 2.06 ± 0.47 3 10 89 ± 6.1 0.87 ± 0.17 0.82 0.61 ± 0.01 103 ± 21.2 0.23 ± 0.09 0.82 ± 0.08 45 68 ± 4.9 0.87 ± 0.20 0.78 0.61 ± 0.01 78 ± 18.9 0.43 ± 0.16 2.02 ± 0.58 Mean 97 ± 8.5 0.75 ± 0.14 0.80 0.61 ± 0.01 138 ± 28.0 0.42 ± 0.18 1.42 ± 0.36 −2 −1 −2 −1 NPQ derived from RLC curves at E of 180 μmol m s , close to maximum E (mean value of 138 μmol m s ) and corresponding light- PAR k −2 −1 saturated photosynthesis (E of 500 μmol m s ). Standard error of average (SE, n = 3) indicated as ± after photosynthetic parameters. Sta- PAR tion 3 is the site of AUV survey in 5  mL of methanol (24  h, 4  °C) and the extracts were [Chl a] in living cells (in vivo and in situ) refiltered through a 0.2 μm sterile Microsart syringe filter to and in extracts (in vitro) remove any potential light-scattering particles and debris. −1 For spectrophotometric [Chl a] measurements, filter The in vitro [Chl a], μg L , measurements were taken from extracts were measured and the optical density (OD) spectra Niskin water samples using both turner designs fluorometer (300–800 nm, 1 cm cuvette) were obtained, using methanol (10-AU-005-CE, Sunnyvale, CA, USA), a Shimadzu Model (MeOH) as blank (OD at 750 nm). The concentration of Chl 1240 UV mini spectrophotometer (see details below) and −3 a (mg m ) was calculated using the OD red peak of Chl a, a high performance liquid chromatograph equipped with a at 665 nm in vitro and the extinction coefficient for Chl a in diode array detector (Hewlett Packard series 1100 HPLC) methanol at 665 nm (Mackinney 1941), detailed in Chauton according to Rodriguez et al. (2006). Comparison between et al. (2004). in  situ fluorescence (CTD and AUV Chl a fluorometer) versus analyses on extracts in the lab (fluorometer, spec - trophotometer and HPLC) are detailed in discussion. Lab HPLC pigment identification and analysis fluorometer (turner designs) was calibrated by the laboratory spectrophotometer using Chl a standard from Sigma-Aldrich Glass fiber filters for HPLC analysis of chlorophylls and (C6144 Chl a) according to Knap et al. (1996). Due to pho- carotenoids were immediately put in a freezer at − 20 °C toacclimation affected by PQ (photosynthetic quenching, during the cruise (1  week) and moved to −  80  °C for ) and NPQ (mainly due to photoprotective pigments PSII 1 month before analysis in laboratory. The frozen filters were and low internal pH inside chloroplast at high photosyn- used to provide pigment extracts for HPLC (extraction time thetic rates), there is not a direct relationship between in vivo 24 h at 4 °C in N (g) atmosphere) in 1.6 mL 100% methanol (often called for semi-quantitative [Chl a] measurements, (in 2 mL glass vials). Extracts were refiltered (3 mL syringe (Babin 2008)) and in  vitro [Chl a]. Therefore, a precise with 0.2 μm PTFE filter) prior to being measured. HPLC calibration of in situ fluorometers is hard to obtain simply was used to separate pigments, using a Waters Symmetry because of light-induced variation of Chl a fluorescence C8 column (4.6 × 150 mm, 3.5 μm pore size), a Hewlett emitted from the cells (reviews by Babin 2008; Suggett et al. Packard Series 1100 HPLC system with a quaternary pump 2011b; Johnsen et al. 2011a). and auto sampler using the procedure outlined in Rodríguez Spectrophotometric and fluorometric [Chl a] analy- et al. (2006). 77 μL sample was mixed with 23 μL distilled ses were performed onboard the vessel, whilst HPLC was water to increase polarity and better separation of pigments. done at Norwegian University of Science and Technology The mobile phases were methanol:acetonitrile:aqueous pyri- (NTNU). Water samples for cell identification, pigment dine (ratio 50:25:25), acetonitrile:acetone (ratio 80:20) and (HPLC) and Chl a samples from CTD Niskin bottles were methanol (for cleaning). The identification of pigments was taken immediately after the CTD profiles and the AUV sur - based on retention time and the OD spectra of the pigment vey and were filtered through Whatman GF/F glass fiber was obtained with a diode array optical density detector filters (25 mm diameter) until the filters were coloured (typi- (350–800 nm) using our own pigment standards according cally 500 ml water volume due to bloom conditions). Prior to to Rodríguez et al. (2006). being measured, the algal pigments on filters were extracted 1 3 Polar Biology (2018) 41:1197–1216 1205 NPQ was calculated using Eq. 5 [Stern–Volmer equation, Photosynthetic parameters (Lakowicz 1983)]: The photosynthetic parameters from which E were calcu- F − F lated, were obtained from RLC (rapid light curve) versus E m NPQ = , (5) measured using the PAM method using a Phyto-PAM (Walz, F Germany) according to Hancke et al. (2008a, b) and Nymark where F denotes the maximum quantum yield of cells et al. (2009). This method measures the operational quan- tum yield of charge separations in PSII (Eq. 1) at increas- acclimated to dark in 5 min prior to RLC incubations (Max- well and Johnson 2000; Valle et al. 2014). ing irradiances (Genty et al. 1989). Three mL seawater was filtered through a 0.2 µm sterile filter (Microsart) and used as a blank. Afterwards, the same cuvette was used for repli- cate samples (n = 3, water from three different Niskin bot- Results tles). Fluorescence cuvette temperature was controlled with a Peltier cell (US-T/S, Walz) mimicking in situ tempera- Remote sensing of phytoplankton spring bloom ture ± 0.2 °C during incubation. Incubation irradiances for RLC were 1, 2, 4, 7, 14, 28, 56, 112, 176, 240, 304, 368, 496 The distribution of the phytoplankton spring bloom from −2 −1 satellite images in May 2010 showed elevated [Chl a] in and 624 μmol photons m s , with 30 s incubation time for each E, giving  according to Nymark et al. (2009). The the areas south of the Marginal Ice Zone (MIZ, Fig. 1) from PSII 77°N to 80.5°N. The monthly average of satellite-derived maximum quantum yield, was obtained after dark acclima- tion of cells in 5 min prior to RLC. [Chl a] in the bloom area of 32000 km varied between 4 −3 and 13 mg Chl a m (Fig. 1). The ice edge position drawn The operational quantum yield of charge separations in PSII in actinic light conditions,  , was calculated accord- in Fig. 1 is indicative of its location on May 17th, the day of PSII the southward transect (St 4–7), showing that station 7 was ing to Genty et al. (1989) (Eq 1): in open waters (sea-ice images May 2010 can be retrieved � � � � = F − F ∕F (1) PSII m 0 m by www.met.no—daily sea-ice edge analysis). Satellite data using the F (“minimum fluorescence”) and F (“maximum (EUMETSAT Ocean and Satellite Application Facility on 0 m fluorescence” during saturating flash) at different actinic Ocean and Sea Ice, OSI SAF) on sea-ice edge position at irradiances. Note that cell samples were dominated by 16 and 17th May also indicated a north to south movement chromophytes and that no phycobiliprotein-containing cryp- of sea ice from 16 to 17th May in the transect area (see ice tophytes or cyanobacteria, with major Chl a emission from edge 16 and 17 May 2010, Fig. 1). PSI, thus affecting  , were observed in LM or by HPLC PSII pigment chemotaxonomy (Johnsen and Sakshaug 2007). In situ measurements of salinity, temperature, The irradiance (E) and  gives the relative electron irradiance and [Chl a] from the research vessel PSII transfer rate, rETR (Eq. 2): Salinity, temperature and in situ fluorescence of [Chl a ] rETR = E ⋅ (2) PSII from ice stations (St. 1–6, with St 6 close to MIZ) to open where the photosynthetic rates (rETR) at a given E was fitted water (St 7) indicate two distinct layers of water in the using a non-linear least-squares Marquardt algorithm and the study area separated vertically by a thermocline and halo- equation from Webb et al. (1974) to provide photosynthetic cline (Fig.  2). The halocline was found at around 40 m parameters based on a rETR versus E curve (RLC vs. E depth at St 4–7 and slightly deeper, around 60 m depth, curve, Eq. 3): at St 1–3. At low temperatures, water density will follow salinity and therefore a strong pycnocline is also present − ⋅ E rETR = rETR 1 − exp , (3) max between the two layers. Below the halocline temperature rETR max and salinity values indicate a strong influence of Atlantic where the photosynthetic parameters are maximum rETR water masses with salinity values  >  34.7 and tempera- (rETR ) and α (maximum light utilization coefficient, tures > 0°C. The waters of Arctic origin above the halo- max ETR also called “photosynthetic efficiency”) using Sigma Plot cline had temperature and salinity characteristic of polar version 11 (Systat Software, USA). surface water [T < 0°C and S 33.8–34.0, (Lind and Ing- The photosynthetic saturation index, E was calculated valdsen 2012)]. These salinity values (34.0 < S < 34.7) are as (Eq. 4): fresher than that defined for Arctic Water. Maximum con- centrations of in situ [Chl a] from CTD ranged between 5 E = rETR ∕ , (4) k max ETR −3 and 8 mg Chl a m , and are confined above the halocline −2. −1 where E is in μmol photons m s . going from open to ice-covered waters (Fig. 2). Typically, 1 3 1206 Polar Biology (2018) 41:1197–1216 the Chl a maxima were closest to the surface under the sea Phytoplankton biodiversity ice (around 10 m depth) and deeper within the MIZ and open waters (Chl a max at 20 m depth). See in vitro [Chl The phytoplankton species composition in the water masses a] estimation in the section below. underneath the sea ice and sea-ice bottom samples was dif- In situ downwelling irradiance profiles were measured in ferent (Table 4, Fig. 6). The phytoplankton cells from the open leads at ice stations 1–3 (Fig. 3, 4). E values at 10 m water were dominated by large and typical spring bloom PAR depth (depth of the AUV horizontal survey, see Fig. 5) at species of diatoms (northern cold or cosmopolitan species) −2 all three stations varied between 1 and 2 μmol photons m followed by prymnesio-, dino-, prasino- and chrysophytes, −1 s . Importantly, as visualized by the AUV photomosaics while the underside of the ice (from ice core samples) was (Fig. 3b–c), the leads represent areas of maximum irradi- characterized by typical ice algal species (Table 4). At St ance. Underwater downwelling irradiance at St 1–3 in the 1–3, Niskin water samples and vertical phytoplankton net upper [Chl a] maximum layer at 10 m depth was close to haul from the under-ice bloom (from 35 m depth to surface) two orders of magnitude lower than the E from in  vivo were dominated by a mixture of chain-forming centric (C) photosynthesis from the phytoplankton cells, see photosyn- and pennate (P) diatoms (Bacillariophyceae), 37 species thetic parameters below (Tables 1, 2, 3). For logistical rea- were identified. Four diatom species comprised the major sons, absolute values of E from directly underneath sea biomass under ice (St 1–3) and in open waters at St 7 (the PAR ice were not possible to obtain. Diurnal surface irradiance 18–20 May), based on relative cell abundance and chemot- south of MIZ 18–19 May at 0.5 m depth (from deck incuba- axonomic pigment tracers. Abundant diatoms were the cen- −2 −1 tors) ranged from 50 μmol photons m s at midnight to tric and spring-forming species Thalassiosira antarctica var. −2 −1 670 μmol photons m s at noon. borealis (Northern cold water to temperate distribution), T. nordenskioeldii (Northern cold water to temperate species) and Chaetoceros socialis species complex (aka C. socialis, Measurements from AUV under sea ice Cosmopolitan species). Also, the prymnesiophyte (cocco- lithophore) Phaeocystis pouchetii was abundant (many small The undulating AUV transects revealed that under-ice water cells identified), but comprised low biomass indicated by temperature was generally at −  1.66  °C from 5 to 25  m HPLC. The overall biomass was dominated by diatoms, con- depth, with corresponding salinity of 33.9 (Fig. 5) charac- firmed by high diatom-specific Chl c and fucoxanthin 1+2 teristic of polar surface waters, PSW (Lind and Ingvaldsen concentrations. The chemotaxonomic tracers for Phaeocystis 2012). The vertical distribution of [Chl a] (Fig. 5) indicated pouchetii indicated low biomass relative to the larger dia- two maximum layers, one at 0–15 meters depth (with high- toms (only traces of Chl c and acyloxy-fucoxanthin found est concentrations close to the underside of the ice) and one in under-ice samples, data not shown). at 30–40 m depth. Values in these layers ranged between 3 Ice core samples of bottom side (lower 3 cm, i.e. inter- −3 and 6 mg Chl a m (Fig. 5). The horizontal AUV survey (at stitial community) from St 1 was characterized by ice algae 10 m depth below the sea ice) indicated patchy, but generally dominated by the pennate sea-ice diatoms Navicula spp. and high in situ [Chl a] with values ranging from 3.7 to 5 mg Chl Nitzschia frigida. Present in low cell numbers were the pen- −3 a m (Fig. 3, 5). Both temperature and salinity showed little nate sea-ice diatoms Entomoneis sp., Hantzschia weiprech- variability at the 10 m layer (Fig. 5). The horizontal variabil- tii, Pseudogomphonema arcticum, Synedropsis hyperborea ity in salinity and temperature at 10 m (across 80,000 m ) and the centric diatom Attheya septentrionalis. The ice algal show that the lowest salinities (range 33.85–33.90) cor- community at St 2 was by numbers totally dominated by the responded with the lowest temperatures (range − 1.66 to dinoflagellate Polarella glacialis with many resting cysts, − 1.61 °C). The under-ice photomosaic images from AUV followed by diatoms Navicula sp. 1–2 and Nitzschia frigida (horizontal mapping grid at 10 m depth at 400 × 200 m area) (Table 4). indicated a patchy and dynamic underwater light field (irra - diance values in section above) that varied as a function HPLC pigment characterization of ice/snow thickness and leads seen as differences in light intensity and colouration (Fig. 3). The brightest areas (white The HPLC pigment chemotaxonomic signatures from St.1 hues in Fig. 3b–c) were found underneath the ice leads and and 3 revealed dominance of diatoms (Chl a, Chl c , 1+2 −2 corresponded to a downwelling E of 1–2 μmol photons m fucoxanthin and diadinoxanthin), verified by microscopy, −1 −2 −1 s at 10 m depth and around 20 μmol photons m s at at all depths. In addition, Chl b indicated the presence of the 5 m. Different hues in blue indicated sea-ice melting ponds observed Pyramimonas sp., which do not obtain the marker (observed at surface), dark green (thick ice) to brighter green carotenoid prasinoxanthin found in pigment group prasino- to brownish indicating different [Chl a ] in the water masses phyceae II (Johnsen and Sakshaug 2007) (verified by light and as biofilm of ice algae at the underside of sea ice. microscopy), was found at 10 m depth at St 3 (Table 1). Chl 1 3 Polar Biology (2018) 41:1197–1216 1207 Table 4 Phytoplankton and ice algae at ice stations (St 1–2)  and open waters (St 7)  Species Water Ice core St 1 Water Ice core St 2 Comments* column column St 1 St 7 Bacillariophyceae Amphora sp. × Often in ice, but also benthic, P Attheya septentrionalis × × + Epiphyte on phytoplankton or ice algae, C Bacterosira bathyomphala + ++ Northern cold water, spring, C Chaetoceros borealis × Cosmopolitan, C C. cf. debilis × Cosmopolitan; C C. convolutus × Cosmopolitan, usually more common during summer, C. decipiens × × Cosmopolitan, season independent, C C. furcellatus × Northern cold water, spring, C C. socialis species complex ++ + Northern cold water, spring, C C. karianus + + Northern cold water, C C. teres × Northern cold water to temperate, C Cylindrotheca closterium × × × Probably cosmopolitan, season independent, P Entomoneis sp. + × Usually spring, often also in sub-ice communities, P Eucampia groenlandica × + Typical northern cold water distribution, C Fragilariopsis cylindrus × × Bipolar? Spring (colonies may also occur in sub-ice communities. Solitary cells have been observed in interstitial communities), P F. oceanica + + Northern cold water, spring (colonies may also occur in sub-ice communities), P Hantzschia weiprechtii × Arctic, ice, P Navicula directa × Cosmopolitan, P N. pelagica × Northern cold water, spring. Usually ice-covered areas, P N. septentrionalis × × Northern cold water, spring, P N. vanhoeffenii × + Northern cold water, spring, P N. sp. 1 × +++ N. sp. 2 ++ N. spp. + + + Well-developed ice algal communities often consist of many different solitary Navicula species, P Nitzschia frigida × + × ++ Arctic, ice, P N. laevissima × × Arctic, ice, P Odontella aurita × Cosmopolitan. C Pauliella taeniata + × Northern cold water, spring, P Pleurosigma spp. × Both phytoplankton and ice, P Porosira glacialis + + Northern cold water to temperate, spring, P Proboscia alata × More common during summer, C Pseudogomphonema arcticum × x × Epiphyte on phytoplankton or ice algae, P Pseudo-nitzschia delicatissima ×x Cosmopolitan, more common during summer, P P. granii × × Northern cold water to temperate, often in Phaeocystis colonies, P P. pseudodelicatissima × Cosmopolitan. More common during summer, P P. seriata × × Northern cold water to temperate. More common dur- ing summer, P Stenoneis inconspicua var. borealis × Arctic, ice, P Synedropsis hyperborea × × + Epiphyte on phytoplankton or ice algae, P Thalassiosira antarctica var. borealis ++ ++ Northern cold water to temperate, spring, C T. bioculata × + Northern cold water, spring. Often close to ice or part of ice algal communities, C 1 3 1208 Polar Biology (2018) 41:1197–1216 Table 4 (continued) Species Water Ice core St 1 Water Ice core St 2 Comments* column column St 1 St 7 T. hyalina × +++ Northern cold water to temperate, spring, C T. nordenskioeldii ++ + Northern cold water to temperate, spring, C Coccolithophyceae Phaeocystis pouchetii ++ × Spring Dinophyceae × Gymnodinium spp. × × Katodinium sp. × Polarella glacialis × +++ Mix of vegetative cells and cysts Protoperidinium spp. × More common during summer, but several season independent species Prasinophyceae Pyramimonas sp. × × Chrysophyceae Dictyocha speculum + Cosmopolitan, season independent Dinobryon balticum + More common during summer (sometimes dominat- ing) Phytoplankton net samples (mesh size 20 μm, haul from 20 m depth to the surface). Ice algal samples from lower 3 cm of ice core. Symbols: × = present, + = regularly occurring, ++ = abundant, +++ = dominant. *Comments include biography, C (centric) and P (pennate) diatoms and own observations. Determination of phytoplankton biogeography follows the reviews (with references therein) in the region by von Quill- feldt (2000), von Quillfeldt et al. 2009 and Sakshaug et al. (2009b) −2 −1 c and 19′-acyloxufucoxanthins were found in trace amounts E of 244 μmol photons m s was found at 2 m depth and 3 k −2 −1 (pigment signature of Chl c -containing prymnesiophytes E of 127 μmol photons m s at 10 m depth followed with 3 k −2 −1 like P. pouchetii, (Johnsen et al. 2011a) at St 1–3 indicating 106 μmol photons m s at 40 m depth, comparable to St that the biomass (detected as total pigments) of Phaeocystis 3. Correspondingly, E at 10-m depth at St 1–3 was remark- pouchetii was very low (trace amounts) relative to diatoms. ably similar with 99 (St 1), 96 (St 2) and 89 (St 3) μmol pho- −2 −1 Light-harvesting pigments (LHP, chl c and fucoxantin) to tons m s , respectively. The maximum photosynthetic rate 1+2 Chl a ratio (w:w) was close to 1 (Table 2) while photoprotec- (rETRmax) was highest in cells from surface and inversely tive carotenoids (PPC, diadino- and diatoxanthin) to Chl a related to α. The mean NPQ at light-saturated photosynthesis −2 −1 ratio (w:w) was approximately 0.2. At 10 m depth at St. 3, (E of 500 μmol m -s ) from cells sampled at St. 1–3 was PAR cells contained both diadino- and diatoxanthin and the PPC 1.42 ± 25.9% (± CV of mean value, Table 3, see discussion). to Chl a ratio (w:w) was 0.22. Chlorophyll a to Chl a + Phaeo (Chl a and its degradation In situ and in vitro [Chl a] products phaeophorbide a, phaeophytin a and chlorophyllide a) ratio of 0.9 (w:w, Table 2) indicated healthy cells at all The mean [Chl a] in vitro (pigment extracts of water samples) depths at St 1 and 3 m, except at 10 m depth at St 3 with Chl from St 1–7 relative to in situ [Chl a] by AUV fluorometer and a to Chl a + Phaeo (W:W) ratio of 0.8, indicating peak of CTD fluorometer from vessel, gave an in vitro/in situ [Chl a ] bloom to start of post-bloom phase (Johnsen and Sakshaug ratio (w:w) of 2.92 ± 0.86 (n = 7) which indicates that in situ 1996). [Chl a] estimations from healthy and photosynthesizing cells (fluorometer from AUV or ship CTD fluorometer) is only Photosynthetic parameters 35% relative to in vitro [Chl a], i.e. autofluorescence (pigment extracts) not affected by PQ and NPQ, see discussion (Fig.  7). The photosynthetic parameters at ice stations indicate HL- acclimated cells characterized with E values ranging from −2 −1 78 to 244 μmol photons m s at all depths (Table 3). E Discussion −2 for 2 m depth was in the range of 105–144 μmol photons m −1 s . Lowest E values was found at 45 m depth at St 3 with In this study we report on the origin and dynamics of an −2 −1 −1 −2 78 μmol photons m s and 103 μmol photons s m at under-ice bloom observed in the Eurasian sector of the 10 m depth, correspondingly (Table 3). At St 2, the highest Arctic Ocean. Data were obtained simultaneously from 1 3 Polar Biology (2018) 41:1197–1216 1209 Fig. 6 Major phytoplankton species from Ice station 1: Chaetoceros gelidus (C. socia- lis, a), Chaetoceros karinus (b), Thalassiosira nordenskio- eldii (c), Phaeocystis pouchetii (d), Fragilariopsis oceanica (e), Navicula vanhoeffeni (f), Bacterosira bathyomphola (g) and Thalassiosira antarctica var. borealis (f). Six species of large diatoms made up the largest parts of the biomass. The high cell numbers of Phaeocys- tis pouchetii did not contribute much when looking at [Chl a] biomass using its tracer pig- ments Chl c and 19′-acyloxy- fucoxanthin (Sakshaug et al. 2009b; Johnsen et al. 2011a). Details in Table 4, see also Meshram et al. 2017 using metabarcoding of the microal- gae studied in this report three complementary platforms: Remote sensing of ocean The oceanographic conditions that were observed dur- colour (satellite), an AUV equipped with environmental ing this study are typical of conditions north and south of sensors and analyses from water samples (from a research the MIZ (Fig. 8). Whilst current vectors were not measured vessel). Recent studies in the MIZ north and west of Sval- directly, the regional flow characteristics (Walczowski et al. bard (Assmy et al. 2017; Kauko et al. 2017) have reported 2012) imply a NW transport of the subsurface waters and a on blooms developing in situ under a cover of sea ice and corresponding southward movement of sea ice in the region snow. We, however, argue that some under-ice blooms may studied in this paper. We are further guided in this assump- originate from blooms initiated in well-lit open water, and tion based on the observations and model results reported subsequently advected under sea ice. Being able to distin- by Assmy et  al. (2017) within the region studied here guish between these two scenarios is of vital importance to (80.67°–80.18°N and 4.84°–5.31°E) and at a similar time be able to provide realistic projections of production regimes of year (May 2010 vs. May–June 2015). Model data from as sea ice retreats commensurate with a reduction of the Assmy et al. (2017) show a strong NW advection of polar −1 Arctic ice cover. surface waters (PSW, mean current velocity ≈ 5 cm s at 20–30 m depth) and a NW transport of AW (mean current 1 3 1210 Polar Biology (2018) 41:1197–1216 Fig. 8 Temperature–salinity (TS) plot of water masses from station 1–7. CTD profile data for each station are shown in grey (MIZ or ice- covered stations) and black (open water station). The [Chl a] maxi- mum values are marked by green dots and are found to be consist- ently in the Polar Surface Water masses (PSW). There are no distinct differences in the TS characteristics of the stations indicating a rather consistent vertical structure of the water—mirrored in Fig. 2. In open water (Station 7), the [Chl a] maximum is deeper (25  m) compared to under the ice (~  6 to 10  m), indicative of vertical redistribution through wind-induced turbulence. The max [Chl a] is found at the bottom of the halocline at this station Fig. 7 a Differences in in  situ (living cells) versus in  vitro (pigment The presence of “Northern cold water” and “temperate” spe- extracts) [Chl a] at ice station (St) 1–3. These differences may mainly cies is typical for spring phytoplankton blooms in northern be due to quenching of Chl a fluorescence due to PQ and NPQ in liv - open waters (Hasle and Syvertsen 1996) and the dominance ing cells. Numbers in parentheses indicate ± SE, n = 3 for lab-based of centric over pennate spring species indicates that the in  vitro measurements. b In  situ [Chl a] from repetitive AUV verti- cal zig-zag  transects at ice station 3  (see also Fig.  5) compared to bloom has been developing for some time (von Quillfeldt corresponding ship based vertical [Chl a]  profile (mean of downcast 2000; Fragoso et al. 2017). The species composition in the and upcast in nearby ice-lead) water column was also different from the species composi- tion found on the underside of the ice, which are typical for −1 well-developed ice algal communities (Syvertsen 1991; von velocity of 5–10 cm s ) in the location of this study, the Quillfeldt et al. 2009). The species composition of the ice SW flank of the Yermak Plateau. Also, sea-ice data dur - algal community was dominated by diatoms at ice St 1 and ing May 2010 in the examined region showed small vari- dinoflagellates at St 2. This indicates that a “seeding stock” ations in extent and sea-ice concentration, but it started to of ice algae was not the origin of the “local bloom”, but, to disintegrate from 27 May 2010 (Norwegian Meteorological the contrary, cells in the water masses were brought under Institute, Svalbard High Resolution Ice Charts; see data base the ice by transportation of water masses from the south for May 2010 at www .polar view .me t.no). Hence, we can of MIZ (Fig. 2, 8). In sum, the continuous Chl a maxima conclude that the prevailing oceanographic and ice condi- support the assumption that the bloom observed along the tions are supporting our assumption that the observed bloom transect was one continuum of PP, rather than two distinct is of an advective origin (Fig. 8). blooms of phytoplankton in open water and under sea ice, In addition to the oceanographic conditions, we exam- respectively. In agreement with the species composition ined the taxonomical and eco-physiological phytoplankton data, the pigment data indicated that the dominating biomass status to evaluate whether the bloom was of an advective or was from diatoms (chl c and fucoxanthin) with only traces an in situ origin. First of all, we note that the phytoplank- 1+2 found of Chl c and acyloxy-fucoxanthin used as chemot- ton species composition was similar from the northernmost 3 axonomic markers for prymnesiophytes such as Phaeocys- ice station to the southernmost station in the open waters tis pouchetii (Rodríguez et al. 2006; Johnsen and Sakshaug south of MIZ (Table 4, see also Meshram et al. 2017 for 2007; Pettersen et al. 2011). We also found high E values metabarcoding data of cells close to station 1 in this study). k −2 −1 (mean E of 138 μmol m s under sea ice) indicating that 1 3 Polar Biology (2018) 41:1197–1216 1211 −2 −1 the under-ice bloom (ambient E < 2 μmol m s at 10 m affecting photosynthetic rates such as differences in incuba- depth) may have developed in open water and HL condi- tion time, absorbed quanta utilized by PSII, diurnal oscilla- tions. In addition, the taxonomic composition of the healthy tions of photosynthetic parameters and lastly, how to convert phytoplankton community (seen by  ) resembles typ- ETR rates (Chl a fluorescence from PSII) into C-fixation PSIImax ical northern planktonic species (different from ice algae rates (Johnsen and Sakshaug 2007; Hancke et al. 2008a, b; found in the sea ice). Finally, a continuum of high [Chl a] at Schuback et al. 2016). What we do know is that the rapid sea surface (remote sensing image of Chl a for May 2010) light curves (RLC) give higher E values than P versus E from open waters to 31 km north of MIZ, and relatively curves derived from PAM with 5 min incubation time or P high amounts of photoprotective carotenoids (PPC: Chl a versus E curves from “Photosynthetron technique” making ratio of 0.17–0.22 at all examined ice stations) also indicates comparison of photoacclimation (comparing E values) dif- HL-acclimated cells (Rodríguez et al. 2006; Schuback et al. ficult. PAM- and O -electrode-derived E values for diatoms 2 k 2016). We discuss each of these lines of evidence in more may be similar to short time C incubations (20–30 min) detail below. using the “Photosyntethron” technique (Lewis and Smith 1983; Johnsen and Hegseth 1991) compared to PAM and Photosynthetic parameters bio-optical-derived data (Hancke et al. 2008a). The findings of Assmy et al. (2017), in the middle part of Yermak plateau −2 −1 The mean E of 138 μmol photons m s (this study) at (May–June 2015) with a significantly lower current speed ice St 1–3 from 2 to 45 m depth indicate HL-acclimated than this study from S-Yermak plateau (May 2010), found −2 −1 cells and is comparable with HL-acclimated cells from the E values ranging from 137 to 584 μmol photons m s Arctic characterized (derived from C incubations) with of blooms dominated by Phaeocystis pouchetii, under leads −2 −1 E  > 60 μmol photons m s (Smith and Sakshaug 1990; in pack ice based on RLC curves (same method as used in Harrison and Cota 1991; Johnsen and Hegseth 1991; Stein this study). However, it should be noted that Phaeocystis and Macdonald 2003; Sakshaug et al. 2009b). These authors pouchetii generally obtain higher E that is 2–3 times higher provided a range for E in LL-acclimated phytoplankton in than diatoms under same ambient light conditions (data from −2 −1 the Arctic from 0.5 to 60 μmol photons m s with low- lab cultures and in situ data from the Barents Sea, (Sakshaug est values for cells acclimated to under-ice irradiances. We et al. 2009b)). obtained mean values of rETR , α and E as 97 ± 8.5 A study comparing photoacclimation status (E ) in max, k k −2 −1 (SE), 0.74 ± 0.09 (SE) and 138 ± 28 (SE) μmol m s , three classes of phytoplankton as a function of temperature respectively (Table  3). Under-ice blooms, dominated by (0–35 °C in diatoms, prymnesiophytes and dinoflagellates) Phaeocystis pouchetii in the mid part of the Yermak Pla- using ETR (PAM), O evolution and C incubation, gave teau with sea ice influenced by leads, leading more light E values in agreement with each other from 0 to 15 °C for into the under-ice water column, using same PAM method diatoms (Hancke et al. 2008a). In the latter study, the O as described here, gave E values of 137–584 μmol pho- method gave generally the highest E values when looking K k −2 −1 tons m s (Assmy et al. 2017). Cultured cells of Arc- at all three phytoplankton classes, followed by rETR and tic diatoms (Chaetoceros furcellatus and Thalassiosira C-method (pigment group dependent). Closest agreement nordenskioeldii) that were HL acclimated (grown at E of was found between E derived from O and rETR data, with k 2 −2 −1 14 400 μmol photons m s at 24 and 12 h day length) and C generally giving lowest E values. Since most E values k k −2 −1 14 LL acclimated (grown at E of 25 μmol photons m s at in literature are based on C method, a direct comparison 24 and 12 h day length) isolated from the same area in the may be hard to interpret. Still, our E data indicate that the Barents Sea (78°N, 30°E) obtained typically an E of 60 diatom cells are HL acclimated and is in the low range of (12 h day length) to 70 (24 h day length) and 31 (12 h) to under-ice E findings by Assmy et al. (2017) of Phaeocystis −2 −1 35 (24 h) μmol photons m s , respectively (Sakshaug pouchetii-dominated bloom, using the same RLC method as et  al. 1991, review in Sakshaug et  al. 2009b). Compar- in this study. For details of underwater irradiance in similar ing C data with ETR data is difficult and currently not type of sea ice at same time of the year, Kauko et al. (2017) resolved. Photosynthetic C- incubators were widely used observed similar light conditions as in this study. until the end of the 1990’s, typically using the “Photosyn- The NPQ of in vivo Chl a fluorescence is a physiological thetron” photosynthesis versus irradiance curve (P vs. E response related to the lowering of pH (during photosynthe- curve) technique, often with 20 min E incubation time sis) in the thylakoids inside the chloroplasts which induces PAR (Lewis and Smith 1983), compared to rETR data derived de-epoxidation of diadinoxanthin to diatoxanthin in diatoms from short incubation time experiments using PAM or FRRF and prymnesiophytes (coccolithophytes) through the “xan- (0.2–5 min incubation time per E to provide ETR vs. E thophyll cycle” of PPC (Brunet et al. 2011; Valle et al. 2014), PAR curves (Suggett et al. 2011a; Schuback et al. 2016)). There state II–I transitions between PSII and PSI (occurs mainly in are still several challenges in understanding the variables green algae and phycobiliprotein-containing phytoplankton) 1 3 1212 Polar Biology (2018) 41:1197–1216 and finally photoinhibitory quenching (Govindjee 1995; to diatoxanthin as a response to rapid (seconds to hour) Falkowski and Raven 1997). Regarding NPQ, we found no increase of irradiance (Brunet et al. 2011). pigment groups (green algae and cryptophytes) that may possess state II–I transitions inducing further lowering of In vitro versus in vivo/in situ Chl a fluorescence fluorescence signal emitted from cells (Brunet et al. 2011; Johnsen et al. 2011a; Schuback et al. 2016) and no photoin- Caution regarding in situ (or in vivo) versus in vitro [Chl hibitory irradiances under ice were observed (E at 10 m a]-based fluorometry for interpretation of data due to PQ PAR 2 −1 depth under sea ice ranged from 1 to 2 μmol m  s ) and and NPQ are discussed below. The observed bloom reached −3 these two factors cannot explain the high E and high PPC 20 mg Chl a m (in vitro), which can be regarded as a high indicating HL-acclimated cells (Table 3). This leaves the density bloom (Roy et al. 2011). In contrast, the correspond- reduction of Chl a fluorescence emitted from cells in situ ing in situ [Chl a] ranged under the sea ice St 1–4 from 4 to −3 (Fig. 7) by the two remaining NPQ mechanisms: first, the pH 8 mg Chl a m and with the highest [Chl a] centred at St 5 −3 gradient across the thylakoid membrane (causing acidic con- (8 mg Chl a m maximum at 10 m depth, Fig. 2, 7). In addi- ditions when photosynthetic rate is high and thus reduction tion, at St 3, the AUV survey indicated patchy distribution of fluorescence emitted from the cells). Secondly, the cor - of [Chl a] under the sea ice (Fig. 5), with a maximum [Chl responding lowering of pH inducing the xanthophyll cycle a] found at 10 and 38 m depth (vertical surveys) and at 10 m of diadino- and diatoxanthin (acting as radiators, emitting depth (horizontal survey) ranging between 3 and 6.5 mg Chl −3 absorbed light energy as heat) and corresponding quench- a m . Differences between in situ and in vitro estimations of ing of Chl a fluorescence, which requires HL conditions far [Chl a] may be, in addition to patchiness of phytoplankton above E level, which were not observed under the sea ice biomass, due to quenching of Chl a fluorescence in living (Table 2, Brunet et al. 2011). This is another indication that cells and differences in methodology (HPLC with separation the cells found under the sea ice may have been advected of Chl a and its degradation products, spectrophotometer and and/or covered by sea ice from well-lit, ice-free surface bench fluorometer). In living cells (in vivo) ϕ denotes the waters from the south of MIZ. A lack of saturation in NPQ fraction of light absorbed and utilized by PSII (Photosystem was observed at ice St 1–3 dominated by diatoms, despite II, the oxygen evolving site) that is converted to Chl a fluo- photophysiologial stress on cells during RLC exposed to rescence [emission detected at 685 nm in most fluorometers, high-light conditions up to 300 times higher than ambient (Babin 2008; Suggett et al. 2011b)]. The in vivo (including under-ice irradiance (Table 3). This suggest the presence of in situ) Chl a fluorescence is different from in vitro measure- HL-acclimated cells still with the capacity of photoprotec- ments because living cells change ϕ by two major processes tive downregulation as shown in Lavaud and Kroth (2006) defined as photochemical quenching (PQ) and NPQ. The and Schuback et al. (2016). PQ of Chl a fluorescence is a function of the fraction of closed reaction centres of PSII,  , see M & M (reviewed PSII Photoprotective carotenoids and light‑harvesting in Babin 2008; Johnsen et al. 2011a; Suggett et al. 2011a; pigments b). In addition, the in vitro versus in situ [Chl a] measure- ments can be used to evaluate the effect of photosynthetic The phytoplankton sampled under the sea ice, character- absorption and utilization of ambient irradiance by looking −2 −1 ized by low under-ice E (1–2 μmol photons m s at at PQ and NPQ. The highest ϕ for Chl a fluorescence will PAR F 10 m depth) was not acclimated to the ambient light inten- be found in vitro (extracted pigments) since the Chl a has sities. High cellular contents of diadinoxanthin, indicating lost its apoproteins and is non-functional in photosynthesis ML- (medium) to HL-acclimated cells, were found at St (the rate constant for photochemistry is 0) and a value of 1 and 3 (Table 2). Cells obtained diadinoxanthin to Chl a ϕ of 30% can be obtained (autofluorescence, Owens 1991; ratios (w:w) ranging from 0.17 to 1.19 typical for ML-HL- Johnsen et al. 2011a). In living and photosynthetically active acclimated cells (Johnsen and Sakshaug 1993; Rodríguez cells, ϕ in situ is typically 0.5–5% re-emission per quanta et  al. 2006; Schuback et  al. 2016). Note that conversion absorbed, about 70% of absorbed quanta will be lost as (epoxidation) from diatoxanthin to diadinoxanthin in dim/ thermal decay (heat), and the rest, 25–30%, will be used in dark–light conditions due to handling of water samples and photochemistry affecting PQ (Falkowski and Raven 1997; filtering of cells in dim light may happen in seconds and Johnsen et al. 2011a; Suggett et al. 2011a). For diatoms a therefore the sum of diadino- and diatoxanthin would often further reduction of Chl a fluorescence emitted from liv - be the best measure of photoacclimation status (Rodríguez ing cells will be due to NPQ processes, mainly due to low et al. 2006; Brunet et al. 2011). Significant amounts of dia- pH (H + produced by the water-splitting complex in PSII, dinoxanthin, characteristic for HL-acclimated phytoplankton creating acidic condition and reducing the fluorescence emit- (Johnsen et al. 1994; Rodríguez et al. 2006; Johnsen and ted from cells) and light absorption and corresponding heat Sakshaug 2007; Brunet et al. 2011), can be de-epoxided radiation from PPC thus not sending the absorbed energy 1 3 Polar Biology (2018) 41:1197–1216 1213 further to PSII (reviews in Govindjee 1995; Falkowski and and south of the MIZ (open waters). We conclude that the Raven 1997; Babin 2008; Brunet et al. 2011; Johnsen et al. bloom observed beneath the sea ice North of Svalbard in 2011a and Suggett et al. 2011a). May 2010 was an advective phenomenon, with cells that By comparing in vitro to in situ [Chl a] ratios (w:w, with had developed in open water further South. This advective in situ and in vivo Chl a fluorescence highly sensitive to PQ mechanism is in contrast to the in situ under-ice bloom sce- and NPQ) from research vessel and/or AUV fluorometer we narios elucidated by Arrigo et al. (2012) and Assmy et al. obtained ratios of 1.3–4.4 with a mean ratio of 2.92 ± 0.86 (2017) where under-ice, advective flow was considerably (n = 7), indicating a significant loss in ϕ due to PQ and weaker compared to our study. Knowledge about the ori- NPQ (Fig. 7). The mean  of 0.61 ± 1.3% (all water sam- gin and dynamics of pelagic blooms in the MIZ and PIZ is PSII ples from st 1–3) at lowest RLC incubation (Phyto-PAM) essential in order to provide realistic projections of produc- −2 −1 irradiance of 1 μmol photons m s (similar to the light tion regimes in a future warmer, and hence less ice-covered intensities observed under sea ice) indicated healthy cells Arctic. This study also imply that comparisons/calibrations with a high fraction of open reaction centres to perform pho- between different P vs E methodologies and incubation tosynthesis, i.e. cells far from light saturation in the under- times are highly needed for a common currency to compare ice ambient light (Table 3, Hancke et al. 2008a, b). between laboratory and in situ measurements as outlined by In vitro analyses provide the most accurate measurements Schuback et al. (2016). of [Chl a] and is often based on spectrophotometric meas- Acknowledgements We thank Carsten Brockmann at Brockmann Con- urements (high precision and accuracy, but low sensitivity). sults for the processing of the MERIS data. Contributions from the In contrast, Chl a fluorometers have about 100–200 times Centre of Excellence at NTNU “AMOS” (NRC, Norwegian Research higher sensitivity than spectrophotometers, but are depend- Council, Project 223254), the NRC Project “Arctic ABC” (NRC ent on calibration from spectrophotometers using Chl a 244319) and the UNIS MSc and PhD course (AB 323-823) “Light regime and primary production in the Arctic” and the crew on RV standards (Johnsen and Sakshaug 2000; Roy et al. 2011; this Helmer Hanssen (University of Tromsø) are greatly acknowledged. study). Note that fluorometer readings of Chl a are highly Contribution by FC is through the NERC-funded Changing Arctic influenced by the different fluorescent fractions of Chl a Ocean Program project “Arctic PRIZE”. such as its degradation products chlorophyllide a and phaeo- Open Access This article is distributed under the terms of the Crea- pigments (Johnsen and Sakshaug 2000; Roy et al. 2011). The tive Commons Attribution 4.0 International License (http://creat iveco Chl a fluorescence is also in vitro highly pH dependent and mmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- adding of acid may be used to look at phaeopigments using tion, and reproduction in any medium, provided you give appropriate broadband fluorometers, spectrofluorometers or spectropho- credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. tometers (Holm-Hansen et al. 1965; Johnsen and Sakshaug 2000; Roy et al. 2011). Many put in 1–2 drops of HCl in different volumes of pigment extracts to estimate the fraction of “phaeopigments” and these readings are highly depended References on pH giving different degradation products of Chl a with different emission strength making quantification difficult Arrigo KR, Perovich DK, Pickart RS, Brown ZW, van Dijken GL, (Johnsen and Sakshaug 1993). Lowry KE, Mills MM, Palmer MA, Balch WM, Bahr F, Bates NR, Benitez-Nelson C, Bowler B, Brownlee E, Ehn JK, Frey KE, Garley R, Laney SR, Lubelczyk L, Mathis J, Matsuoka A, Mitchell BG, Moore GWK, Ortega-Retuerta E, Pal S, Polashenski Conclusions CM, Reynolds RA, Schieber B, Sosik HM, Stephens M, Swift JH (2012) Massive phytoplankton blooms under Arctic sea ice. A coordinated campaign with the main aim of studying the Science 336:1408–1408. https://doi.or g/10.1126/science.12150 65 Arrigo KR, Perovich DK, Pickart RS, Brown ZW, van Dijken GL, dynamics and origin of an under-ice pelagic bloom north of Lowry KE, Mills MM, Palmer MA, Balch WM, Bates NR, Svalbard was carried out in May 2010. A continuum of high Benitez-Nelson CR, Brownlee E, Frey KE, Laney SR, Mathis J, [Chl a] above the halocline was found along the transect Matsuoka A, Greg Mitchell B, Moore GWK, Reynolds RA, Sosik from open water and 31 km into the pack ice. The bloom HM, Swift JH (2014) Phytoplankton blooms beneath the sea ice in the Chukchi sea. Deep-Sea Res Part II 105:1–16. https ://doi. was characterized with high-light-acclimated phytoplankton org/10.1016/j.dsr2.2014.03.018 cells (based on E , PPC and NPQ) along the entire transect, Assmy P, Fernández-Méndez M, Duarte P, Meyer A, Randelhoff A, including the ice stations deepest into the sea ice. Typically, Mundy CJ, Olsen LM, Kauko HM, Bailey A, Chierici M, Cohen the Chl a maximum was closer to the surface under the ice L, Doulgeris AP, Ehn JK, Fransson A, Gerland S, Hop H, Hudson SR, Hughes N, Itkin P, Johnsen G, King JA, Koch BP, Koenig Z, (around 10 m depth) and deeper within the MIZ and open Kwasniewski S, Laney SR, Nicolaus M, Pavlov AK, Polashenski waters (20 m depth). Also, the, species composition was CM, Provost C, Rösel A, Sandbu M, Spreen G, Smedsrud LH, similar at northernmost ice stations (31 km into the sea ice) Sundfjord A, Taskjelle T, Tatarek A, Wiktor J, Wagner PM, Wold 1 3 1214 Polar Biology (2018) 41:1197–1216 A, Steen H, Granskog MA (2017) Leads in Arctic pack ice enable Hasle GR, Syvertsen EE (1996) Marine diatoms. In: Tomas CR (ed) early phytoplankton blooms below snow-covered sea ice. Sci Rep Identifying marine diatoms and dinoflagellates. Academic Press, 7:40850. https ://doi.org/10.1038/srep4 0850 San Diego Babin M (2008) Phytoplankton fluorescence: theory, current literature Holm-Hansen O, Lorenzen CJ, Holmes RW, Strickland JDH (1965) and in situ measurement. In: Babin M, Roesler CS, Cullen JJ (eds) Fluorometric determination of chlorophyll. J Cons Perm Int Real-time coastal observing systems for ecosystem dynamics and Explor Mer 30:3–15 harmful algal blooms: theory, instrumentation and modelling. Hovland EK, Dierssen HM, Ferreira AS, Johnsen G (2013) Dynamics UNESCO Publishing, Paris, pp 237–280 regulating major trends in Barents Sea temperatures and subse- Behrenfeld M, Prasil O, Babin M, Bruyant F (2004) In search of a quent ee ff ct on remotely sensed particulate inorganic carbon. Mar physiological basis for covariations in light-limited and light- Ecol Prog Ser 484:17–32. https ://doi.org/10.3354/meps1 0277 saturated photosynthesis. J Phycol 40:4–25 Hovland EK, Hancke K, Alver MO, Drinkwater K, Hokedal J, Johnsen Berge J, Renaud P, Darnis G, Cottier F, Last K, Gabrielsen T, John- G, Moline M, Sakshaug E (2014) Optical impact of an Emiliania sen G, Seuthe L, Weslawski JM, Leu E, Moline M, Nahrgang huxleyi bloom in the frontal region of the Barents Sea. J Mar Syst J, Søreide JE, Varpe Ø, Lønne OJ, Daase M, Falk-Petersen S 130:228–240. https ://doi.org/10.1016/j.jmars ys.2012.07.002 (2015) In the dark: a review of ecosystem processes during the Ingvaldsen R, Loeng H (2009) Physical oceanography. In: Sakshaug Arctic polar night. Prog Oceanogr 139:258–271. h t t p s : / / d o i . E, Johnsen G, Kovacs KM (eds) Ecosystem Barents Sea. Tapir org/10.1016/j.pocea n.2015.08.005 Academic Press, Trondheim, pp 33–64 Boetius A, Albrecht S, Bakker K, Bienhold C, Felden J, Fernandez- Johnsen G, Hegseth EN (1991) Photoadaptation of sea-ice microalgae Mendez M, Hendricks S, Katlein C, Lalande C, Krumpen T, Nico- in the Barents sea. Polar Biol 11:179–184 laus M, Peeken I, Rabe B, Rogacheva A, Rybakova E, Somavilla Johnsen G, Sakshaug E (1993) Biooptical characteristics and photoa- R, Wenzhofer F, Sc RVPA–S (2013) Export of algal biomass from daptive responses in the toxic and bloom-forming dinoflagellates the melting Arctic sea ice. Science 339:1430–1432. https ://doi. Gyrodinium aureolum, Gymnodinium galatheanum, and two org/10.1126/scien ce.12313 46 strains of Prorocentrum minimum. J Phycol 29:627–642 Brunet C, Johnsen G, Lavaud J, Roy S (2011) Pigments and photoaccli- Johnsen G, Sakshaug E (1996) Light harvesting in bloom-forming mation processes. In: Roy S, Llewellyn CA, Egeland ES, Johnsen marine phytoplankton: species-specificity and photoacclimation. G (eds) Phytoplankton pigments. Cambridge Press University, Sci Mar 60:47–56 Cambridge, pp 445–471 Johnsen G, Sakshaug E (2000) Monitoring of harmful algal blooms Chauton M, Tilstone G, Legrand C, Johnsen G (2004) Changes in along the Norwegian Coast using bio-optical methods. S Afr J pigmentation, bio-optical characteristics and photo-physiology, Mar Sci 22:309–321 during a phytoplankton succession in mesocosms. J Plankt Res Johnsen G, Sakshaug E (2007) Bio-optical characteristics of PSII and 26:315–324 PSI in 33 species (13 pigment groups) of marine phytoplank- Drinkwater KF (2011) The influence of climate variability and change ton, and the relevance for PAM and FRR fluorometry. J Phycol on the ecosystems of the Barents Sea and adjacent waters: review 43:1236–1251 and synthesis of recent studies from the NESSAS Project. Prog Johnsen G, Samset O, Granskog L, Sakshaug E (1994) In-vivo absorp- Oceanogr 90:47–61 tion characteristics in 10 classes of bloom-forming phytoplank- Falkowski PG, Raven JA (1997) Aquatic photosynthesis. Blackwell ton—taxonomic characteristics and responses to photoadaptation Science, Oxford by means of discriminant and HPLC analysis. Mar Ecol Prog Ser Fragoso GM, Poulton AJ, Yashayaev IM, Head EJH, Purdie DA 105:149–157 (2017) Spring phytoplankton communities of the Labrador Sea Johnsen G, Volent Z, Sakshaug E, Sigernes F, Pettersson LH (2009) (2005–2014): pigment signatures, photophysiology and elemen- Remote sensing in the Barents Sea. In: Sakshaug E, Johnsen G, tal ratios. Biogeosciences 14:1235–1259. https://d oi.org/10.5194/ Kovacs KM (eds) Ecosystem Barents Sea. Tapir Academic Press, bg-14-1235-2017 Trondheim, pp 139–166 Genty B, Briantais JM, Baker NR (1989) The relationship between the Johnsen G, Bricaud A, Nelson N, Prézelin B, Bidigare R (2011a) In quantum yield of photosynthetic electron transport and quenching vivo bio-optical properties of phytoplankton pigments. In: Roy of chlorophyll fluorescence. Biochem Biophys Acta 990:87–92 S, Llewellyn CA, Egeland ES, Johnsen G (eds) Phytoplankton Govindjee (1995) Sixty-three years since Kautsky: chlorophyll a fluo- pigments: characterization, chemotaxonomy and applications rescence. Aust J Plant Physiol 22:131–160 in oceanography. Cambridge University Press, Cambridge, pp Gradinger R (1996) Occurrence of an algal bloom under Arctic pack 496–537 ice. Mar Ecol Prog Ser 131:301–305 Johnsen G, Moline M, Pettersson LH, Pinckney JL, Pozdnyakov D, Hancke K, Hancke TB, Olsen LM, Johnsen G, Glud RN (2008a) Tem- Egeland ES, Schofield OM (2011b) Optical monitoring of phyto- perature effects on microalgal photosynthesis-light responses plankton bloom pigment signatures. In: Roy S et al (eds), Phyto- measured by O production, pulse-amplitude-modulated fluores- plankton pigments: characterization, chemotaxonomy and applica- cence, and C assimilation. J Phycol 44:501–514. https://doi.or g tions in oceanography. Cambridge University Press, Cambridge, /10.1111/j.1529-8817.2008.00487 .x pp 538–581 Hancke TB, Hancke K, Johnsen G, Sakshaug E (2008b) Rate of O Kauko HM, Taskjelle T, Pavlov AK, Mundy CJ, Assmy P, Duarte production derived from pulse-amplitude-modulated fluores- P, Fernández Méndez M, Olsen LM, Hudson SR, Johnsen G, cence: testing three biooptical approaches against measured O Granskog MA (2017) Windows in Arctic sea ice: light transmis- production rate. J Phycol 44:803–813. h t t p s : / /d o i . or g / 1 0. 1 1 1 sion and the role of ice algae in a refrozen lead. J Geophys Res 1/j.1529-8817.2008.00509 .x 122:1486–1505 Hancke K, Hovland EK, Volent Z, Pettersen R, Johnsen G, Moline Knap A, Michaels A, Close A, Ducklow H, Dickson A (1996) Meas- M, Sakshaug E (2014) Optical properties of CDOM across the urement of chlorophyll a and phaeopigments by fluorometric Polar Front in the Barents Sea: origin, distribution and signifi- analysis. JGOFS Rep 19:118–122 cance. J Mar Syst 130:219–227. https ://doi.org/10.1016/j.jmars Kolber Z, Falkowski PG (1993) Use of active fluorescence to esti- ys.2012.06.006 mate phytoplankton photosynthesis in  situ. Limnol Oceanogr Harrison WG, Cota GF (1991) Primary production in polar waters— 38:1646–1665 relation to nutrient availability. Polar Res 10:87–104 1 3 Polar Biology (2018) 41:1197–1216 1215 Kromkamp JC, Forster RM (2003) The use of variable fluorescence Rudels B, Bjork G, Nilsson J, Winsor P, Lake I, Nohr C (2005) The measurements in aquatic ecosystems: die ff rence between multiple interaction between waters from the Arctic Ocean and the Nordic and single turnover measuring protocols and suggested terminol- Seas north of Fram Strait and along the East Greenland Current: ogy. Eur J Phycol 38:103–112 results from the Arctic Ocean-02 Oden expedition. J Mar Syst Lakowicz JR (1983) Principles of fluorescence spectroscopy. Plenum 55:1–30. https ://doi.org/10.1016/j.jmars ys.2004.06.008 Press, New York Sakshaug E, Johnsen G, Andresen K, Vernet M (1991) Modeling of Lavaud J, Kroth PG (2006) In diatoms, the transthylakoid proton gradi- light-dependent algal photosynthesis and growth—experiments ent regulates the photoprotective non-photochemical u fl orescence with the Barents Sea diatoms Thalassiosira nordenskioeldii and quenching beyond its control on the xanthophyll cycle. Plant Cell Chaetoceros furcellatus. Deep-Sea Res Part A 38:415–430 Physiol 47:1010–1016 Sakshaug E, Bricaud A, Dandonneau Y, Falkowski PG, Kiefer DA, Leu E, Søreide JE, Hessen DO, Falk-Petersen S, Berge J (2011) Con- Legendre L, Morel A, Parslow J, Takahashi M (1997) Param- sequences of changing sea-ice cover for primary and secondary eters of photosynthesis: definitions, theory and interpretation of producers in the European Arctic shelf seas: timing, quantity, and results. J Plankt Res 19:1637–1670. https://doi.or g/10.1093/plank quality. Prog Oceanogr 90:18–32 t/19.11.1637 Lewis MR, Smith JC (1983) A small volume, short-incubation-time Sakshaug E, Johnsen G, Kovacs KM (eds) (2009a) Ecosystem Barents method for measurement of photosynthesis as a function of inci- Sea. Tapir Academic Press, Trondheim, p 589 dent irradiance. Mar Ecol Prog Ser 13:99–102 Sakshaug E, Johnsen G, Kristiansen S, Quillfeldt CH, Rey F, Slagstad Lind S, Ingvaldsen RB (2012) Variability and impacts of Atlantic D, Thingstad F (2009b) Phytoplankton and primary production. Water entering the Barents Sea from the north. Deep-Sea Res In: Sakshaug E, Johnsen G, Kovacs KM (eds) Ecosystem Barents Part I 62:70–88. https ://doi.org/10.1016/j.dsr.2011.12.007 Sea. Tapir Academic Press, Trondheim, pp 167–208 MacIntyre HL, Kana TM, Geider RJ (2000) The effect of water motion Schuback N, Flecken M, Maldonado MT, Tortell PD (2016) Diurnal on short-term rates of photosynthesis by marine phytoplank- variation in the coupling of photosynthetic electron transport and ton. Trends Plant Sci 5:12–17. h t t p s : / / d o i . o r g / 1 0 . 1 0 1 6 / s 1 3 6 0 carbon fixation in iron-limited phytoplankton in the NE subarctic -1385(99)01504 -6 Pacific. Biogeosciences 13:1019–1035. https ://doi.org/10.5194/ Mackinney G (1941) Absorption of light by chlorophyll solutions. J bg-13-1019-2016 Biol Chem 140:315–322 Slagstad D, Wassmann PFJ, Ellingsen I (2015) Physical constrains and Maxwell K, Johnson GN (2000) Chlorophyll fluorescence—a practical productivity in the future Arctic Ocean. Front Mar Sci. https://doi. guide. J Exp Bot 51:659–668org/10.3389/fmars .2015.00085 McMinn A, Hegseth EN (2007) Sea ice primary productivity in the Smith WO, Sakshaug E (1990) Polar Phytoplankton. In: Smith WO Jr northern Barents Sea, spring 2004. Polar Biol 30:289–294. https (ed) Polar oceanography part B chemistry, biology and geology. ://doi.org/10.1007/s0030 0-006-0182-x Academic Press Inc., San Diego, pp 477–525 Meshram AR, Vader A, Kristiansen S, Gabrielsen TM (2017) Micro- Søreide JE, Leu E, Berge J, Graeve M, Falk-Petersen S (2010) Timining bial Eukaryotes in an Arctic Under-Ice Spring Bloom North of in blooms, algal food quality and Calanus glacialis reproduction Svalbard. Front Microbiol 8:1099. https ://doi.org/10.3389/fmicb and growth in a changing Arctic. Glob Change Biol 16:3154– .2017.01099 3163. https ://doi.org/10.1111/j.1365-2486.2010.02175 Moline MA, Blackwell SM, Von Alt C, Allen B, Austin T, Case J, Stein R, Macdonald RW (2003) The organic carbon cycle in the Arctic Forrester N, Goldsborough R, Purcell M, Stokey R (2005) Ocean. Springer, Berlin Remote environmental monitoring units: an autonomous vehi- Suggett DJ, Moore CM, Hickman AE, Geider RJ (2009) Interpreta- cle for characterizing coastal environments. J Atmos Ocean Tech tion of fast repetition rate (FRR) fluorescence: signatures of phy - 22:1797–1808 toplankton community structure versus physiological state. Mar Mundy CJ, Gosselin M, Ehn J, Gratton Y, Rossnagel A, Barber D, Ecol Prog Ser 376:1–19. https ://doi.org/10.3354/meps0 7830 Tremblay J-E, Palmer M, Arrigo KR, Darnis G, Fortier L, Else Suggett DJ, Moore CM, Geider RJ (2011a) Estimating aquatic produc- B, Papakyriakou T (2009) Contribution of under-ice primary tivity from active fluorescence measurements. In: Suggett DJ, Pra- production to an ice-edge upwelling phytoplankton bloom sil O, Borowitzka MA (eds) Chlorophyll a fluorescence in aquatic in the Canadian Beaufort Sea. Geophys Res Lett. https ://doi. sciences. Springer, Dordrecht, pp 103–128 org/10.1029/2009G L0388 37 Suggett DJ, Prasil O, Borowitzka MA (2011b) Chlorophyll a fluores- Nymark M, Valle KC, Brembu T, Hancke K, Winge P, Andresen K, cence in aquatic sciences. Springer, Dordrecht Johnsen G, Bones AM (2009) An integrated analysis of molecular Syvertsen EE (1991) Ice algae in the Barents Sea—types of assem- acclimation to high light in the marine diatom Phaeodactylum blages, origin, fate and role in the ice-edge phytoplankton bloom. tricornutum. PLoS ONE 11:e7743. https://doi.or g/10.1371/journ Polar Res 10:277–287 al.pone.00077 43 Valle KC, Nymark M, Aamot I, Hancke K, Winge P, Andresen K, John- Pettersen R, Johnsen G, Berge J, Hovland EK (2011) Phytoplankton sen G, Brembu T, Bones AM (2014) System responses to equal chemotaxonomy in waters around the Svalbard archipelago reveals doses of photosynthetically usable radiation of blue, green, and high amounts of Chl b and presence of gyroxanthin-diester. Polar red light in the marine diatom Phaeodactylum tricornutum. PLoS Biol 34:627–635. https ://doi.org/10.1007/s0030 0-010-0917-6 ONE 9:e114211. https ://doi.org/10.1371/journ al.pone.01142 11 Rodríguez F, Chauton M, Johnsen G, Andresen K, Olsen LM, Zapata Vancoppenolle M, Meiners KM, Michel C, Bopp L, Brabant F, Carnat M (2006) Photoacclimation in phytoplankton: implications for G, Delille B, Lannuzel D, Madec G, Moreau S, Tison JL, van Der biomass estimates, pigment functionality and chemotaxonomy. Merwe P (2013) Role of sea ice in global biogeochemical cycles: Mar Biol 148:963–971 Emerging views and challenges. Quat Sci Rev 79:207–230 Roy S, Llewellyn CA, Egeland ES, Johnsen G (2011) Phytoplankton von Quillfeldt CH (2000) Common diatom species in Arctic spring pigments: updates on characterization, chemotaxonomy and appli- blooms: their distribution and abundance. Bot Mar 43:499–516 cations in oceanography. Cambridge University Press, Cambridge 1 3 1216 Polar Biology (2018) 41:1197–1216 von Quillfeldt CH, Hegseth EN, Johnsen G, Sakshaug E, Syvertsen Arctic marine climate. ICES J Mar Sci 69:864–869. https ://doi. EE (2009) Ice algae. In: Sakshaug E, Johnsen G, Kovacs KM org/10.1093/icesj ms/fss06 8 (eds) Ecosystem Barents Sea. Tapir Academic Press, Trondheim, Webb WL, Newton M, Starr D (1974) Carbon dioxide exchange of pp 285–302 Alnus rubra: a mathematical model. Oecologia 17:281–291 Walczowski W, Piechura J, Goszczko I, Wieczorek P (2012) Changes in Atlantic water properties: an important factor in the European 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Polar Biology Springer Journals

The advective origin of an under-ice spring bloom in the Arctic Ocean using multiple observational platforms

Free
20 pages
Loading next page...
 
/lp/springer_journal/the-advective-origin-of-an-under-ice-spring-bloom-in-the-arctic-ocean-gN0kBomjBZ
Publisher
Springer Berlin Heidelberg
Copyright
Copyright © 2018 by The Author(s)
Subject
Life Sciences; Ecology; Oceanography; Microbiology; Plant Sciences; Zoology
ISSN
0722-4060
eISSN
1432-2056
D.O.I.
10.1007/s00300-018-2278-5
Publisher site
See Article on Publisher Site

Abstract

Under-ice blooms of phytoplankton in the Chukchi Sea have been observed, with strong implications for our understand- ing of the production regimes in the Arctic Ocean. Using a combination of satellite remote sensing of phytoplankton bio- mass, in situ observations under sea ice from an autonomous underwater vehicle (AUV), and in vivo photophysiology, we examined the composition, magnitude and origin of a bloom detected beneath the sea ice Northwest of Svalbard (Southern −3 Yermak Plateau) in May 2010. In situ concentration of up to 20 mg chlorophyll a [Chl a] m , were dominated by the north- ern planktonic spring species of diatoms, Thalassiosira nordenskioeldii, T. antarctica var. borealis, Chaetoceros socialis species complex and Fragilariopsis oceanica. These species were also found south of the marginal ice zone (MIZ). Cells in the water column under the sea ice were typically high-light acclimated, with a mean light saturation index (E ) of 138 μmol −2 −1 photons m  s and a ratio between photoprotective carotenoids (PPC) and Chl a (w:w) of 0.2. Remotely sensed data of [Chl a] showed a 32,000 km bloom developing south of the MIZ. In effect, our data suggest that the observed under-ice bloom was in fact a bloom developed in open waters south of the ice edge, and that a combination of northward-flowing water masses and southward drifting sea ice effectively positioned the bloom under the sea ice. This have implications for our general understanding of under-ice blooms, suggesting that their origin and connection with open water may be differ - ent in different regions of the Arctic. Keywords Sea-ice algae · Under-ice bloom · Phytoplankton · Photoacclimation · Advection of cells · Arctic · Satellite · Autonomous underwater vehicle · Chlorophyll a · Pulse amplitude modulated fluorescence (PAM) * Geir Johnsen School of Marine Science and Policy, University geir.johnsen@ntnu.no of Delaware, Newark, DE 19958, USA California Polytechnic State University, San Luis Obispo, Centre for Autonomous Marine Operations and Systems, CA 93407, USA Department of Biology, Norwegian University of Technology and Science (NTNU), 7491 Trondheim, Department of Environment and Mapping, Norwegian Polar Norway Institute (NP), 9296 Tromsø, Norway 2 8 Department of Arctic Biology, University Centre on Svalbard Scottish Association for Marine Science (SAMS), Oban, (UNIS), 9171 Longyearbyen, Norway Argyll PA37 1QA, UK Department of Arctic and Marine Biology, UiT The Arctic University of Norway, 9037 Tromsø, Norway Section of Marine Biogeochemistry and Oceanography, Norwegian Institute of Water Research (NIVA), 0349 Oslo, Norway Vol.:(0123456789) 1 3 1198 Polar Biology (2018) 41:1197–1216 physiological status of phytoplankton (Leu et al. 2011; Introduction Pettersen et al. 2011; Hovland et al. 2013; Hancke et al. 2014; Hovland et al. 2014). The key properties of the water Sea ice plays a dual role in the control of high-latitude masses and ice conditions in this region are well docu- phytoplankton blooms by influencing both light shading mented in terms of current speed and direction, tempera- and stratification (Arrigo et al. 2012; Assmy et al. 2017; ture, salinity and density (Rudels et al. 2005; Ingvaldsen Kauko et  al. 2017). Importantly, primary production and Loeng 2009; Lind and Ingvaldsen 2012; Assmy et al. (PP) in regions with an annual ice cover comes from two 2017). main sources, i.e. ice-associated and open water blooms Pelagic phytoplankton blooms in the MIZ are typically (Søreide et al. 2010). These two main sources of PP have initiated from mid-April to June in the waters west of Spits- traditionally been considered as temporally and spatially bergen and in the Barents Sea, and are associated with both distinct. It has been suggested that withdrawal of ice will Arctic cold water species and/or temperate species (McMinn generally lead to an increase in PP (Drinkwater 2011; and Hegseth 2007; Sakshaug et al. 2009a, b). In contrast, Slagstad et al. 2015), and that the nutrient-dependent new blooms in the perennial ice zone (PIZ) are generally thought production may not increase proportionally with increas- of as mainly being associated with ice algae, with irradiance ing light intensity due to insufficient access to essential in the water column below ice too low to support a pelagic nutrients. Recently, new reports of a “third” source of PP bloom (Sakshaug et al. 2009b; Berge et al. 2015). Recently, were presented, under-ice phytoplankton blooms (see e.g. Arrigo et al. (2012) provided new and important insights Arrigo et al. 2012). A phytoplankton bloom beneath the into the occurrence and development of pelagic blooms sea ice is a phenomenon that is poorly understood, they are under Arctic sea ice. A recent report from the same area as hard to detect and effectively invisible from satellites, and this study (NW of Svalbard), shows the dynamics between they occur in habitats that are logistically hard to sample. sea ice, the local current systems and a phytoplankton bloom Yet, there are several reports documenting their existence dominated by Phaeocystis pouchetii (Assmy et al. 2017). (Gradinger 1996; Mundy et al. 2009; Boetius et al. 2013; The light regime (light climate) is a function of three Assmy et al. 2017), but few that estimate their role in the variables, the irradiance (E) in the PAR region, E in total productivity and carbon budget of the Arctic. In the PAR −2 −1 μmol quanta m s (photosynthetic active radiation, Chukchi Sea, Arrigo et  al. (2012, 2014) estimated that 400–700 nm), the spectral irradiance, E (λ) in μmol quanta 90% of the total PP occurred under sea ice. Importantly, as −2 −1 −1 m s nm , and day length in hours (Falkowski and most current models rely on remote sensing for input and Raven 1997; Sakshaug et al. 1997, 2009b). Photoacclima- validation of phytoplankton biomass, occurring beneath an tion (short-term physiological adjustment of growth rate to ice cover, phytoplankton biomass in an under-ice bloom light climate) is divided in short- (seconds–minutes) and is at best poorly represented in contemporary modelling long-term (hours–days) responses (MacIntyre et al. 2000; efforts (Vancoppenolle et al. 2013). Under-ice blooms are Brunet et al. 2011; Valle et al. 2014). Short-term photoac- therefore important to map and understand, as they may climation includes the xanthophyll cycle for photoprotec- represent a significant component that is missing in the tion, an important part of non-photochemical quenching total production regime of the Arctic Ocean. Also, the (NPQ). Long-term photoacclimation is defined as change sea-ice zone has been identified as the region with largest in ratio between light-harvesting pigments (LHP) and pho- model uncertainty regarding Arctic PP and biophysical toprotective carotenoids (PPC), photosynthetic parameters, interactions (Assmy et al. 2017; Kauko et al. 2017). Know- enzymatic activities involved in photosynthesis and respira- ing that under-ice blooms have been reported from across tion, and changes in cell volume and chemical composition the central Arctic Ocean and surrounding shelf seas, yet (Brunet et al. 2011). not taken into account in contemporary modelling efforts, Qualitative and quantitative pigment analyses of chloro- there is a clear and important gap in knowledge that needs phylls and carotenoids by high performance liquid chroma- to be filled: A quantitative and mechanistic understanding tography (HPLC) can be used for chemotaxonomical and of the processes that control PP in ice-covered waters. photoacclimation information. Different pigment groups The concentration of Chlorophyll a, [Chl a], is the main can be identified (Johnsen and Sakshaug 2007; Roy et al. source of information to detect, map and monitor phyto- 2011) such as the carotenoid fucoxanthin (spring blooms plankton biomass (review in Johnsen et al. 2011a, b). The of diatoms), Chlorophyll c (prymnesiophytes such as differences in water mass-related effects on phytoplankton Emiliania huxleyi and Phaeocystis pouchetii) and small dynamics in the Svalbard area, often seen as the balance prasinoxanthin-containing prasinophytes, typical for this between the warm and saline water masses of Atlantic area (Pettersen et al. 2011). Likewise, HPLC-isolated pig- origin versus the colder and fresher water masses of Arctic ments comprise information of light-harvesting pigments origin, is reflected in the differences in biodiversity and (LHP) and photoprotective carotenoids (PPC), (Johnsen 1 3 Polar Biology (2018) 41:1197–1216 1199 and Sakshaug 2007; Brunet et  al. 2011; Johnsen et  al. vertically and horizontally below 1 m depth and under sea 2011a). The degradation state of chlorophylls can be also ice, in contrast to satellite-derived data. used to indicate the physiological state of the phytoplank- Based on this, we report on an under-ice bloom consisting ton cells, i.e. pre-bloom, bloom and post-bloom state of HL-acclimated cells in the Eurasian sector of the Arctic (Johnsen and Sakshaug 2000; Roy et al. 2011). Ocean (southern part of the Yermak Plateau, Fig. 1)resulting Through photoacclimation, the cells are optimizing the from northward advection of water masses. This study shows light harvesting properties, the corresponding photosyn- that by combining information from three observational plat- thetic electron transport rates (ETR, see Materials and forms (satellite, AUV and ship) we obtain a “picture” of Methods) and pigment functions (light harvesting ver- overall biomass in both open waters (satellite) and under sus photoprotective) to provide maximum growth rates sea ice (AUV and ship). The ship-based CTD equipped with (Falkowski and Raven 1997; Sakshaug et al. 1997; Brunet an in situ Chl a fluorometer and Niskin bottles for water et al. 2011). Chlorophyll a fluorescence-based photosyn- sampling (phytoplankton) made it possible to study the thetic relative ETR rates (rETR), are typically measured by taxa diversity and photoacclimation status (photosynthetic means of PAM (Genty et al. 1989; Kromkamp and Forster 2003; Hancke et al. 2008a, b) or by fast repetition rate fluorometer, FRRF (Kolber and Falkowski 1993; Suggett et al. 2009). The photoacclimation status of phytoplankton can be identified by looking at the photosynthetic param- eters termed maximum light utilization coefficient (α , ETR similar to α for C-based P vs. E curves), Chl a fluo- rescence-based maximum rETR rate, rETR (similar to max maximum photosynthetic rate, P for C-based P vs. E max curves), and the corresponding light saturation parameter −2 −1 (E , = rETR /α in μmol quanta m s ) (Behrenfeld k max ETR et al. 2004; Hancke et al. 2008b; Schuback et al. 2016). Using E we can divide cells into high-light- (HL, cells growing in “surface open waters”) and low-light- (LL, cells grown “under sea ice” or deep waters) acclimated cells (Sakshaug et al. 2009b). The in situ E (μmol photons −2 −1 m s ) of spring bloom forming phytoplankton are typi- −2 −1 cally 50–60 μmol photons m s at 0–10 m and 30–50 m depth, respectively in May–June (numbers from the Bar- ents Sea using C-incubated cells, at latitudes comparable to the west of Spitsbergen at 78°N (reviewed by Sakshaug et al. 2009b). To address our key scientific question “are the observed under-ice phytoplankton blooms a result of local production, or are most of the algal cells transported from open and well- lit waters south of the Marginal Ice Zone (MIZ)”, we com- bined information of phytoplankton biomass [Chl a] data from: Satellite (open water bloom south of ice edge covering 32,000  km ), an AUV deployed under sea ice [providing areal coverage of under-ice light transmission patchiness, (Chl a), temperature and salinity; horizontally 80,000 m and vertically 16,500 m ], vessel-based in situ profiles and corresponding work on cells in the laboratory (in vivo and Fig. 1 Remotely sensed sea surface phytoplankton bloom covering 2 −3 32,000  km (green–yellow area), defined as [Chl a]  >  4  mg  m , in in vitro). The major “pro” of using polar orbiting satellites May 2010 (mean values) W and N of the Svalbard archipelago (black equipped with ocean colour multispectral imagers for [Chl −3 area). The phytoplankton biomass (colour bar in mg Chl a m ), sea- a] are the covering of large areas while the “cons” are that ice distribution (white areas) with ice stations (red circles, northern- the information is restricted to water surface (upper 1–2 m), most stations at 31 km north of MIZ). The ice edge at 16 and 17 May indicated a southward ice movement during the campaign. Note that dependent on illumination from the sun and absence of the southernmost station was in open waters during the actual tran- clouds (review in Johnsen et al. 2011b). Autonomous under- sect. Image based on Envisat satellite with MERIS multispectral water vehicles (propelled AUV’s) may cover the [Chl a] imager. For details, see Materials and methods 1 3 1200 Polar Biology (2018) 41:1197–1216 parameters) in addition to pigment speciation and function of the living cells sampled under the sea ice. Materials and methods The measurements were made to the north and south of the Marginal Ice Zone (MIZ) north of Svalbard 15–18 May 2010 (Fig. 1, Table 1). During sampling, the MIZ was at station (St) 6, St 7 in open waters (Fig. 1–2, Table 1) and the sea-ice coverage at St 1–5 was close to 100% with an ice thickness of 1.1–1.9 m and with some leads present. At station 1–3 the ice cores were ranging from 122 cm ± 12 (n = 11, SD) to 146 cm ± 40 (n = 8, SD). In addition, the corresponding snow depth on top of sea ice were ranging from 23 to 30 cm (n = 16). Ice cores were melted for ice algal analyses in cool- ing lab at 4°C in research vessel under dim light. Observational platforms and sensors Envisat satellite data from the MEdium resolution imaging spectrometer (MERIS) were used to identify and map the −3 extent of the phytoplankton bloom (Chl a, mg m ) south of the MIZ (Fig. 1–2) through a monthly mean [Chl a] map of the area. All available reduced resolution data (MERIS-RR) from May 2010 with a pixel size of 1 km from the MERIS archive at Brockmann Consults were used. The Algal-1 Chl a product algorithm from the 3rd reprocessing was pro- cessed using the software BEAM version 5.0.1 to produce the L3 binning product. A REMUS 100 AUV from Hydroid Inc. (Moline et al. 2005) was used under the sea ice at St 3 the 16 May 2010 at 11:00–12:15 local time (Figs. 3, 4, 5, Tables 1, 2, 3). Fig. 2 Ship-based under-ice 0–200  m vertical distribution of salin- The REMUS 100 was equipped with both ultra short base ity (upper panel), temperature (mid panel) and [Chl a] (lower panel) line (USBL) and long base line (LBL) transducers for through a transect line of 31  km north of the MIZ. Sea-ice stations positioning and navigation under the ice. With a moving 1–6, station 3 was the site for AUV survey, station 1–2 northernmost ice field (typically at ≈ 0.5 knots, data from ship log), it stations (right side), MIZ is the marginal ice zone and open water (station 7, southernmost) Table 1 Overview of ice station 1–7 with date, time of day (local depth of water samples for pigment (all pigments by HPLC and cor- time), latitude, longitude, CTD and Chl a fluorometer vertical tran- responding in vitro Chl a analyses) and in vivo photosynthesis sect from 0 to 200  m depth (all stations) from RV Helmer Hanssen, Station Date May Time Latitude Longitude Water sample Photosynthesis sam- Pigments sample AUV sam- depths (m) ple depths (m) depths (m) ple depths (m) 1 15 10–13 80.665°N 5.048°E 2, 10 2, 10 2, 10 2 15 22–01 80.673°N 5.100°E 2, 10, 40 2, 10, 40 2, 10, 40 3 16 10–18 80.640°N 5.024°E 10, 45 10, 45 10, 45 5–55 4 16 23–24 80.607°N 4.841°E 5 17–18 24–01 80.566°N 4.918°E 6 17–18 24–01 80.528°N 5.097°E 7 18 03–04 80.416°N 5.307°E Downwelling irradiance E and vertical net hauls of phytoplankton (both from surface to 50 m depths) were carried immediately after CTD profiles PAR 1 3 Polar Biology (2018) 41:1197–1216 1201 Fig. 3 AUV-based under sea-ice photomosaic images at 10  m depth images along track line shown on right side. White frame indicates (horizontal survey, covering 80,000  m ) under the sea ice the 16th area in b and c. b Complete photomosaics along grid line at 10  m May 2010 at  11–12:15 local time at ice station 3. Deployment at depth. c Details (from b, white frame) of geo-tagged photomosaic −1 80°640′ N, 5°024′ E at average speed of 1.5  m  s . At noon, under- images of under-ice light climate and ice morphology as a function of −2 −1 ice E was between 1.2 and 2.1 μmol photons m s at 10 m depth thick sea ice (dark picture frames), leads and thin ice (bright frames), (St 1–3, Fig. 4). At 22:00, E was typically about 30% of noon values blue areas indicating melt ponds (blue frames) and areas with high (see results). a Schematic image of underwater AUV photomosaic [Chl a] (green frames, indicating ice algae and phytoplankton). Cor- images of underside of sea ice with camera facing upwards at 10  m responding horizontal and vertical AUV transect lines for Chl a, T depth during horizontal transect grid line. Indications on photomosaic and S in Fig. 5 Fig. 4 Downwelling irradiance (E ) as a function of depth at ice station 1–3. The profiles indi- cate maximum irradiance values (measured from leads) was important to ensure that the AUV was navigating rela- Progress of the AUV mission was monitored in real time tive to the ice features. The average velocity of AUV was via acoustic modem (hydroid ranger). Two mission plans −1 1.5 m s (2.9 knots) and the total mission time was 1.15 h were run; the first was an under-ice zig-zag pattern gener - with total mission length of 6663 m. Three transponders ating three vertical transects spanning depths of 5–50 m, were placed under the sea ice; two were used for LBL the second was an under-ice grid pattern (20 m spacing) navigation for both the AUV data collection providing ver- at 10 m fixed depth to ensure clearance of pressure ridges. tical profiles and horizontal area coverage at 10 m depth The REMUS 100 was equipped with a Neil Brown CTD for under sea ice. A third transponder (homing transponder) temperature and salinity and depth measurements. In situ was used for USBL navigation to ensure return of the vehi- [Chl a] was measured with a BBFL2 Eco-Puck, with excita- cle to a particular opening in the ice for retrieval. tion wavelength maximum at 470 nm and corresponding Chl 1 3 1202 Polar Biology (2018) 41:1197–1216 Fig. 5 AUV-based vertical (left panels, black lines) and horizontal (right panels, black lines) mapping of phytoplank- ton biomass ([Chl a], lower panel), temperature (mid panel) and salinity (upper panel) of water masses under sea ice at ice station 3. Spatial coverage of −3 Chl a (mg m ), salinity (PSU) and temperature (°C) using Remus 100 AUV in vertical zig-zag mode transect covering 16,500 m and a horizontal grid area at 10 m depth covering 80,000 m (400 × 200 m) a emission measured at 695 nm (Wet Labs, Oregon, USA, 200 m depth along a transect from stations 1–7 using a calibrated by producer prior to cruise). In addition to these Sea-bird 911 CTD (Sea-Bird Electronics, Inc., USA). In sensors, an upward-looking DC1400 SeaLife digital under- addition, the Sea-bird 911 was equipped with a calibrated water camera was mounted in the nose of AUV to provide (prior to cruise) in situ Seapoint (Seapoint Inc, USA) Chl −3 geo-tagged (through time-synchronization with AUV data) a fluorometer (Chl a, mg  m ), and Niskin water sam- qualitative pictures of the underside of the sea ice provid- pler rosette (5 L) for collecting water samples for in vivo ing information of light field, ice morphology, water colour analyses of phytoplankton taxa, photosynthetic parameters (blue, green and white) related to inherent optical proper- (onboard research vessel) and for harvesting cells on fil - ties (IOP) of water and its constituents [phytoplankton, col- ters for HPLC pigment characterization (light-harvesting oured dissolved organic matter and total suspended matter, pigments, photoprotective carotenoids and degraded pig- (Johnsen et al. 2009)]. Vehicle position, time and the lens ments). At each station, immediately after the CTD pro- geometry were used in the mosaic reconstruction of the grid files, vertical net hauls of phytoplankton samples (from portion of the mission (Fig. 3). below Chl a maximum to the surface) were carried out The research vessel “RV Helmer Hanssen” (UiT, the using a phytoplankton net with 20 μm mesh size. Water Arctic University of Norway) was used for water col- sample depths are indicated in Tables 1, 2, 3. Phytoplank- umn profiles of salinity, temperature and depth from 0 to ton samples were kept alive in polyethylene containers at 1 3 Polar Biology (2018) 41:1197–1216 1203 4 °C. Cell counts and identification were performed on live cells 1–4 h after sampling. No fixatives were added to ensure that fragile naked cells were included in the analy- sis. Detailed investigation of algal eco-physiology was undertaken at sea-ice sites (St 1–3); note St 3 was also the AUV survey location. This was followed by a transect line southward (St 4–7) with ship-based in situ CTD and Chl a fluorescence profiles from surface to 200 m depth (Fig. 1, 2, 3, Table 1). −2 Downwelling irradiance (E , μmol photons m d-PAR −1 s ), was measured using two different light sensors and data loggers. Irradiance depth profiles (0–40 m at station 1–3) were measured using a calibrated Li-Cor UW1800 2π irradiance (cosine corrected) light collector connected to a Li-Cor 1100 data logger with an underwater cable to 50 m depth (1-m depth increment from surface to 10 m, 5-m depth increments for depths > 10 m, Fig. 4). E d-PAR in open deck incubators (measuring photosynthesis versus irradiance in kelp, data not shown) from a DIVING-PAM irradiance sensor (planar), measuring diurnal time series (15 min between measurements) at 0.5 m depth. Identification of phytoplankton groups Light microscopic (LM) identification of living cells from phytoplankton nets was performed to identify major taxa (class to species level). Cell identification by microscopy (200 and 400 × magnification, using at least 4 subsam- ples) was further used to separate species into “northern cold water species” and “cosmopolitan species” (Sakshaug et al. 2009b). Ice algae were sampled from the lower part of ice cores and compared with species composition from the water column in order to be able to determine pos- sible origin of the species in the water column according to (von Quillfeldt et al. 2009). In addition, HPLC-derived pigment chemotaxonomy gives information of major pig- ment groups and pigment-based biomass (including small and fragile cells that are hard to detect using LM), such as sub-groups of Chromophytes (Chl c-containing algae), Chlorophytes (Chl b-containing classes) and phycobilipro- tein-containing cyanobacteria and eukaryotic algae present (Johnsen et al. 1994; Johnsen and Sakshaug 2007; Johnsen et al. 2011a, b; Pettersen et al. 2011; Roy et al. 2011). No traces of prasinoxanthin (marker for small prasinophytes, pigment group 6 in Johnsen and Sakshaug (2007)), phy- cobiliprotein-containing cryptophytes (alloxanthin) and cyanobacteria (zeaxanthin and b-carotene), affecting pho - tosynthetic parameters (  ) were found (Table  2, see PSII photosynthetic parameters below). 1 3 −3 Table 2 Phytoplankton pigment data (HPLC, mg pigment m ) at ice station (st) 1 and 3 Station Depth (m) Chl a + Phaeo Chl a Chlc Chl b Fuco Diadino Diato LHP/Chl a + Phaeo LHP/Chl a PPC/Chl a + Phaeo PPC/Chl a 1+2 1 2 10.8 ± 0.5 9.8 ± 0.5 3.6 ± 0.17 6.0 ± 0.3 1.9 ± 0.1 0.89 ± 0.01 0.98 ± 0.01 0.17 ± 0.01 0.19 ± 0.01 10 10.7 ± 0.8 9.6 ± 0.6 3.4 ± 0.25 5.6 ± 0.4 1.6 ± 0.1 0.84 ± 0.02 0.94 ± 0.02 0.15 ± 0.01 0.17 ± 0.00 3 10 9.9 ± 0.3 7.9 ± 0.4 2.0 ± 0.43 0.2 ± 0.01 4.0 ± 0.1 1.2 ± 0.07 0.5 ± 0.01 0.63 ± 0.07 0.79 ± 0.1 0.17 ± 0.01 0.22 ± 0.02 45 7.2 ± 0.6 6.2 ± 0.4 1.9 ± 0.09 3.4 ± 0.1 1.1 ± 0.03 0.75 ± 0.06 0.87 ± 0.06 0.15 ± 0.01 0.17 ± 0.01 Total Chl a + Phaeo includes phaeophorbide a, phaeophytin a and chlorophyllide a. Standard error of average (SE, n = 3) indicated as ± after [pigment]. Light-harvesting pigments (LHP): Chl a, Chl c , Fucoxanthin (Fuco). Photoprotective carotenoids (PPC): Sum of diadinoxanthin (Diadino) and diatoxanthin (Diato). Pigment ratios in w:w. Station 3 is the site of AUV survey. Only 1+2 −3 traces (< 0.02 mg pigment m ) of Chl c and acyloxy-fucoxanthins were found under ice (indicating low biomass of prymnesiophytes such as Phaeocystis pouchetii, data not shown) 3 1204 Polar Biology (2018) 41:1197–1216 −2 −1 Table 3 Photosynthetic parameters from ice station 1–3: rETR (  × E ),  ,  , E (μmol  m s ) and non-photosynthetic max PAR ETR k PSII PSII - max quenching (NPQ) of phytoplankton Station Depth (m) rETR  (± SE) R  E NPQ E  = 180 NPQ E  = 500 max ETR k PAR PAR PSII - max (± SE) 1 2 105 ± 7.1 0.69 ± 0.10 0.88 0.60 ± 0.01 152 ± 24.3 0.35 ± 0.03 1.47 ± 0.20 10 99 ± 7.6 0.64 ± 0.10 0.85 0.62 ± 0.01 155 ± 27.0 0.23 ± 0.07 0.81 ± 0.08 2 2 144 ± 20.8 0.59 ± 0.11 0.79 0.61 ± 0.02 244 ± 57.7 0.23 ± 0.10 1.41 ± 0.30 10 96 ± 7.7 0.76 ± 0.15 0.80 0.62 ± 0.02 127 ± 27.0 0.89 ± 0.37 1.36 ± 0.30 40 81 ± 5.4 0.77 ± 0.14 0.84 0.63 ± 0.02 106 ± 20.4 0.58 ± 0.16 2.06 ± 0.47 3 10 89 ± 6.1 0.87 ± 0.17 0.82 0.61 ± 0.01 103 ± 21.2 0.23 ± 0.09 0.82 ± 0.08 45 68 ± 4.9 0.87 ± 0.20 0.78 0.61 ± 0.01 78 ± 18.9 0.43 ± 0.16 2.02 ± 0.58 Mean 97 ± 8.5 0.75 ± 0.14 0.80 0.61 ± 0.01 138 ± 28.0 0.42 ± 0.18 1.42 ± 0.36 −2 −1 −2 −1 NPQ derived from RLC curves at E of 180 μmol m s , close to maximum E (mean value of 138 μmol m s ) and corresponding light- PAR k −2 −1 saturated photosynthesis (E of 500 μmol m s ). Standard error of average (SE, n = 3) indicated as ± after photosynthetic parameters. Sta- PAR tion 3 is the site of AUV survey in 5  mL of methanol (24  h, 4  °C) and the extracts were [Chl a] in living cells (in vivo and in situ) refiltered through a 0.2 μm sterile Microsart syringe filter to and in extracts (in vitro) remove any potential light-scattering particles and debris. −1 For spectrophotometric [Chl a] measurements, filter The in vitro [Chl a], μg L , measurements were taken from extracts were measured and the optical density (OD) spectra Niskin water samples using both turner designs fluorometer (300–800 nm, 1 cm cuvette) were obtained, using methanol (10-AU-005-CE, Sunnyvale, CA, USA), a Shimadzu Model (MeOH) as blank (OD at 750 nm). The concentration of Chl 1240 UV mini spectrophotometer (see details below) and −3 a (mg m ) was calculated using the OD red peak of Chl a, a high performance liquid chromatograph equipped with a at 665 nm in vitro and the extinction coefficient for Chl a in diode array detector (Hewlett Packard series 1100 HPLC) methanol at 665 nm (Mackinney 1941), detailed in Chauton according to Rodriguez et al. (2006). Comparison between et al. (2004). in  situ fluorescence (CTD and AUV Chl a fluorometer) versus analyses on extracts in the lab (fluorometer, spec - trophotometer and HPLC) are detailed in discussion. Lab HPLC pigment identification and analysis fluorometer (turner designs) was calibrated by the laboratory spectrophotometer using Chl a standard from Sigma-Aldrich Glass fiber filters for HPLC analysis of chlorophylls and (C6144 Chl a) according to Knap et al. (1996). Due to pho- carotenoids were immediately put in a freezer at − 20 °C toacclimation affected by PQ (photosynthetic quenching, during the cruise (1  week) and moved to −  80  °C for ) and NPQ (mainly due to photoprotective pigments PSII 1 month before analysis in laboratory. The frozen filters were and low internal pH inside chloroplast at high photosyn- used to provide pigment extracts for HPLC (extraction time thetic rates), there is not a direct relationship between in vivo 24 h at 4 °C in N (g) atmosphere) in 1.6 mL 100% methanol (often called for semi-quantitative [Chl a] measurements, (in 2 mL glass vials). Extracts were refiltered (3 mL syringe (Babin 2008)) and in  vitro [Chl a]. Therefore, a precise with 0.2 μm PTFE filter) prior to being measured. HPLC calibration of in situ fluorometers is hard to obtain simply was used to separate pigments, using a Waters Symmetry because of light-induced variation of Chl a fluorescence C8 column (4.6 × 150 mm, 3.5 μm pore size), a Hewlett emitted from the cells (reviews by Babin 2008; Suggett et al. Packard Series 1100 HPLC system with a quaternary pump 2011b; Johnsen et al. 2011a). and auto sampler using the procedure outlined in Rodríguez Spectrophotometric and fluorometric [Chl a] analy- et al. (2006). 77 μL sample was mixed with 23 μL distilled ses were performed onboard the vessel, whilst HPLC was water to increase polarity and better separation of pigments. done at Norwegian University of Science and Technology The mobile phases were methanol:acetonitrile:aqueous pyri- (NTNU). Water samples for cell identification, pigment dine (ratio 50:25:25), acetonitrile:acetone (ratio 80:20) and (HPLC) and Chl a samples from CTD Niskin bottles were methanol (for cleaning). The identification of pigments was taken immediately after the CTD profiles and the AUV sur - based on retention time and the OD spectra of the pigment vey and were filtered through Whatman GF/F glass fiber was obtained with a diode array optical density detector filters (25 mm diameter) until the filters were coloured (typi- (350–800 nm) using our own pigment standards according cally 500 ml water volume due to bloom conditions). Prior to to Rodríguez et al. (2006). being measured, the algal pigments on filters were extracted 1 3 Polar Biology (2018) 41:1197–1216 1205 NPQ was calculated using Eq. 5 [Stern–Volmer equation, Photosynthetic parameters (Lakowicz 1983)]: The photosynthetic parameters from which E were calcu- F − F lated, were obtained from RLC (rapid light curve) versus E m NPQ = , (5) measured using the PAM method using a Phyto-PAM (Walz, F Germany) according to Hancke et al. (2008a, b) and Nymark where F denotes the maximum quantum yield of cells et al. (2009). This method measures the operational quan- tum yield of charge separations in PSII (Eq. 1) at increas- acclimated to dark in 5 min prior to RLC incubations (Max- well and Johnson 2000; Valle et al. 2014). ing irradiances (Genty et al. 1989). Three mL seawater was filtered through a 0.2 µm sterile filter (Microsart) and used as a blank. Afterwards, the same cuvette was used for repli- cate samples (n = 3, water from three different Niskin bot- Results tles). Fluorescence cuvette temperature was controlled with a Peltier cell (US-T/S, Walz) mimicking in situ tempera- Remote sensing of phytoplankton spring bloom ture ± 0.2 °C during incubation. Incubation irradiances for RLC were 1, 2, 4, 7, 14, 28, 56, 112, 176, 240, 304, 368, 496 The distribution of the phytoplankton spring bloom from −2 −1 satellite images in May 2010 showed elevated [Chl a] in and 624 μmol photons m s , with 30 s incubation time for each E, giving  according to Nymark et al. (2009). The the areas south of the Marginal Ice Zone (MIZ, Fig. 1) from PSII 77°N to 80.5°N. The monthly average of satellite-derived maximum quantum yield, was obtained after dark acclima- tion of cells in 5 min prior to RLC. [Chl a] in the bloom area of 32000 km varied between 4 −3 and 13 mg Chl a m (Fig. 1). The ice edge position drawn The operational quantum yield of charge separations in PSII in actinic light conditions,  , was calculated accord- in Fig. 1 is indicative of its location on May 17th, the day of PSII the southward transect (St 4–7), showing that station 7 was ing to Genty et al. (1989) (Eq 1): in open waters (sea-ice images May 2010 can be retrieved � � � � = F − F ∕F (1) PSII m 0 m by www.met.no—daily sea-ice edge analysis). Satellite data using the F (“minimum fluorescence”) and F (“maximum (EUMETSAT Ocean and Satellite Application Facility on 0 m fluorescence” during saturating flash) at different actinic Ocean and Sea Ice, OSI SAF) on sea-ice edge position at irradiances. Note that cell samples were dominated by 16 and 17th May also indicated a north to south movement chromophytes and that no phycobiliprotein-containing cryp- of sea ice from 16 to 17th May in the transect area (see ice tophytes or cyanobacteria, with major Chl a emission from edge 16 and 17 May 2010, Fig. 1). PSI, thus affecting  , were observed in LM or by HPLC PSII pigment chemotaxonomy (Johnsen and Sakshaug 2007). In situ measurements of salinity, temperature, The irradiance (E) and  gives the relative electron irradiance and [Chl a] from the research vessel PSII transfer rate, rETR (Eq. 2): Salinity, temperature and in situ fluorescence of [Chl a ] rETR = E ⋅ (2) PSII from ice stations (St. 1–6, with St 6 close to MIZ) to open where the photosynthetic rates (rETR) at a given E was fitted water (St 7) indicate two distinct layers of water in the using a non-linear least-squares Marquardt algorithm and the study area separated vertically by a thermocline and halo- equation from Webb et al. (1974) to provide photosynthetic cline (Fig.  2). The halocline was found at around 40 m parameters based on a rETR versus E curve (RLC vs. E depth at St 4–7 and slightly deeper, around 60 m depth, curve, Eq. 3): at St 1–3. At low temperatures, water density will follow salinity and therefore a strong pycnocline is also present − ⋅ E rETR = rETR 1 − exp , (3) max between the two layers. Below the halocline temperature rETR max and salinity values indicate a strong influence of Atlantic where the photosynthetic parameters are maximum rETR water masses with salinity values  >  34.7 and tempera- (rETR ) and α (maximum light utilization coefficient, tures > 0°C. The waters of Arctic origin above the halo- max ETR also called “photosynthetic efficiency”) using Sigma Plot cline had temperature and salinity characteristic of polar version 11 (Systat Software, USA). surface water [T < 0°C and S 33.8–34.0, (Lind and Ing- The photosynthetic saturation index, E was calculated valdsen 2012)]. These salinity values (34.0 < S < 34.7) are as (Eq. 4): fresher than that defined for Arctic Water. Maximum con- centrations of in situ [Chl a] from CTD ranged between 5 E = rETR ∕ , (4) k max ETR −3 and 8 mg Chl a m , and are confined above the halocline −2. −1 where E is in μmol photons m s . going from open to ice-covered waters (Fig. 2). Typically, 1 3 1206 Polar Biology (2018) 41:1197–1216 the Chl a maxima were closest to the surface under the sea Phytoplankton biodiversity ice (around 10 m depth) and deeper within the MIZ and open waters (Chl a max at 20 m depth). See in vitro [Chl The phytoplankton species composition in the water masses a] estimation in the section below. underneath the sea ice and sea-ice bottom samples was dif- In situ downwelling irradiance profiles were measured in ferent (Table 4, Fig. 6). The phytoplankton cells from the open leads at ice stations 1–3 (Fig. 3, 4). E values at 10 m water were dominated by large and typical spring bloom PAR depth (depth of the AUV horizontal survey, see Fig. 5) at species of diatoms (northern cold or cosmopolitan species) −2 all three stations varied between 1 and 2 μmol photons m followed by prymnesio-, dino-, prasino- and chrysophytes, −1 s . Importantly, as visualized by the AUV photomosaics while the underside of the ice (from ice core samples) was (Fig. 3b–c), the leads represent areas of maximum irradi- characterized by typical ice algal species (Table 4). At St ance. Underwater downwelling irradiance at St 1–3 in the 1–3, Niskin water samples and vertical phytoplankton net upper [Chl a] maximum layer at 10 m depth was close to haul from the under-ice bloom (from 35 m depth to surface) two orders of magnitude lower than the E from in  vivo were dominated by a mixture of chain-forming centric (C) photosynthesis from the phytoplankton cells, see photosyn- and pennate (P) diatoms (Bacillariophyceae), 37 species thetic parameters below (Tables 1, 2, 3). For logistical rea- were identified. Four diatom species comprised the major sons, absolute values of E from directly underneath sea biomass under ice (St 1–3) and in open waters at St 7 (the PAR ice were not possible to obtain. Diurnal surface irradiance 18–20 May), based on relative cell abundance and chemot- south of MIZ 18–19 May at 0.5 m depth (from deck incuba- axonomic pigment tracers. Abundant diatoms were the cen- −2 −1 tors) ranged from 50 μmol photons m s at midnight to tric and spring-forming species Thalassiosira antarctica var. −2 −1 670 μmol photons m s at noon. borealis (Northern cold water to temperate distribution), T. nordenskioeldii (Northern cold water to temperate species) and Chaetoceros socialis species complex (aka C. socialis, Measurements from AUV under sea ice Cosmopolitan species). Also, the prymnesiophyte (cocco- lithophore) Phaeocystis pouchetii was abundant (many small The undulating AUV transects revealed that under-ice water cells identified), but comprised low biomass indicated by temperature was generally at −  1.66  °C from 5 to 25  m HPLC. The overall biomass was dominated by diatoms, con- depth, with corresponding salinity of 33.9 (Fig. 5) charac- firmed by high diatom-specific Chl c and fucoxanthin 1+2 teristic of polar surface waters, PSW (Lind and Ingvaldsen concentrations. The chemotaxonomic tracers for Phaeocystis 2012). The vertical distribution of [Chl a] (Fig. 5) indicated pouchetii indicated low biomass relative to the larger dia- two maximum layers, one at 0–15 meters depth (with high- toms (only traces of Chl c and acyloxy-fucoxanthin found est concentrations close to the underside of the ice) and one in under-ice samples, data not shown). at 30–40 m depth. Values in these layers ranged between 3 Ice core samples of bottom side (lower 3 cm, i.e. inter- −3 and 6 mg Chl a m (Fig. 5). The horizontal AUV survey (at stitial community) from St 1 was characterized by ice algae 10 m depth below the sea ice) indicated patchy, but generally dominated by the pennate sea-ice diatoms Navicula spp. and high in situ [Chl a] with values ranging from 3.7 to 5 mg Chl Nitzschia frigida. Present in low cell numbers were the pen- −3 a m (Fig. 3, 5). Both temperature and salinity showed little nate sea-ice diatoms Entomoneis sp., Hantzschia weiprech- variability at the 10 m layer (Fig. 5). The horizontal variabil- tii, Pseudogomphonema arcticum, Synedropsis hyperborea ity in salinity and temperature at 10 m (across 80,000 m ) and the centric diatom Attheya septentrionalis. The ice algal show that the lowest salinities (range 33.85–33.90) cor- community at St 2 was by numbers totally dominated by the responded with the lowest temperatures (range − 1.66 to dinoflagellate Polarella glacialis with many resting cysts, − 1.61 °C). The under-ice photomosaic images from AUV followed by diatoms Navicula sp. 1–2 and Nitzschia frigida (horizontal mapping grid at 10 m depth at 400 × 200 m area) (Table 4). indicated a patchy and dynamic underwater light field (irra - diance values in section above) that varied as a function HPLC pigment characterization of ice/snow thickness and leads seen as differences in light intensity and colouration (Fig. 3). The brightest areas (white The HPLC pigment chemotaxonomic signatures from St.1 hues in Fig. 3b–c) were found underneath the ice leads and and 3 revealed dominance of diatoms (Chl a, Chl c , 1+2 −2 corresponded to a downwelling E of 1–2 μmol photons m fucoxanthin and diadinoxanthin), verified by microscopy, −1 −2 −1 s at 10 m depth and around 20 μmol photons m s at at all depths. In addition, Chl b indicated the presence of the 5 m. Different hues in blue indicated sea-ice melting ponds observed Pyramimonas sp., which do not obtain the marker (observed at surface), dark green (thick ice) to brighter green carotenoid prasinoxanthin found in pigment group prasino- to brownish indicating different [Chl a ] in the water masses phyceae II (Johnsen and Sakshaug 2007) (verified by light and as biofilm of ice algae at the underside of sea ice. microscopy), was found at 10 m depth at St 3 (Table 1). Chl 1 3 Polar Biology (2018) 41:1197–1216 1207 Table 4 Phytoplankton and ice algae at ice stations (St 1–2)  and open waters (St 7)  Species Water Ice core St 1 Water Ice core St 2 Comments* column column St 1 St 7 Bacillariophyceae Amphora sp. × Often in ice, but also benthic, P Attheya septentrionalis × × + Epiphyte on phytoplankton or ice algae, C Bacterosira bathyomphala + ++ Northern cold water, spring, C Chaetoceros borealis × Cosmopolitan, C C. cf. debilis × Cosmopolitan; C C. convolutus × Cosmopolitan, usually more common during summer, C. decipiens × × Cosmopolitan, season independent, C C. furcellatus × Northern cold water, spring, C C. socialis species complex ++ + Northern cold water, spring, C C. karianus + + Northern cold water, C C. teres × Northern cold water to temperate, C Cylindrotheca closterium × × × Probably cosmopolitan, season independent, P Entomoneis sp. + × Usually spring, often also in sub-ice communities, P Eucampia groenlandica × + Typical northern cold water distribution, C Fragilariopsis cylindrus × × Bipolar? Spring (colonies may also occur in sub-ice communities. Solitary cells have been observed in interstitial communities), P F. oceanica + + Northern cold water, spring (colonies may also occur in sub-ice communities), P Hantzschia weiprechtii × Arctic, ice, P Navicula directa × Cosmopolitan, P N. pelagica × Northern cold water, spring. Usually ice-covered areas, P N. septentrionalis × × Northern cold water, spring, P N. vanhoeffenii × + Northern cold water, spring, P N. sp. 1 × +++ N. sp. 2 ++ N. spp. + + + Well-developed ice algal communities often consist of many different solitary Navicula species, P Nitzschia frigida × + × ++ Arctic, ice, P N. laevissima × × Arctic, ice, P Odontella aurita × Cosmopolitan. C Pauliella taeniata + × Northern cold water, spring, P Pleurosigma spp. × Both phytoplankton and ice, P Porosira glacialis + + Northern cold water to temperate, spring, P Proboscia alata × More common during summer, C Pseudogomphonema arcticum × x × Epiphyte on phytoplankton or ice algae, P Pseudo-nitzschia delicatissima ×x Cosmopolitan, more common during summer, P P. granii × × Northern cold water to temperate, often in Phaeocystis colonies, P P. pseudodelicatissima × Cosmopolitan. More common during summer, P P. seriata × × Northern cold water to temperate. More common dur- ing summer, P Stenoneis inconspicua var. borealis × Arctic, ice, P Synedropsis hyperborea × × + Epiphyte on phytoplankton or ice algae, P Thalassiosira antarctica var. borealis ++ ++ Northern cold water to temperate, spring, C T. bioculata × + Northern cold water, spring. Often close to ice or part of ice algal communities, C 1 3 1208 Polar Biology (2018) 41:1197–1216 Table 4 (continued) Species Water Ice core St 1 Water Ice core St 2 Comments* column column St 1 St 7 T. hyalina × +++ Northern cold water to temperate, spring, C T. nordenskioeldii ++ + Northern cold water to temperate, spring, C Coccolithophyceae Phaeocystis pouchetii ++ × Spring Dinophyceae × Gymnodinium spp. × × Katodinium sp. × Polarella glacialis × +++ Mix of vegetative cells and cysts Protoperidinium spp. × More common during summer, but several season independent species Prasinophyceae Pyramimonas sp. × × Chrysophyceae Dictyocha speculum + Cosmopolitan, season independent Dinobryon balticum + More common during summer (sometimes dominat- ing) Phytoplankton net samples (mesh size 20 μm, haul from 20 m depth to the surface). Ice algal samples from lower 3 cm of ice core. Symbols: × = present, + = regularly occurring, ++ = abundant, +++ = dominant. *Comments include biography, C (centric) and P (pennate) diatoms and own observations. Determination of phytoplankton biogeography follows the reviews (with references therein) in the region by von Quill- feldt (2000), von Quillfeldt et al. 2009 and Sakshaug et al. (2009b) −2 −1 c and 19′-acyloxufucoxanthins were found in trace amounts E of 244 μmol photons m s was found at 2 m depth and 3 k −2 −1 (pigment signature of Chl c -containing prymnesiophytes E of 127 μmol photons m s at 10 m depth followed with 3 k −2 −1 like P. pouchetii, (Johnsen et al. 2011a) at St 1–3 indicating 106 μmol photons m s at 40 m depth, comparable to St that the biomass (detected as total pigments) of Phaeocystis 3. Correspondingly, E at 10-m depth at St 1–3 was remark- pouchetii was very low (trace amounts) relative to diatoms. ably similar with 99 (St 1), 96 (St 2) and 89 (St 3) μmol pho- −2 −1 Light-harvesting pigments (LHP, chl c and fucoxantin) to tons m s , respectively. The maximum photosynthetic rate 1+2 Chl a ratio (w:w) was close to 1 (Table 2) while photoprotec- (rETRmax) was highest in cells from surface and inversely tive carotenoids (PPC, diadino- and diatoxanthin) to Chl a related to α. The mean NPQ at light-saturated photosynthesis −2 −1 ratio (w:w) was approximately 0.2. At 10 m depth at St. 3, (E of 500 μmol m -s ) from cells sampled at St. 1–3 was PAR cells contained both diadino- and diatoxanthin and the PPC 1.42 ± 25.9% (± CV of mean value, Table 3, see discussion). to Chl a ratio (w:w) was 0.22. Chlorophyll a to Chl a + Phaeo (Chl a and its degradation In situ and in vitro [Chl a] products phaeophorbide a, phaeophytin a and chlorophyllide a) ratio of 0.9 (w:w, Table 2) indicated healthy cells at all The mean [Chl a] in vitro (pigment extracts of water samples) depths at St 1 and 3 m, except at 10 m depth at St 3 with Chl from St 1–7 relative to in situ [Chl a] by AUV fluorometer and a to Chl a + Phaeo (W:W) ratio of 0.8, indicating peak of CTD fluorometer from vessel, gave an in vitro/in situ [Chl a ] bloom to start of post-bloom phase (Johnsen and Sakshaug ratio (w:w) of 2.92 ± 0.86 (n = 7) which indicates that in situ 1996). [Chl a] estimations from healthy and photosynthesizing cells (fluorometer from AUV or ship CTD fluorometer) is only Photosynthetic parameters 35% relative to in vitro [Chl a], i.e. autofluorescence (pigment extracts) not affected by PQ and NPQ, see discussion (Fig.  7). The photosynthetic parameters at ice stations indicate HL- acclimated cells characterized with E values ranging from −2 −1 78 to 244 μmol photons m s at all depths (Table 3). E Discussion −2 for 2 m depth was in the range of 105–144 μmol photons m −1 s . Lowest E values was found at 45 m depth at St 3 with In this study we report on the origin and dynamics of an −2 −1 −1 −2 78 μmol photons m s and 103 μmol photons s m at under-ice bloom observed in the Eurasian sector of the 10 m depth, correspondingly (Table 3). At St 2, the highest Arctic Ocean. Data were obtained simultaneously from 1 3 Polar Biology (2018) 41:1197–1216 1209 Fig. 6 Major phytoplankton species from Ice station 1: Chaetoceros gelidus (C. socia- lis, a), Chaetoceros karinus (b), Thalassiosira nordenskio- eldii (c), Phaeocystis pouchetii (d), Fragilariopsis oceanica (e), Navicula vanhoeffeni (f), Bacterosira bathyomphola (g) and Thalassiosira antarctica var. borealis (f). Six species of large diatoms made up the largest parts of the biomass. The high cell numbers of Phaeocys- tis pouchetii did not contribute much when looking at [Chl a] biomass using its tracer pig- ments Chl c and 19′-acyloxy- fucoxanthin (Sakshaug et al. 2009b; Johnsen et al. 2011a). Details in Table 4, see also Meshram et al. 2017 using metabarcoding of the microal- gae studied in this report three complementary platforms: Remote sensing of ocean The oceanographic conditions that were observed dur- colour (satellite), an AUV equipped with environmental ing this study are typical of conditions north and south of sensors and analyses from water samples (from a research the MIZ (Fig. 8). Whilst current vectors were not measured vessel). Recent studies in the MIZ north and west of Sval- directly, the regional flow characteristics (Walczowski et al. bard (Assmy et al. 2017; Kauko et al. 2017) have reported 2012) imply a NW transport of the subsurface waters and a on blooms developing in situ under a cover of sea ice and corresponding southward movement of sea ice in the region snow. We, however, argue that some under-ice blooms may studied in this paper. We are further guided in this assump- originate from blooms initiated in well-lit open water, and tion based on the observations and model results reported subsequently advected under sea ice. Being able to distin- by Assmy et  al. (2017) within the region studied here guish between these two scenarios is of vital importance to (80.67°–80.18°N and 4.84°–5.31°E) and at a similar time be able to provide realistic projections of production regimes of year (May 2010 vs. May–June 2015). Model data from as sea ice retreats commensurate with a reduction of the Assmy et al. (2017) show a strong NW advection of polar −1 Arctic ice cover. surface waters (PSW, mean current velocity ≈ 5 cm s at 20–30 m depth) and a NW transport of AW (mean current 1 3 1210 Polar Biology (2018) 41:1197–1216 Fig. 8 Temperature–salinity (TS) plot of water masses from station 1–7. CTD profile data for each station are shown in grey (MIZ or ice- covered stations) and black (open water station). The [Chl a] maxi- mum values are marked by green dots and are found to be consist- ently in the Polar Surface Water masses (PSW). There are no distinct differences in the TS characteristics of the stations indicating a rather consistent vertical structure of the water—mirrored in Fig. 2. In open water (Station 7), the [Chl a] maximum is deeper (25  m) compared to under the ice (~  6 to 10  m), indicative of vertical redistribution through wind-induced turbulence. The max [Chl a] is found at the bottom of the halocline at this station Fig. 7 a Differences in in  situ (living cells) versus in  vitro (pigment The presence of “Northern cold water” and “temperate” spe- extracts) [Chl a] at ice station (St) 1–3. These differences may mainly cies is typical for spring phytoplankton blooms in northern be due to quenching of Chl a fluorescence due to PQ and NPQ in liv - open waters (Hasle and Syvertsen 1996) and the dominance ing cells. Numbers in parentheses indicate ± SE, n = 3 for lab-based of centric over pennate spring species indicates that the in  vitro measurements. b In  situ [Chl a] from repetitive AUV verti- cal zig-zag  transects at ice station 3  (see also Fig.  5) compared to bloom has been developing for some time (von Quillfeldt corresponding ship based vertical [Chl a]  profile (mean of downcast 2000; Fragoso et al. 2017). The species composition in the and upcast in nearby ice-lead) water column was also different from the species composi- tion found on the underside of the ice, which are typical for −1 well-developed ice algal communities (Syvertsen 1991; von velocity of 5–10 cm s ) in the location of this study, the Quillfeldt et al. 2009). The species composition of the ice SW flank of the Yermak Plateau. Also, sea-ice data dur - algal community was dominated by diatoms at ice St 1 and ing May 2010 in the examined region showed small vari- dinoflagellates at St 2. This indicates that a “seeding stock” ations in extent and sea-ice concentration, but it started to of ice algae was not the origin of the “local bloom”, but, to disintegrate from 27 May 2010 (Norwegian Meteorological the contrary, cells in the water masses were brought under Institute, Svalbard High Resolution Ice Charts; see data base the ice by transportation of water masses from the south for May 2010 at www .polar view .me t.no). Hence, we can of MIZ (Fig. 2, 8). In sum, the continuous Chl a maxima conclude that the prevailing oceanographic and ice condi- support the assumption that the bloom observed along the tions are supporting our assumption that the observed bloom transect was one continuum of PP, rather than two distinct is of an advective origin (Fig. 8). blooms of phytoplankton in open water and under sea ice, In addition to the oceanographic conditions, we exam- respectively. In agreement with the species composition ined the taxonomical and eco-physiological phytoplankton data, the pigment data indicated that the dominating biomass status to evaluate whether the bloom was of an advective or was from diatoms (chl c and fucoxanthin) with only traces an in situ origin. First of all, we note that the phytoplank- 1+2 found of Chl c and acyloxy-fucoxanthin used as chemot- ton species composition was similar from the northernmost 3 axonomic markers for prymnesiophytes such as Phaeocys- ice station to the southernmost station in the open waters tis pouchetii (Rodríguez et al. 2006; Johnsen and Sakshaug south of MIZ (Table 4, see also Meshram et al. 2017 for 2007; Pettersen et al. 2011). We also found high E values metabarcoding data of cells close to station 1 in this study). k −2 −1 (mean E of 138 μmol m s under sea ice) indicating that 1 3 Polar Biology (2018) 41:1197–1216 1211 −2 −1 the under-ice bloom (ambient E < 2 μmol m s at 10 m affecting photosynthetic rates such as differences in incuba- depth) may have developed in open water and HL condi- tion time, absorbed quanta utilized by PSII, diurnal oscilla- tions. In addition, the taxonomic composition of the healthy tions of photosynthetic parameters and lastly, how to convert phytoplankton community (seen by  ) resembles typ- ETR rates (Chl a fluorescence from PSII) into C-fixation PSIImax ical northern planktonic species (different from ice algae rates (Johnsen and Sakshaug 2007; Hancke et al. 2008a, b; found in the sea ice). Finally, a continuum of high [Chl a] at Schuback et al. 2016). What we do know is that the rapid sea surface (remote sensing image of Chl a for May 2010) light curves (RLC) give higher E values than P versus E from open waters to 31 km north of MIZ, and relatively curves derived from PAM with 5 min incubation time or P high amounts of photoprotective carotenoids (PPC: Chl a versus E curves from “Photosynthetron technique” making ratio of 0.17–0.22 at all examined ice stations) also indicates comparison of photoacclimation (comparing E values) dif- HL-acclimated cells (Rodríguez et al. 2006; Schuback et al. ficult. PAM- and O -electrode-derived E values for diatoms 2 k 2016). We discuss each of these lines of evidence in more may be similar to short time C incubations (20–30 min) detail below. using the “Photosyntethron” technique (Lewis and Smith 1983; Johnsen and Hegseth 1991) compared to PAM and Photosynthetic parameters bio-optical-derived data (Hancke et al. 2008a). The findings of Assmy et al. (2017), in the middle part of Yermak plateau −2 −1 The mean E of 138 μmol photons m s (this study) at (May–June 2015) with a significantly lower current speed ice St 1–3 from 2 to 45 m depth indicate HL-acclimated than this study from S-Yermak plateau (May 2010), found −2 −1 cells and is comparable with HL-acclimated cells from the E values ranging from 137 to 584 μmol photons m s Arctic characterized (derived from C incubations) with of blooms dominated by Phaeocystis pouchetii, under leads −2 −1 E  > 60 μmol photons m s (Smith and Sakshaug 1990; in pack ice based on RLC curves (same method as used in Harrison and Cota 1991; Johnsen and Hegseth 1991; Stein this study). However, it should be noted that Phaeocystis and Macdonald 2003; Sakshaug et al. 2009b). These authors pouchetii generally obtain higher E that is 2–3 times higher provided a range for E in LL-acclimated phytoplankton in than diatoms under same ambient light conditions (data from −2 −1 the Arctic from 0.5 to 60 μmol photons m s with low- lab cultures and in situ data from the Barents Sea, (Sakshaug est values for cells acclimated to under-ice irradiances. We et al. 2009b)). obtained mean values of rETR , α and E as 97 ± 8.5 A study comparing photoacclimation status (E ) in max, k k −2 −1 (SE), 0.74 ± 0.09 (SE) and 138 ± 28 (SE) μmol m s , three classes of phytoplankton as a function of temperature respectively (Table  3). Under-ice blooms, dominated by (0–35 °C in diatoms, prymnesiophytes and dinoflagellates) Phaeocystis pouchetii in the mid part of the Yermak Pla- using ETR (PAM), O evolution and C incubation, gave teau with sea ice influenced by leads, leading more light E values in agreement with each other from 0 to 15 °C for into the under-ice water column, using same PAM method diatoms (Hancke et al. 2008a). In the latter study, the O as described here, gave E values of 137–584 μmol pho- method gave generally the highest E values when looking K k −2 −1 tons m s (Assmy et al. 2017). Cultured cells of Arc- at all three phytoplankton classes, followed by rETR and tic diatoms (Chaetoceros furcellatus and Thalassiosira C-method (pigment group dependent). Closest agreement nordenskioeldii) that were HL acclimated (grown at E of was found between E derived from O and rETR data, with k 2 −2 −1 14 400 μmol photons m s at 24 and 12 h day length) and C generally giving lowest E values. Since most E values k k −2 −1 14 LL acclimated (grown at E of 25 μmol photons m s at in literature are based on C method, a direct comparison 24 and 12 h day length) isolated from the same area in the may be hard to interpret. Still, our E data indicate that the Barents Sea (78°N, 30°E) obtained typically an E of 60 diatom cells are HL acclimated and is in the low range of (12 h day length) to 70 (24 h day length) and 31 (12 h) to under-ice E findings by Assmy et al. (2017) of Phaeocystis −2 −1 35 (24 h) μmol photons m s , respectively (Sakshaug pouchetii-dominated bloom, using the same RLC method as et  al. 1991, review in Sakshaug et  al. 2009b). Compar- in this study. For details of underwater irradiance in similar ing C data with ETR data is difficult and currently not type of sea ice at same time of the year, Kauko et al. (2017) resolved. Photosynthetic C- incubators were widely used observed similar light conditions as in this study. until the end of the 1990’s, typically using the “Photosyn- The NPQ of in vivo Chl a fluorescence is a physiological thetron” photosynthesis versus irradiance curve (P vs. E response related to the lowering of pH (during photosynthe- curve) technique, often with 20 min E incubation time sis) in the thylakoids inside the chloroplasts which induces PAR (Lewis and Smith 1983), compared to rETR data derived de-epoxidation of diadinoxanthin to diatoxanthin in diatoms from short incubation time experiments using PAM or FRRF and prymnesiophytes (coccolithophytes) through the “xan- (0.2–5 min incubation time per E to provide ETR vs. E thophyll cycle” of PPC (Brunet et al. 2011; Valle et al. 2014), PAR curves (Suggett et al. 2011a; Schuback et al. 2016)). There state II–I transitions between PSII and PSI (occurs mainly in are still several challenges in understanding the variables green algae and phycobiliprotein-containing phytoplankton) 1 3 1212 Polar Biology (2018) 41:1197–1216 and finally photoinhibitory quenching (Govindjee 1995; to diatoxanthin as a response to rapid (seconds to hour) Falkowski and Raven 1997). Regarding NPQ, we found no increase of irradiance (Brunet et al. 2011). pigment groups (green algae and cryptophytes) that may possess state II–I transitions inducing further lowering of In vitro versus in vivo/in situ Chl a fluorescence fluorescence signal emitted from cells (Brunet et al. 2011; Johnsen et al. 2011a; Schuback et al. 2016) and no photoin- Caution regarding in situ (or in vivo) versus in vitro [Chl hibitory irradiances under ice were observed (E at 10 m a]-based fluorometry for interpretation of data due to PQ PAR 2 −1 depth under sea ice ranged from 1 to 2 μmol m  s ) and and NPQ are discussed below. The observed bloom reached −3 these two factors cannot explain the high E and high PPC 20 mg Chl a m (in vitro), which can be regarded as a high indicating HL-acclimated cells (Table 3). This leaves the density bloom (Roy et al. 2011). In contrast, the correspond- reduction of Chl a fluorescence emitted from cells in situ ing in situ [Chl a] ranged under the sea ice St 1–4 from 4 to −3 (Fig. 7) by the two remaining NPQ mechanisms: first, the pH 8 mg Chl a m and with the highest [Chl a] centred at St 5 −3 gradient across the thylakoid membrane (causing acidic con- (8 mg Chl a m maximum at 10 m depth, Fig. 2, 7). In addi- ditions when photosynthetic rate is high and thus reduction tion, at St 3, the AUV survey indicated patchy distribution of fluorescence emitted from the cells). Secondly, the cor - of [Chl a] under the sea ice (Fig. 5), with a maximum [Chl responding lowering of pH inducing the xanthophyll cycle a] found at 10 and 38 m depth (vertical surveys) and at 10 m of diadino- and diatoxanthin (acting as radiators, emitting depth (horizontal survey) ranging between 3 and 6.5 mg Chl −3 absorbed light energy as heat) and corresponding quench- a m . Differences between in situ and in vitro estimations of ing of Chl a fluorescence, which requires HL conditions far [Chl a] may be, in addition to patchiness of phytoplankton above E level, which were not observed under the sea ice biomass, due to quenching of Chl a fluorescence in living (Table 2, Brunet et al. 2011). This is another indication that cells and differences in methodology (HPLC with separation the cells found under the sea ice may have been advected of Chl a and its degradation products, spectrophotometer and and/or covered by sea ice from well-lit, ice-free surface bench fluorometer). In living cells (in vivo) ϕ denotes the waters from the south of MIZ. A lack of saturation in NPQ fraction of light absorbed and utilized by PSII (Photosystem was observed at ice St 1–3 dominated by diatoms, despite II, the oxygen evolving site) that is converted to Chl a fluo- photophysiologial stress on cells during RLC exposed to rescence [emission detected at 685 nm in most fluorometers, high-light conditions up to 300 times higher than ambient (Babin 2008; Suggett et al. 2011b)]. The in vivo (including under-ice irradiance (Table 3). This suggest the presence of in situ) Chl a fluorescence is different from in vitro measure- HL-acclimated cells still with the capacity of photoprotec- ments because living cells change ϕ by two major processes tive downregulation as shown in Lavaud and Kroth (2006) defined as photochemical quenching (PQ) and NPQ. The and Schuback et al. (2016). PQ of Chl a fluorescence is a function of the fraction of closed reaction centres of PSII,  , see M & M (reviewed PSII Photoprotective carotenoids and light‑harvesting in Babin 2008; Johnsen et al. 2011a; Suggett et al. 2011a; pigments b). In addition, the in vitro versus in situ [Chl a] measure- ments can be used to evaluate the effect of photosynthetic The phytoplankton sampled under the sea ice, character- absorption and utilization of ambient irradiance by looking −2 −1 ized by low under-ice E (1–2 μmol photons m s at at PQ and NPQ. The highest ϕ for Chl a fluorescence will PAR F 10 m depth) was not acclimated to the ambient light inten- be found in vitro (extracted pigments) since the Chl a has sities. High cellular contents of diadinoxanthin, indicating lost its apoproteins and is non-functional in photosynthesis ML- (medium) to HL-acclimated cells, were found at St (the rate constant for photochemistry is 0) and a value of 1 and 3 (Table 2). Cells obtained diadinoxanthin to Chl a ϕ of 30% can be obtained (autofluorescence, Owens 1991; ratios (w:w) ranging from 0.17 to 1.19 typical for ML-HL- Johnsen et al. 2011a). In living and photosynthetically active acclimated cells (Johnsen and Sakshaug 1993; Rodríguez cells, ϕ in situ is typically 0.5–5% re-emission per quanta et  al. 2006; Schuback et  al. 2016). Note that conversion absorbed, about 70% of absorbed quanta will be lost as (epoxidation) from diatoxanthin to diadinoxanthin in dim/ thermal decay (heat), and the rest, 25–30%, will be used in dark–light conditions due to handling of water samples and photochemistry affecting PQ (Falkowski and Raven 1997; filtering of cells in dim light may happen in seconds and Johnsen et al. 2011a; Suggett et al. 2011a). For diatoms a therefore the sum of diadino- and diatoxanthin would often further reduction of Chl a fluorescence emitted from liv - be the best measure of photoacclimation status (Rodríguez ing cells will be due to NPQ processes, mainly due to low et al. 2006; Brunet et al. 2011). Significant amounts of dia- pH (H + produced by the water-splitting complex in PSII, dinoxanthin, characteristic for HL-acclimated phytoplankton creating acidic condition and reducing the fluorescence emit- (Johnsen et al. 1994; Rodríguez et al. 2006; Johnsen and ted from cells) and light absorption and corresponding heat Sakshaug 2007; Brunet et al. 2011), can be de-epoxided radiation from PPC thus not sending the absorbed energy 1 3 Polar Biology (2018) 41:1197–1216 1213 further to PSII (reviews in Govindjee 1995; Falkowski and and south of the MIZ (open waters). We conclude that the Raven 1997; Babin 2008; Brunet et al. 2011; Johnsen et al. bloom observed beneath the sea ice North of Svalbard in 2011a and Suggett et al. 2011a). May 2010 was an advective phenomenon, with cells that By comparing in vitro to in situ [Chl a] ratios (w:w, with had developed in open water further South. This advective in situ and in vivo Chl a fluorescence highly sensitive to PQ mechanism is in contrast to the in situ under-ice bloom sce- and NPQ) from research vessel and/or AUV fluorometer we narios elucidated by Arrigo et al. (2012) and Assmy et al. obtained ratios of 1.3–4.4 with a mean ratio of 2.92 ± 0.86 (2017) where under-ice, advective flow was considerably (n = 7), indicating a significant loss in ϕ due to PQ and weaker compared to our study. Knowledge about the ori- NPQ (Fig. 7). The mean  of 0.61 ± 1.3% (all water sam- gin and dynamics of pelagic blooms in the MIZ and PIZ is PSII ples from st 1–3) at lowest RLC incubation (Phyto-PAM) essential in order to provide realistic projections of produc- −2 −1 irradiance of 1 μmol photons m s (similar to the light tion regimes in a future warmer, and hence less ice-covered intensities observed under sea ice) indicated healthy cells Arctic. This study also imply that comparisons/calibrations with a high fraction of open reaction centres to perform pho- between different P vs E methodologies and incubation tosynthesis, i.e. cells far from light saturation in the under- times are highly needed for a common currency to compare ice ambient light (Table 3, Hancke et al. 2008a, b). between laboratory and in situ measurements as outlined by In vitro analyses provide the most accurate measurements Schuback et al. (2016). of [Chl a] and is often based on spectrophotometric meas- Acknowledgements We thank Carsten Brockmann at Brockmann Con- urements (high precision and accuracy, but low sensitivity). sults for the processing of the MERIS data. Contributions from the In contrast, Chl a fluorometers have about 100–200 times Centre of Excellence at NTNU “AMOS” (NRC, Norwegian Research higher sensitivity than spectrophotometers, but are depend- Council, Project 223254), the NRC Project “Arctic ABC” (NRC ent on calibration from spectrophotometers using Chl a 244319) and the UNIS MSc and PhD course (AB 323-823) “Light regime and primary production in the Arctic” and the crew on RV standards (Johnsen and Sakshaug 2000; Roy et al. 2011; this Helmer Hanssen (University of Tromsø) are greatly acknowledged. study). Note that fluorometer readings of Chl a are highly Contribution by FC is through the NERC-funded Changing Arctic influenced by the different fluorescent fractions of Chl a Ocean Program project “Arctic PRIZE”. such as its degradation products chlorophyllide a and phaeo- Open Access This article is distributed under the terms of the Crea- pigments (Johnsen and Sakshaug 2000; Roy et al. 2011). The tive Commons Attribution 4.0 International License (http://creat iveco Chl a fluorescence is also in vitro highly pH dependent and mmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- adding of acid may be used to look at phaeopigments using tion, and reproduction in any medium, provided you give appropriate broadband fluorometers, spectrofluorometers or spectropho- credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. tometers (Holm-Hansen et al. 1965; Johnsen and Sakshaug 2000; Roy et al. 2011). Many put in 1–2 drops of HCl in different volumes of pigment extracts to estimate the fraction of “phaeopigments” and these readings are highly depended References on pH giving different degradation products of Chl a with different emission strength making quantification difficult Arrigo KR, Perovich DK, Pickart RS, Brown ZW, van Dijken GL, (Johnsen and Sakshaug 1993). Lowry KE, Mills MM, Palmer MA, Balch WM, Bahr F, Bates NR, Benitez-Nelson C, Bowler B, Brownlee E, Ehn JK, Frey KE, Garley R, Laney SR, Lubelczyk L, Mathis J, Matsuoka A, Mitchell BG, Moore GWK, Ortega-Retuerta E, Pal S, Polashenski Conclusions CM, Reynolds RA, Schieber B, Sosik HM, Stephens M, Swift JH (2012) Massive phytoplankton blooms under Arctic sea ice. A coordinated campaign with the main aim of studying the Science 336:1408–1408. https://doi.or g/10.1126/science.12150 65 Arrigo KR, Perovich DK, Pickart RS, Brown ZW, van Dijken GL, dynamics and origin of an under-ice pelagic bloom north of Lowry KE, Mills MM, Palmer MA, Balch WM, Bates NR, Svalbard was carried out in May 2010. A continuum of high Benitez-Nelson CR, Brownlee E, Frey KE, Laney SR, Mathis J, [Chl a] above the halocline was found along the transect Matsuoka A, Greg Mitchell B, Moore GWK, Reynolds RA, Sosik from open water and 31 km into the pack ice. The bloom HM, Swift JH (2014) Phytoplankton blooms beneath the sea ice in the Chukchi sea. Deep-Sea Res Part II 105:1–16. https ://doi. was characterized with high-light-acclimated phytoplankton org/10.1016/j.dsr2.2014.03.018 cells (based on E , PPC and NPQ) along the entire transect, Assmy P, Fernández-Méndez M, Duarte P, Meyer A, Randelhoff A, including the ice stations deepest into the sea ice. Typically, Mundy CJ, Olsen LM, Kauko HM, Bailey A, Chierici M, Cohen the Chl a maximum was closer to the surface under the ice L, Doulgeris AP, Ehn JK, Fransson A, Gerland S, Hop H, Hudson SR, Hughes N, Itkin P, Johnsen G, King JA, Koch BP, Koenig Z, (around 10 m depth) and deeper within the MIZ and open Kwasniewski S, Laney SR, Nicolaus M, Pavlov AK, Polashenski waters (20 m depth). Also, the, species composition was CM, Provost C, Rösel A, Sandbu M, Spreen G, Smedsrud LH, similar at northernmost ice stations (31 km into the sea ice) Sundfjord A, Taskjelle T, Tatarek A, Wiktor J, Wagner PM, Wold 1 3 1214 Polar Biology (2018) 41:1197–1216 A, Steen H, Granskog MA (2017) Leads in Arctic pack ice enable Hasle GR, Syvertsen EE (1996) Marine diatoms. In: Tomas CR (ed) early phytoplankton blooms below snow-covered sea ice. Sci Rep Identifying marine diatoms and dinoflagellates. Academic Press, 7:40850. https ://doi.org/10.1038/srep4 0850 San Diego Babin M (2008) Phytoplankton fluorescence: theory, current literature Holm-Hansen O, Lorenzen CJ, Holmes RW, Strickland JDH (1965) and in situ measurement. In: Babin M, Roesler CS, Cullen JJ (eds) Fluorometric determination of chlorophyll. J Cons Perm Int Real-time coastal observing systems for ecosystem dynamics and Explor Mer 30:3–15 harmful algal blooms: theory, instrumentation and modelling. Hovland EK, Dierssen HM, Ferreira AS, Johnsen G (2013) Dynamics UNESCO Publishing, Paris, pp 237–280 regulating major trends in Barents Sea temperatures and subse- Behrenfeld M, Prasil O, Babin M, Bruyant F (2004) In search of a quent ee ff ct on remotely sensed particulate inorganic carbon. Mar physiological basis for covariations in light-limited and light- Ecol Prog Ser 484:17–32. https ://doi.org/10.3354/meps1 0277 saturated photosynthesis. J Phycol 40:4–25 Hovland EK, Hancke K, Alver MO, Drinkwater K, Hokedal J, Johnsen Berge J, Renaud P, Darnis G, Cottier F, Last K, Gabrielsen T, John- G, Moline M, Sakshaug E (2014) Optical impact of an Emiliania sen G, Seuthe L, Weslawski JM, Leu E, Moline M, Nahrgang huxleyi bloom in the frontal region of the Barents Sea. J Mar Syst J, Søreide JE, Varpe Ø, Lønne OJ, Daase M, Falk-Petersen S 130:228–240. https ://doi.org/10.1016/j.jmars ys.2012.07.002 (2015) In the dark: a review of ecosystem processes during the Ingvaldsen R, Loeng H (2009) Physical oceanography. In: Sakshaug Arctic polar night. Prog Oceanogr 139:258–271. h t t p s : / / d o i . E, Johnsen G, Kovacs KM (eds) Ecosystem Barents Sea. Tapir org/10.1016/j.pocea n.2015.08.005 Academic Press, Trondheim, pp 33–64 Boetius A, Albrecht S, Bakker K, Bienhold C, Felden J, Fernandez- Johnsen G, Hegseth EN (1991) Photoadaptation of sea-ice microalgae Mendez M, Hendricks S, Katlein C, Lalande C, Krumpen T, Nico- in the Barents sea. Polar Biol 11:179–184 laus M, Peeken I, Rabe B, Rogacheva A, Rybakova E, Somavilla Johnsen G, Sakshaug E (1993) Biooptical characteristics and photoa- R, Wenzhofer F, Sc RVPA–S (2013) Export of algal biomass from daptive responses in the toxic and bloom-forming dinoflagellates the melting Arctic sea ice. Science 339:1430–1432. https ://doi. Gyrodinium aureolum, Gymnodinium galatheanum, and two org/10.1126/scien ce.12313 46 strains of Prorocentrum minimum. J Phycol 29:627–642 Brunet C, Johnsen G, Lavaud J, Roy S (2011) Pigments and photoaccli- Johnsen G, Sakshaug E (1996) Light harvesting in bloom-forming mation processes. In: Roy S, Llewellyn CA, Egeland ES, Johnsen marine phytoplankton: species-specificity and photoacclimation. G (eds) Phytoplankton pigments. Cambridge Press University, Sci Mar 60:47–56 Cambridge, pp 445–471 Johnsen G, Sakshaug E (2000) Monitoring of harmful algal blooms Chauton M, Tilstone G, Legrand C, Johnsen G (2004) Changes in along the Norwegian Coast using bio-optical methods. S Afr J pigmentation, bio-optical characteristics and photo-physiology, Mar Sci 22:309–321 during a phytoplankton succession in mesocosms. J Plankt Res Johnsen G, Sakshaug E (2007) Bio-optical characteristics of PSII and 26:315–324 PSI in 33 species (13 pigment groups) of marine phytoplank- Drinkwater KF (2011) The influence of climate variability and change ton, and the relevance for PAM and FRR fluorometry. J Phycol on the ecosystems of the Barents Sea and adjacent waters: review 43:1236–1251 and synthesis of recent studies from the NESSAS Project. Prog Johnsen G, Samset O, Granskog L, Sakshaug E (1994) In-vivo absorp- Oceanogr 90:47–61 tion characteristics in 10 classes of bloom-forming phytoplank- Falkowski PG, Raven JA (1997) Aquatic photosynthesis. Blackwell ton—taxonomic characteristics and responses to photoadaptation Science, Oxford by means of discriminant and HPLC analysis. Mar Ecol Prog Ser Fragoso GM, Poulton AJ, Yashayaev IM, Head EJH, Purdie DA 105:149–157 (2017) Spring phytoplankton communities of the Labrador Sea Johnsen G, Volent Z, Sakshaug E, Sigernes F, Pettersson LH (2009) (2005–2014): pigment signatures, photophysiology and elemen- Remote sensing in the Barents Sea. In: Sakshaug E, Johnsen G, tal ratios. Biogeosciences 14:1235–1259. https://d oi.org/10.5194/ Kovacs KM (eds) Ecosystem Barents Sea. Tapir Academic Press, bg-14-1235-2017 Trondheim, pp 139–166 Genty B, Briantais JM, Baker NR (1989) The relationship between the Johnsen G, Bricaud A, Nelson N, Prézelin B, Bidigare R (2011a) In quantum yield of photosynthetic electron transport and quenching vivo bio-optical properties of phytoplankton pigments. In: Roy of chlorophyll fluorescence. Biochem Biophys Acta 990:87–92 S, Llewellyn CA, Egeland ES, Johnsen G (eds) Phytoplankton Govindjee (1995) Sixty-three years since Kautsky: chlorophyll a fluo- pigments: characterization, chemotaxonomy and applications rescence. Aust J Plant Physiol 22:131–160 in oceanography. Cambridge University Press, Cambridge, pp Gradinger R (1996) Occurrence of an algal bloom under Arctic pack 496–537 ice. Mar Ecol Prog Ser 131:301–305 Johnsen G, Moline M, Pettersson LH, Pinckney JL, Pozdnyakov D, Hancke K, Hancke TB, Olsen LM, Johnsen G, Glud RN (2008a) Tem- Egeland ES, Schofield OM (2011b) Optical monitoring of phyto- perature effects on microalgal photosynthesis-light responses plankton bloom pigment signatures. In: Roy S et al (eds), Phyto- measured by O production, pulse-amplitude-modulated fluores- plankton pigments: characterization, chemotaxonomy and applica- cence, and C assimilation. J Phycol 44:501–514. https://doi.or g tions in oceanography. Cambridge University Press, Cambridge, /10.1111/j.1529-8817.2008.00487 .x pp 538–581 Hancke TB, Hancke K, Johnsen G, Sakshaug E (2008b) Rate of O Kauko HM, Taskjelle T, Pavlov AK, Mundy CJ, Assmy P, Duarte production derived from pulse-amplitude-modulated fluores- P, Fernández Méndez M, Olsen LM, Hudson SR, Johnsen G, cence: testing three biooptical approaches against measured O Granskog MA (2017) Windows in Arctic sea ice: light transmis- production rate. J Phycol 44:803–813. h t t p s : / /d o i . or g / 1 0. 1 1 1 sion and the role of ice algae in a refrozen lead. J Geophys Res 1/j.1529-8817.2008.00509 .x 122:1486–1505 Hancke K, Hovland EK, Volent Z, Pettersen R, Johnsen G, Moline Knap A, Michaels A, Close A, Ducklow H, Dickson A (1996) Meas- M, Sakshaug E (2014) Optical properties of CDOM across the urement of chlorophyll a and phaeopigments by fluorometric Polar Front in the Barents Sea: origin, distribution and signifi- analysis. JGOFS Rep 19:118–122 cance. J Mar Syst 130:219–227. https ://doi.org/10.1016/j.jmars Kolber Z, Falkowski PG (1993) Use of active fluorescence to esti- ys.2012.06.006 mate phytoplankton photosynthesis in  situ. Limnol Oceanogr Harrison WG, Cota GF (1991) Primary production in polar waters— 38:1646–1665 relation to nutrient availability. Polar Res 10:87–104 1 3 Polar Biology (2018) 41:1197–1216 1215 Kromkamp JC, Forster RM (2003) The use of variable fluorescence Rudels B, Bjork G, Nilsson J, Winsor P, Lake I, Nohr C (2005) The measurements in aquatic ecosystems: die ff rence between multiple interaction between waters from the Arctic Ocean and the Nordic and single turnover measuring protocols and suggested terminol- Seas north of Fram Strait and along the East Greenland Current: ogy. Eur J Phycol 38:103–112 results from the Arctic Ocean-02 Oden expedition. J Mar Syst Lakowicz JR (1983) Principles of fluorescence spectroscopy. Plenum 55:1–30. https ://doi.org/10.1016/j.jmars ys.2004.06.008 Press, New York Sakshaug E, Johnsen G, Andresen K, Vernet M (1991) Modeling of Lavaud J, Kroth PG (2006) In diatoms, the transthylakoid proton gradi- light-dependent algal photosynthesis and growth—experiments ent regulates the photoprotective non-photochemical u fl orescence with the Barents Sea diatoms Thalassiosira nordenskioeldii and quenching beyond its control on the xanthophyll cycle. Plant Cell Chaetoceros furcellatus. Deep-Sea Res Part A 38:415–430 Physiol 47:1010–1016 Sakshaug E, Bricaud A, Dandonneau Y, Falkowski PG, Kiefer DA, Leu E, Søreide JE, Hessen DO, Falk-Petersen S, Berge J (2011) Con- Legendre L, Morel A, Parslow J, Takahashi M (1997) Param- sequences of changing sea-ice cover for primary and secondary eters of photosynthesis: definitions, theory and interpretation of producers in the European Arctic shelf seas: timing, quantity, and results. J Plankt Res 19:1637–1670. https://doi.or g/10.1093/plank quality. Prog Oceanogr 90:18–32 t/19.11.1637 Lewis MR, Smith JC (1983) A small volume, short-incubation-time Sakshaug E, Johnsen G, Kovacs KM (eds) (2009a) Ecosystem Barents method for measurement of photosynthesis as a function of inci- Sea. Tapir Academic Press, Trondheim, p 589 dent irradiance. Mar Ecol Prog Ser 13:99–102 Sakshaug E, Johnsen G, Kristiansen S, Quillfeldt CH, Rey F, Slagstad Lind S, Ingvaldsen RB (2012) Variability and impacts of Atlantic D, Thingstad F (2009b) Phytoplankton and primary production. Water entering the Barents Sea from the north. Deep-Sea Res In: Sakshaug E, Johnsen G, Kovacs KM (eds) Ecosystem Barents Part I 62:70–88. https ://doi.org/10.1016/j.dsr.2011.12.007 Sea. Tapir Academic Press, Trondheim, pp 167–208 MacIntyre HL, Kana TM, Geider RJ (2000) The effect of water motion Schuback N, Flecken M, Maldonado MT, Tortell PD (2016) Diurnal on short-term rates of photosynthesis by marine phytoplank- variation in the coupling of photosynthetic electron transport and ton. Trends Plant Sci 5:12–17. h t t p s : / / d o i . o r g / 1 0 . 1 0 1 6 / s 1 3 6 0 carbon fixation in iron-limited phytoplankton in the NE subarctic -1385(99)01504 -6 Pacific. Biogeosciences 13:1019–1035. https ://doi.org/10.5194/ Mackinney G (1941) Absorption of light by chlorophyll solutions. J bg-13-1019-2016 Biol Chem 140:315–322 Slagstad D, Wassmann PFJ, Ellingsen I (2015) Physical constrains and Maxwell K, Johnson GN (2000) Chlorophyll fluorescence—a practical productivity in the future Arctic Ocean. Front Mar Sci. https://doi. guide. J Exp Bot 51:659–668org/10.3389/fmars .2015.00085 McMinn A, Hegseth EN (2007) Sea ice primary productivity in the Smith WO, Sakshaug E (1990) Polar Phytoplankton. In: Smith WO Jr northern Barents Sea, spring 2004. Polar Biol 30:289–294. https (ed) Polar oceanography part B chemistry, biology and geology. ://doi.org/10.1007/s0030 0-006-0182-x Academic Press Inc., San Diego, pp 477–525 Meshram AR, Vader A, Kristiansen S, Gabrielsen TM (2017) Micro- Søreide JE, Leu E, Berge J, Graeve M, Falk-Petersen S (2010) Timining bial Eukaryotes in an Arctic Under-Ice Spring Bloom North of in blooms, algal food quality and Calanus glacialis reproduction Svalbard. Front Microbiol 8:1099. https ://doi.org/10.3389/fmicb and growth in a changing Arctic. Glob Change Biol 16:3154– .2017.01099 3163. https ://doi.org/10.1111/j.1365-2486.2010.02175 Moline MA, Blackwell SM, Von Alt C, Allen B, Austin T, Case J, Stein R, Macdonald RW (2003) The organic carbon cycle in the Arctic Forrester N, Goldsborough R, Purcell M, Stokey R (2005) Ocean. Springer, Berlin Remote environmental monitoring units: an autonomous vehi- Suggett DJ, Moore CM, Hickman AE, Geider RJ (2009) Interpreta- cle for characterizing coastal environments. J Atmos Ocean Tech tion of fast repetition rate (FRR) fluorescence: signatures of phy - 22:1797–1808 toplankton community structure versus physiological state. Mar Mundy CJ, Gosselin M, Ehn J, Gratton Y, Rossnagel A, Barber D, Ecol Prog Ser 376:1–19. https ://doi.org/10.3354/meps0 7830 Tremblay J-E, Palmer M, Arrigo KR, Darnis G, Fortier L, Else Suggett DJ, Moore CM, Geider RJ (2011a) Estimating aquatic produc- B, Papakyriakou T (2009) Contribution of under-ice primary tivity from active fluorescence measurements. In: Suggett DJ, Pra- production to an ice-edge upwelling phytoplankton bloom sil O, Borowitzka MA (eds) Chlorophyll a fluorescence in aquatic in the Canadian Beaufort Sea. Geophys Res Lett. https ://doi. sciences. Springer, Dordrecht, pp 103–128 org/10.1029/2009G L0388 37 Suggett DJ, Prasil O, Borowitzka MA (2011b) Chlorophyll a fluores- Nymark M, Valle KC, Brembu T, Hancke K, Winge P, Andresen K, cence in aquatic sciences. Springer, Dordrecht Johnsen G, Bones AM (2009) An integrated analysis of molecular Syvertsen EE (1991) Ice algae in the Barents Sea—types of assem- acclimation to high light in the marine diatom Phaeodactylum blages, origin, fate and role in the ice-edge phytoplankton bloom. tricornutum. PLoS ONE 11:e7743. https://doi.or g/10.1371/journ Polar Res 10:277–287 al.pone.00077 43 Valle KC, Nymark M, Aamot I, Hancke K, Winge P, Andresen K, John- Pettersen R, Johnsen G, Berge J, Hovland EK (2011) Phytoplankton sen G, Brembu T, Bones AM (2014) System responses to equal chemotaxonomy in waters around the Svalbard archipelago reveals doses of photosynthetically usable radiation of blue, green, and high amounts of Chl b and presence of gyroxanthin-diester. Polar red light in the marine diatom Phaeodactylum tricornutum. PLoS Biol 34:627–635. https ://doi.org/10.1007/s0030 0-010-0917-6 ONE 9:e114211. https ://doi.org/10.1371/journ al.pone.01142 11 Rodríguez F, Chauton M, Johnsen G, Andresen K, Olsen LM, Zapata Vancoppenolle M, Meiners KM, Michel C, Bopp L, Brabant F, Carnat M (2006) Photoacclimation in phytoplankton: implications for G, Delille B, Lannuzel D, Madec G, Moreau S, Tison JL, van Der biomass estimates, pigment functionality and chemotaxonomy. Merwe P (2013) Role of sea ice in global biogeochemical cycles: Mar Biol 148:963–971 Emerging views and challenges. Quat Sci Rev 79:207–230 Roy S, Llewellyn CA, Egeland ES, Johnsen G (2011) Phytoplankton von Quillfeldt CH (2000) Common diatom species in Arctic spring pigments: updates on characterization, chemotaxonomy and appli- blooms: their distribution and abundance. Bot Mar 43:499–516 cations in oceanography. Cambridge University Press, Cambridge 1 3 1216 Polar Biology (2018) 41:1197–1216 von Quillfeldt CH, Hegseth EN, Johnsen G, Sakshaug E, Syvertsen Arctic marine climate. ICES J Mar Sci 69:864–869. https ://doi. EE (2009) Ice algae. In: Sakshaug E, Johnsen G, Kovacs KM org/10.1093/icesj ms/fss06 8 (eds) Ecosystem Barents Sea. Tapir Academic Press, Trondheim, Webb WL, Newton M, Starr D (1974) Carbon dioxide exchange of pp 285–302 Alnus rubra: a mathematical model. Oecologia 17:281–291 Walczowski W, Piechura J, Goszczko I, Wieczorek P (2012) Changes in Atlantic water properties: an important factor in the European 1 3

Journal

Polar BiologySpringer Journals

Published: Feb 13, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off