microbial food web responses to light and nutrients ... terrado08.pdf · microbial food web...

Available online at www.sciencedirect.com

s 74 (2008) 964–977www.elsevier.com/locate/jmarsys

Journal of Marine System

Microbial food web responses to light and nutrients beneath thecoastal Arctic Ocean sea ice during the winter–spring transition

Ramon Terrado a,⁎, Connie Lovejoy a, Ramon Massana b, Warwick F. Vincent c

a Département de Biologie Université Laval & Québec-Ocean, Québec City, Québec, Canada G1K 7P4b Institut de Ciències del Mar CMIMA (CSIC), Dept. Biologia Marina i Oceanografia, Passeig Marítim de la Barceloneta,

37-49, 08003, Barcelona, Spainc Département de Biologie, Université Laval & Centre d'Études Nordiques, Québec City, Québec, Canada G1K 7P4

Received 8 February 2007; received in revised form 29 October 2007; accepted 18 November 2007Available online 28 November 2007

Abstract

Wemeasured the abundance and biomass of phototrophic and heterotrophic microbes in the upper mixed layer of thewater columnin ice-covered Franklin Bay, Beaufort Sea, Canada, from December 2003 to May 2004, and evaluated the influence of light andnutrients on these communities by way of a shipboard enrichment experiment. Bacterial cell concentrations showed no consistenttrends throughout the sampling period, averaging (±SD) 2.4 (0.9)×108 cells L−1; integrated bacterial biomass for the upper mixedlayer ranged from 1.33 mg C m−3 to 3.60 mg C m−3. Small cells numerically dominated the heterotrophic protist community in bothwinter and spring, but in terms of biomass, protists with a diameter N10 µm generally dominated the standing stocks. Heterotrophicprotist biomass integrated over the upper mixed layer ranged from 1.23mg Cm−3 to 6.56mg Cm−3. Phytoplankton biomass was lowand variable, but persisted during the winter period. The standing stock of pigment-containing protists ranged from a minimum valueof 0.38 mg C m−3 in winter to a maximal value of 6.09 mg C m−3 in spring and the most abundant taxa wereMicromonas-like cells.These picoprasinophytes began to increase under the ice in February and their population size was positively correlated with surfaceirradiance. Despite the continuing presence of sea ice, phytoplankton biomass rose by more than an order of magnitude in the uppermixed layer by May. The shipboard experiment in April showed that this phototrophic increase in the community was not responsiveto pulsed nutrient enrichment, with all treatments showing a strong growth response to improved irradiance conditions. Molecular(DGGE) and microscopic analyses indicated that most components of the eukaryotic community responded positively to the lighttreatment. These results show the persistence of a phototrophic inoculum throughout winter darkness, and the strong seasonalresponse by arctic microbial food webs to sub-ice irradiance in early spring.© 2007 Elsevier B.V. All rights reserved.

Keywords: Protists; Phytoplankton; Bacteria; Nutrients; Mixed layer; Sea ice; Light response

1. Introduction

Microbes are a heterogeneous group that includesBacteria, Archaea and small Eukarya. Given their large

⁎ Corresponding author. Fax: +1 418 656 2339.E-mail address: [email protected] (R. Terrado).

0924-7963/$ - see front matter © 2007 Elsevier B.V. All rights reserved.doi:10.1016/j.jmarsys.2007.11.001

standing stocks and multiple biogeochemical roles,these communities are a key to understanding thedynamics of marine ecosystems (Karl, 2002; DeLongand Karl, 2005; Giovannoni and Sting, 2005; Suttle,2005). Although bacteria and protists have beenintensively studied in the World Ocean over the lastthree decades (Azam et al., 1983; DeLong and Karl,

mailto:[email protected]

http://dx.doi.org/10.1016/j.jmarsys.2007.11.001

965R. Terrado et al. / Journal of Marine Systems 74 (2008) 964–977

2005) there are few data from arctic waters, especiallythroughout the winter season (Ducklow, 2000; Sherr andSherr, 2002; Calbet and Landry, 2004). Oceanographicresearch in the Arctic is greatly restricted through mostof the year because of the presence of thick ice cover.The majority of studies that have reported on microbialdynamics in the Arctic are limited to summer months(Thibault et al., 1999; Auel and Hagen, 2002; Hopcroftet al., 2005; Lee and Whitledge, 2005). A notableexception is the SHEBA/JOIS project in 1997/98 thatsampled throughout winter (Sherr and Sherr, 2003;Sherr et al., 2003). This pioneering study showed thatmicrobes were active during arctic winter darkness,although at a lower rate in comparison with summer.

Microbial communities in the Arctic are likely to beconstrained by environmental factors and the widefluctuation in physical and chemical conditions through-out the year. Primary production depends upon thestrong seasonality of incident irradiance, the extent andduration of sea ice and the availability of inorganicnutrients (Smith and Sakshaug, 1990). Bacterial meta-bolism is slowed by low temperatures (Pomeroy andDeibel, 1986), while the availability of substrates forbacterial growth is affected by autochthonous primaryproduction (Azam et al., 1993; Arrigo et al., 1999) andthe influx of allochthonous organic matter from rivers(Opsahl et al., 1999; Wheeler et al., 1997), both subjectto strong seasonal variations.

The Arctic Ocean has a low-salinity, low nutrientsurface mixed layer that is separated by a strongpycnocline from deeper, more nutrient-rich waters.Despite this barrier to mixing, episodes of convectiveexchange occur due to the formation of dense water at thesurface during freezing (Solomon, 1973; Macdonaldet al., 1989). If enough dense water is formed, it sinksand flows off the shelf, ultimately reaching a depth whereit matches the density of the surrounding oceanic water.As it sinks it is replaced by nutrient-rich, off-shelf watersthat advect nutrients inshore. There is considerableinterannual variation in deep water formation (Melling,1993; Melling and Moore, 1995) and therefore in the netnutrient influx to surface waters over the shelf.

Spring is especially important in Arctic waters. Itrepresents the transition from winter darkness and is aperiod of rapidly increasing daylength (Burn, 1996).Snow cover plays an important role in the penetration oflight underneath the ice, which in turn strongly affectsautotrophic production (Vincent and Belzile, 2003;Mundy et al., 2005; Riedel et al., 2006; Mundy et al.,2007). The transition from sea ice to open waterconditions in summer causes a major increase inpenetration of light into the water column for photo-

synthetic activity (Smith and Sakshaug, 1990). Forzooplankton, these changes represent the beginning oftheir phase of active feeding and reproduction, whichcan be strongly coupled to changes at the base of themicrobial food web (Bamstedt et al., 1999; Kosobo-kova, 1999; Lischka and Hagen, 2005). However, littleis known about microbial dynamics during spring, priorto ice melt, because of the extreme difficulty of shipaccess for sampling at that time of year.

The Canadian Arctic Shelf Exchange Study (CASES)during 2003/04 provided a unique opportunity to evaluatemicrobial dynamics during the period of continuous icecover. Our research platform (CCGS Amundsen) wasfrozen into the ice in Franklin Bay, adjacent to theBeaufort Sea, and allowed continuous sampling through-out winter and spring. Our objectives in the present studywere twofold. Firstly, we aimed to follow microbial foodweb dynamics over the winter–spring period to determinethe magnitude of overwintering stocks in the arctic shelfenvironment, and the timing of growth responses to thereturn of sunlight. Secondly we aimed to determine therelative importance of light exposure and nutrient supplyin controlling microbial food web dynamics duringspring. To address the latter objective we undertook ashipboard enrichment study, and examined the responsesby microscopy and molecular methods. We applied apulsed versus batch set of nutrient additions using anexperimental protocol that has previously shown strongresponses in an arctic polynya.

2. Materials and methods

2.1. Sample collection

As part of the CASES program, the icebreaker CCGSAmundsen was stationed in Franklin Bay (70° 1.3' N,126° 25.2' W) for continuous biological and physicalstudies during winter 2003 and spring 2004. Thispermitted under-ice sampling of the water columnevery 6 days over nearly 6 months. Samples werecollected at different depths with a rosette systemequipped with 12-L Niskin bottles and a CTD sensorto obtain physical measurements down the watercolumn. During the overwintering period, the depths of20 m and below were sampled through an opening to theocean from inside the ship (“moon pool”). Depths of 2 mand 10 m were sampled with a Go-Flo (GeneralOceanics) bottle from a hole on the ice 500 m awayfrom CCGS Amundsen, in the upstream direction of thedominant current flow to minimize any influence fromthe ship. Surface irradiance measurements were takenfrom an ice camp situated 1 km from the ship.

Table 1Nutrient concentrations present in the replacement waters and atdifferent time intervals of the incubations

Day Nitrate SRP Si

Treatment

HIGH replacement 17.2 1.3 30.3LOW replacement 2.1 0.8 7.0Ti 2.1 0.8 7.0

TsHIGH 1.1±0.6 0.5±0.2 3.3±2.0LOW 1.8±0.0 0.7±0.0 6.6±0.1BATCH 0.6±0.9 0.3±0.4 2.1±2.9

Ts + 4HIGH 5.6±2.5 0.7±0.3 11.4±6.0LOW 1.3±0.3 0.6±0.1 5.0±0.9BATCH 1.0±0.1 0.5±0.1 4.0±0.5

Ts + 8HIGH 11.1±0.1 1.1±0.0 22.1±0.1LOW 1.6±0.0 0.7±0.0 6.8±0.4BATCH 1.6±0.1 0.7±0.0 7.0±0.3

Ts + 12HIGH 11.0±0.9 0.9±0.1 23.5±2.4LOW 0.1±0.2 0.4±0.3 5.0±3.2BATCH 0.6±0.5 0.5±0.2 4.8±3.0

Nitrate, soluble reactive phosphorus (SRP) and silica (Si) are given inμmol L−1. Initial values at time of collection are given (Ti). Ts: startvalue for the experiment after 2 days preincubation (average±SD,n=2); Ts + 4, Ts + 8, Ts + 12: concentrations at 4 days intervals since startof the treatments (average±SD, n=2).

966 R. Terrado et al. / Journal of Marine Systems 74 (2008) 964–977

2.2. Experimental approach

To determine the effect of nutrients on the microbialgrowth, water was collected on 23 April 2003 and semi-continuous cultures were set up using a protocol that hasbeen previously applied to the North Water polynya(Lovejoy et al., 2002b; Lovejoy et al., 2004). Surfacewater was distributed into 6 acid cleaned 4-L Nalgenepolycarbonate bottles, after rinsing 3 times with samplewater. Initial samples for chlorophyll a (chl a), flagellatesand bacteria from the same cast were collected. Additionalsurface and deep sea water were filtered through 0.2 µmmembranes tominimize biological activity in these watersand stored in cleaned, acid-rinsed containers in the darkand at 1 °C until utilization in the semi-continuouscultures.

The 4-L Nalgene bottles used as microcosms wereplaced in an incubator (Sanyo versatile environmental testchamber) set at−0.4 °C in a 0 °C room in a cycle of 12 h ofdark and 12 h of light; cells were exposed to irradiance of233±25 µmol photons m−2 s−1 after covering with aneutral density mesh, measured with a BSI QSL-1000spherical PAR sensor. The bottles were pre-incubated for2 days, after which three different treatments were appliedto duplicate bottles. The first day that the treatments wereapplied was designated TS (starting time). One treatmentwas designed to simulate a low nutrient supply (LOW),and a second a high nutrient supply rate (HIGH). Thesemi-continuous regime was designed to mimic in situsinking and grazing loss rates in the natural environment(Hargrave et al., 2002). One liter was removed everysampling day and replaced with comparatively lownutrient surface 0.2 µm-filtered seawater (LOW) or withdeep nutrient-rich 0.2 µm-filtered seawater (HIGH)(Table 1). A duplicate control treatment consisted of abatch culture (BATCH), with no replacement of the water.Microcosms were sampled every 2 days for chl a,nutrients and epifluorescence cell counts; at the end of theincubation 1 L was collected from each bottle for DNAanalysis. The subsampling time scales were chosen tomatch average growth rates for polar phytoplankton(Smith et al., 1999; Mei et al., 2002). Sampling was at thesame time of day throughout the experiment to minimizethe confounding effects of diurnal patterns, such as thosepreviously reported (Wheeler et al., 1989) that couldinterfere with the detection of longer term trends.

2.3. Pigment and nutrient analyses

Samples for total chlorophyll a (chl a) were filtered ontoWhatman GF/F filters; the chl a for the b3 µm(picophytoplankton) was prefiltered through a 3 µm filter

and then onto a Whatman GF/F filter. Samples were storedfrozen until they were extracted onshore in boiling ethanoland measured fluorometrically using a Cary Eclipsespectrofluorometer (Nusch, 1980). Water for nutrientanalysis was collected into 15 mL acid rinsed Falcon™centrifuge tubes after prefiltration through a 0.2 µmSartorius Minisart syringe filter. Ammonium concentra-tions were estimated onboard within 24 h of collection ofsamples using the method given in (Bower and Holm-Hansen, 1980). Nitrate+nitrite (nitrate), soluble reactivephosphorus (SRP) and silicate (Si) were analyzed onboardthe ship, using a Brann and Luebbe autoanalyzer II andstandard protocols (Grasshoff, 1999).

2.4. Microscopy

Samples for analysis by epifluorescence microscopywere fixed with glutaraldehyde (Electron MicroscopySciences) at a final concentration of 1%. Fixation tookplace in the dark at 4 °C for 1 to 24 h. After fixation,samples were filtered onto 0.2 µm (for bacteria) or 0.6 µm(for flagellates) black 25 mm-diameter polycarbonate


filters (Poretics) and stained with 4'6-diamino-2-pheny-lindole (DAPI, 5 µg mL−1 final concentration, Sigma),for at least 5 min. Filters were mounted onto slides withnon-fluorescent mounting oil (Zeiss Immersol 518 N).Flagellate counts were made with a Zeiss Axiovert 100fluorescence microscope equipped with a mercury lightsource. Excitation/emission filters were 330/400 forDAPI and 450–490/515 for chlorophyll. Total flagellatecountswere grouped into categories based on size (b2, 2–5, 5–10 and N10 µm) and the presence/absence ofchloroplast fluorescence. The N10 µm fraction repre-sented cells with a diameter between 10 and 20 µm, thusexcluding identifiable ciliates, but it included diatomswith a larger diameter in one axis. Bacteria wereenumerated following DAPI staining (Porter and Feig,1980) either onboard the CCGS Amundsen at 1000×magnification using an Olympus BX51 microscope or atUniversité Laval with a Zeiss Axiovert 100.

Samples for inverted light microscopy were fixedwith a mixture of glutaraldehyde and buffered paraf-ormaldehyde (1% final concentration) and stored in thedark at 4 °C (Lovejoy et al., 2002a). With the addition ofthe nucleic acid stain DAPI and use of UV and bluefluorescent excitation filters, this technique facilitatesthe differentiation of pigmented cells from hetero-trophic, non-chlorophyll containing cells. After amaximum of 8 months after collection 50 mL wereallowed to settle in sedimentation chambers for at least24 h. Between 40% and 80% of the chamber area(491 mm2) was examined at 400×; between 50 and 200nanoflagellates (cellsN10 µm in diameter) per samplewere counted. Given the small volume settled, concen-trations of larger rare species would be poorly estimated.These counts were performed with a Zeiss Axiovert 100inverted microscope at Université Laval.

2.5. DNA extraction and DGGE

During the DNA sample collection, water wasfiltered through a 3 µm pore size polycarbonate filterand the same water was then filtered through a 0.2 µmpore size polycarbonate filter. The 3 µm and 0.2 µmfilters were conserved in separate cryovials with 2 mL ofbuffer (1.8 mL of 40 mmol L−1 EDTA; 50 mmol L−1

Tris pH=8.3; 0.75 mol L−1 sucrose) at −80 °C untilextraction. DNA extraction was carried out with a DNAextraction kit (DNeasy Plant Mini Kit, Qiagen) follow-ing the manufacturer's instructions. The PCR amplifica-tion for the DGGE analyses was done in a BioRadIcycler thermal cycler. The general eukaryotic primersEuk1A and Euk516r-GC (Díez et al., 2001) were used.PCR mixtures (25 µL) contained deoxynucleotide

triphosphates (dNTPs) at a concentration of 200 µM;2 mMMgCl2; each primer at a concentration of 0.3 µM;2.5 units of Taq DNA polymerase (New EnglandBiolabs); and the PCR buffer supplied with the Taqenzyme. The PCR program and DGGE was carried outas described in Díez et al. (2004) using a DGGE 2401-Rev B model (CBS Scientific Company, Inc.).

2.6. Numerical calculations

Cell counts were converted to biovolume (Hillebrandet al., 1999). As a general rule, all cells except diatomswere assumed to be oblate spheres with a diameter of1.66 µm for cells b2 µm (as used in Booth and Horner,(1997)), 4 µm for the fraction between 2 and 5 µm, 7 µmdiameter for the 5 to 10 µm fraction, and 15 µm diameterfor the N10 µm. For diatoms an average value obtainedafter the measurement of 100 cells was used (61 µmlong and 7 µm wide). Biomass values were calculatedfrom biovolume using the carbon:volume relationshipsin Menden-Deuer and Lessard (2000) for non-diatomcells (C=0.216×V0.939, where C is the biomassexpressed in pg carbon and V the volume expressed incubic µm) and for diatoms (C=0.288×V0.811). Apply-ing these relationships yields values of 0.49 pg C cell−1

for the b2 µm cells, 5.8 pg C cell−1 for cells between 2and 5 µm, 28.3 pg C cell−1 for cells between 5 and10 µm, 242 pg C cell−1 for cells N10 µm (other thandiatoms), and 113 pg C cell−1 for diatoms. These valueswere comparable to those used by Sherr et al. (2003).Bacterial biomass was estimated using a value of 10 fgC cell−1, a value for oligotrophic waters (see Section 4).

Biomass was integrated for the surface mixed layerusing Simpson's integration with the formula Cz1, z2=ΔZ×(Xz1+Xz2+ (Xz1×Xz2)

1/2) / 3 where Cz1, z2 is theintegrated variable between depths z1 and z2, ΔZ thedistance between the two depths in m, and X is thevariable value at a given depth. These integrated valueswere divided by the total depth of integration to obtainmean concentrations.

For the experiments, in situ growth rates wereestimated from log-linear regressions of the cell countsversus experiment day. The net change in the experimentvariables was calculated for each 2-day interval, correct-ing for the removal and replacement of the seawaterculture with new water. For all treatments this was:

DCt;tþ2 ¼ Ctþ2 � 1� qð ÞCt þ qCr

where Cr was the concentration of the replacement waterand ρ was the 2-day dilution factor. For the BATCHcontrol ρ was equal to 0 (no water replacement) and for

Table 2Comparison of winter (December–March) and spring (April–May)cell abundance (cells L−1) in surface, 10 m, and inversion temperaturedepth, of bacteria and size fractionated protists (average±SD)

Depth Abundance (cells L−1)

Winter Spring

BacteriaSurface 2.4±1.0×108 2.7±1.0×108

10 m 2.3±0.7×108 2.6±1.2×108

Inversion temperature 2.1±0.5×108 1.9±0.7×108

Heterotrophic protistsb2 μm Surface 1.9±0.7×105 2.4±1.1×105

10 m 2.2±1.1×105 2.2±0.8×105


2–5 μm Surface 7.2±4.2×104 9.5±3.9×104

10 m 9.2±4.9×104 8.0±3.9×104


5–10 μm Surface 2.1±1.1×104 4.5±1.0×104

10 m 2.8±1.8×104 3.4±1.5×104


N10 μm Surface 7.3±5.3×103 1.7±1.0×104

10 m 5.4±6.8×103 1.6±1.0×104


Autotrophic protistsb2 μm Surface 7.0±4.7×104 2.9±1.6×105

10 m 7.1±4.8×104 1.8±1.8×105


2–5 μm Surface 1.9±1.8×104 1.9±0.8×105

10 m 1.9±0.8×104 1.2±0.8×105


5–10 μm Surface 1.1±0.9×104 4.0±3.3×104

10 m 3.8±4.0×103 3.0±3.4×104


N10 μm Surface 3.9±4.7×103 1.5±0.8×104

10 m 3.6±3.7×103 1.6±1.0×104



HIGH and LOW ρ was equal to 0.25. For nutrients, Cr

had a different value depending on the treatment. For therest of microbial variables, Cr was equal to 0 since allcomponents were removed by 0.2 µm filtration. Allstatistical analyses were performed with SigmaStat 3.1(Systat Software Inc.).

3. Results

3.1. Franklin Bay: physical and biological parameters

Throughout the period of sampling in winter andspring, the water column was ice-covered and salinitystratified with a characteristic halocline. A salinity of31.1 was used as an indicator of the upper limit of thishalocline, and over winter and spring this occurred at anaverage depth of circa 25 m. The bottom limit of thecold-water mass below the surface halocline was at adepth of circa 125 m. Below this was a layer of waterwith salinity greater than 33 and relatively warmertemperatures (ranging −1.25° to 0 °C). This layercontinued to the maximum water column depth of200 m. During winter and spring a temperatureinversion could be observed due to the presence ofwarmer water between two cold-water masses in thesurface layer, at depth 32.4±7.3 m (average±SD). Thistemperature inversion was associated with the haloclineso it was considered as a reliable index of the depth ofthe surface mixed layer.

Nutrient concentrations remained constant through-out winter and spring. Nitrate concentrations in theupper mixed layer were usually b3 μmol L−1 and SRPconcentrations b1 μmol L−1; maximum silicate con-centrations were ca. 10 μmol L−1 in winter, with lowervalues in spring (circa 7 μmol L−1). Waters below thehalocline showed higher nutrient concentrations,achieving maximum values of 17 μmol L−1 for nitrate,1.7 for SRP, and 33 μmol L−1 for silica (data providedby J-E Tremblay, NM Price and K. Simpson, McGillUniversity, Québec, Canada).

Bacterial concentrations (Table 2) were similarbetween winter and spring (demarcated as before andafter March 21). The bacterial population was fairlyhomogeneous down the water column and average(±SD) concentrations were 2.4 (0.9)×108 cells L−1 inthe upper mixed layer.

Eukaryotic microbes were present throughout thewater column but concentrations were greater in theupper mixed layer. The most abundant size fraction forboth phototrophic and heterotrophic organisms was theb2 µm size category (Table 2). Although the exactidentification of phytoplankton is not possible under

epifluorescence microscopy, we noted that the b2 µmphototrophs mostly corresponded to the characteristicMicromonas-like cell form, with one acentric chlor-oplast. For the photosynthetic fraction of 2 to 5 µm, aPhaeocystis-like cell form was the dominant, with twochloroplasts on either side of the nucleus. In the largerfraction (N10 µm diameter), dinoflagellate cysts wereobserved during winter and spring in the surface waters,and in winter they were the dominant cells in this sizefraction. Diatoms in winter represented roughly 20% ofN10 µm cells, but this increased to 30% in April and60% in early May.

Estimates of chl a concentrations generally followedthe same patterns as phytoplankton cell densities.Surface chl a increased as winter progressed into spring(Fig. 1). Size fractionation analyses showed that theb3 µm fraction represented on average (±SD) 60 (15) %

Fig. 2. Average heterotrophic protist, autotrophic protist and bacterialbiomass at the overwintering station 200 (mg C m−3) for the surfacemixed layer.


of the total chl a in surface and 64 (18) % of the total chla at 10 m. For the inversion temperature depth a lowerproportion of chl a was in the b3 μm fraction: 39(24) %. Chl a was positively correlated with PAR(Pearson correlation, r=0.778, pb0.01) and negativelycorrelated with nitrate (Pearson correlation, r=−0.771,pb0.01) (Fig. 1).

The average bacterial biomass concentration from thesurface to the inversion temperature depth ranged from1.33 mg C m−3 to 3.60 mg C m−3 (Fig. 2) with nosignificant changes between winter and spring(ANOVA, F=0.0143, df=21, pN0.05). Mean plank-tonic heterotrophic protist biomass for the same layer(Fig. 2) ranged from 1.23 mg C m−3 to 6.56 mg C m−3

(Table 3), with no significant difference between winterand spring (ANOVA, F=1.797, df=18, pN0.05).Picoflagellates were the most abundant eukaryoticgroup by cell numbers. However in terms of biomass,they represented 3.7±2.5% (average±SD) of the totalintegrated protist heterotrophic biomass.

Phototrophic protist biomass concentrations (Fig. 2)showed significant differences between winter andspring (ANOVA, F=11.193, df=18, pb0.01), andranged from a minimum value of 0.38 mg C m−3 inwinter to a maximal value of 6.09 mg C m−3 in spring(Table 3). As with the heterotrophic protists, thepicophytoflagellates were generally the most abundantin terms of cell numbers but in terms of biomassrepresented only 5.1±3.2% (average±SD) in winter and2.5±1.0% in spring of the total integrated phototrophicbiomass.

Fig. 1. Surface chl a concentrations, incident PAR (daily integrals) andnitrate concentrations in the surface layer during winter and spring inFranklin Bay.

3.2. Nutrient influence on the development of a springcommunity: experimental results

Nutrient concentrations in the surface water used tostart the incubations were moderately low (Table 1).Ammonium was below the detection limit and remainedundetectable throughout the entire incubation. Statisticalanalysis of the accumulated consumption of nutrientsduring the incubations showed no significant differencesamong the three different treatments for nitrate, SRP andsilica (Two way ANOVA, Nitrate: F=8.515, df=35,pN0.05; SRP: F=3.193, df=35, pN0.05; Silica:F=3,258, df=35, pN0.05). Bacterial concentrations

Table 3Comparison of winter (December–March) and spring (April–May)average mixed layer biomass (mg C m−3) of bacteria and sizefractionated protists (average±SD)

Biomass (mg C m−3)

Winter Spring

Bacteria 2.1±0.5 2.3±0.8Total protists 3.3±1.6 6.2±3.4Heterotrophic protists 2.5±1.4 3.5±1.9b2 μm 0.08±0.03 0.08±0.022–5 μm 0.4±0.1 0.4±0.25–10 μm 0.6±0.3 0.7±0.3N10 μm 1.4±1.1 2.4±1.6

Autotrophic protists 0.8±0.7 2.6±1.8b2 μm 0.03±0.03 0.06±0.022–5 μm 0.1±0.08 0.4±0.35–10 μm 0.1±0.09 0.6±0.7N10 μm 0.5±0.5 1.6±0.9


were initially 2.02×105 cells mL− 1 and average (±SD)final concentrations were similar in all treatments(3.80±1.16×105 cells mL− 1).

Heterotrophic flagellate concentrations and netbiomass production increased in all three treatments(Fig. 3). The greatest value for net heterotrophicproduction was in the LOW treatment, followed bythe HIGH and the BATCH, but these differences werenot statistically significant (Two way ANOVA,F=1.461, df=35, pN0.05). The last day of theincubation had the highest concentrations of chl a, andin both LOWand HIGH treatments the greatest increasewas from TS+ 8 to TS+ 10 (no pigment data were available

Fig. 3. Average (±SE) net cumulative biomass production ofheterotrophic protists (μmol C L−1), from Ts to Ts + 12, B: Average(±SE) net cumulative biomass production of photosynthetic protists(μmol C L−1) from Ts to Ts + 12.

for the BATCH treatment because of the limited samplevolume). The concentration of phototrophic protistsincreased in all three treatments, with cell concentrationsincreasing an order of magnitude from that measured atTi. For net phototrophic biomass (Fig. 3), LOWachieved the greatest values, followed by the BATCHand the HIGH; however, these differences were notstatistically significant (Two way ANOVA, F=2.310,df=35, pN0.05). The b2 µm cell fraction mostlyconsisted of Micromonas-like cells and the fractionbetween 2 and 5 µm was mostly Phaeocystis-like cells.Diatoms were not abundant at the start of the experiment(20% of the N10 μm fraction) however between TS+ 10and TS+ 12 they increased markedly and accounted for90% of the protists in the N10 μm fraction. Nitzschiaspp. colonies were the most abundant for all the threetreatments, with Fragilariopsis spp., Navicula spp. andSkeletonema spp. also present.

Net in situ growth rates were estimated for thedifferent size fractions. Doubling times (calculated as ln2 /µ) for heterotrophic flagellates were 18 days (b2 µmfraction), 15 days (2–5 µm fraction) and 11 days(N10 µm fraction). Phototrophic flagellates presenteddoubling times of 10 days (for both the b2 µm and 2–5 µm fractions) and 7 days (N10 µm fraction). A growthrate for the 5–10 µm fractions could not be calculatedbecause of the variable responses through time.

All incubations had similar biomass proportionsamong phototrophs and heterotrophs at the end of theincubation (Fig. 4). Also, the genetic fingerprint of thefinal community as assessed by DGGE was similar in allthe incubations (Fig. 5). The b3 µm fraction showedsome minor differences; however, the major bands werepresent in all samples. The N3 µm fraction showed theexact band composition among treatments, confirmingthe microscope observations that all treatments hadsimilar final communities.

4. Discussion

Microbial biomass at all sampling times in theFranklin Bay shelf site was low compared with otherstudies in the Arctic and elsewhere (Table 4). In part thisis related to cell concentrations. Low cell numbers werefound throughout winter and spring, resulting in meanand integrated biomass values that are at the lower endof the range previously reported for polar oceans.However, another factor that affects biomass estimatesis the selection of an appropriate cell number to biomassconversion factor, especially for bacteria. The exactcarbon content of marine bacteria is still uncertain. Arange of different conversion factors have been

Fig. 4. Biomass size structure of heterotrophic and autotrophic protist community. Initial values at time of collection are given (Ti) and average valuesfor each of the treatments at Ts+12. HP: heterotrophic protist; AP: autotrophic protist.


proposed that relate either cell numbers or biovolume tobiomass (Watson et al., 1977; Bratbak and Dundas,1984; Bratbak, 1985; Nagata, 1986; Lee and Fuhrman,1987; Borsheim et al., 1990; Psenner, 1990; Fukudaet al., 1998; Ducklow, 2000). In arctic waters a factor of20 fg C cell−1 factor has been used (Sherr et al., 1997,2003; Sherr and Sherr, 2003), however given theoligotrophic nature of this polar ocean, lower valuesare likely to be more appropriate. For a low conversionfactor of 120 fg C µm−3 (Gasol et al., 1997) an averagevalue of 10 fg C cell−1 is obtained using the biovolumedata for the Arctic Ocean presented by Sherr et al.(1997), 13 fg C cell−1 for bacterial biovolume data ofDisko Bay (Nielsen and Hansen, 1995) and 7 fg C cell−1

for the Antarctic Peninsula (Vaqué et al., 2002). In thepresent study in the Beaufort Sea, considering the smallcell characteristics of the bacteria observed and theoligotrophic nature of the sampled zone, a value of 10 fgC cell−1 was considered most appropriate to convert

bacterial counts to bacterial biomass instead of thewidely used value of 20 fg C cell−1. In the case ofprotists, an average value was considered for each sizegroup counted under the microscope, which gavebiomass conversion factors comparable to those foundin the literature for the Arctic Ocean (Sherr et al., 2003).

Heterotrophic bacteria and flagellates mostly domi-nated winter microbial biomass. Bacteria were activethrough winter and production peaked in summer(Garneau et al., in press). Ciliates were also presentduring winter and spring, although their biomass waslower than those of heterotrophic flagellates (Vaquéet al., submitted for publication). Our microscopeobservations revealed the persistence of pigmentedcells throughout winter despite the extremely lowirradiance levels caused by the snow and ice cover incombination with low incident irradiance. The chl a andcell concentrations were small (Fig. 1; Table 2), and thedominant fraction of this phototrophic community

Fig. 5. Negative images of DGGE gel of the N3 µm and the b3 µm assemblage at the end of the incubation of the different microcosms.


corresponded to Micromonas- like cells, consistent withHPLC pigment analyses of the community structure(Lovejoy et al., 2007). Larger Phaeocystis-like cells anddiatoms were also common. Most of the diatomssampled at this time of year did not appear to be healthyand their chlorophyll appeared degraded under thefluorescence microscope. In contrast, the chlorophyll in

Table 4Average biomass values (mg C m−3) of bacteria, heterotrophic protists and

Region Month–year Average biomass (

Bacteria Heterotrb20 μm

Equatorial Pacific Ocean July 1998 – 6.8North Atlantic May 1989 36.9 30.4Sub arctic Pacific Feb 1994, May 1994/95 – –Disko Bay (Greenland) June–July 1992 53.5 11.5Antarctic Peninsula November–February 1995/96 6.1 0.7Arctic Ocean July–September 1994 9.7 6.6Arctic Ocean November 1997–June 1998 3.5 1.9Arctic Ocean June–August 1998 6.4 4.5Beaufort Sea December 2003–March 2004 4.2 (2.1) 2.5Beaufort Sea March 2004–May 2004 4.6 (2.3) 3.5

Values were obtained from the literature directly or transformed to present theC:chl a average ratio from (Booth and Horner 1997). Transformations toconversion value of 20 fg C per cell (Lee and Fuhrman 1987) and for thispresented in parenthesis (see Section 4 for details).

the Micromonas-like cells was brightly fluorescent. It isinteresting that a positive environmental growth ratefrom February to April was found for this population(Lovejoy et al., 2007) indicating that they retainedviability over winter and an ability to respondimmediately to improved light conditions. Sherr et al.(2003) suggested that Micromonas sp. might be capable

autotrophic protists (b20 µm) for different oceanic regions

mg C m−3)

ophic protists Autotrophicprotists

Integration Reference

19.0 Mixed layer (Yang et al., 2004)58.3 30 m (Sieracki et al., 1993)4.9 Mixed layer (Boyd and Harrison 1999)83 Photic zone (Nielsen and Hansen 1995)2.9 40 m (Vaqué et al., 2002)40.5 50 m (Sherr et al., 1997)0.4 40 m (Sherr et al., 2003)10.9 40 m (Sherr et al., 2003)0.8 Mixed layer This study.2.6 Mixed layer This study.

integrated values. Transformations from chl a values were done usingbacterial biomass from bacteria concentrations were done using thestudy the values using a conversion factor of 10 fg C per cell is also


of phagotrophy, which could be a mechanism contribut-ing to their winter survival, reported for some photo-trophic flagellates in polar lakes (Bell and Laybourn-Parry, 2003; Laybourn-Parry et al., 2005). Cellsidentified as Micromonas pusilla are reported to becapable of phagotrophy elsewhere (González et al.,1993). We have no direct evidence from the presentstudy, since no Micromonas-like cells were recordedwith DAPI-stained fluorescent particles inside, and thispossibility requires further evaluation.

Winter community biomass was dominated byheterotrophic protists, and as spring advanced, photo-trophic organisms represented an increasing proportionof the total biomass (Fig. 2). Chl a was positivelycorrelated with PAR, implying a growth response toirradiance. The availability of this light for phytoplank-ton growth may also have been influenced by the snowmelt from the ice cover, as noted by Sherr et al. (2003).Chl a was negatively correlated with nitrate concentra-tions, implying active growth and nutrient depletion.Although nitrate is often thought of as a substrate forlarger phytoplankton rather than small cells, recentgenomic analyses of Ostreococcus, another picoprasi-nophyte closely related to Micromonas, have showngenes coding for transport systems for nitrate in additionto urea and ammonium (Derelle et al., 2006). Parallel tothe increase in phototrophic biomass there was a changein community structure. With increasing irradianceduring spring, an increase in diatom biomass wasobserved, although they did not reach bloom concentra-tions. Diatoms in spring represented as much as 60% ofthe N10 μm cells while in winter they only represented20% and microscope observations showed these cellswere mostly degraded. The carbon estimates for largephytoplankton are equivalent to the pigment valuespresented in Lovejoy et al. (2007). On the other hand,our values of carbon biomass for small-celled phyto-plankton are lower. If a ratio carbon:chl a is calculatedfor the N3 µm fraction, the average ratio is 40. For theb3 µm, this ratio falls to 10. This suggests that ourcarbon values for the picophytoplankton may beunderestimated, which may be an artifact of ourmicroscope counts. The signal from DAPI stainedcells when observed under the microscope is likely tofade more rapidly for small cells than large cells.

The pycnocline was well established throughout theperiod of sampling and effectively sealed the nutrients inthe deep water, preventing replenishment of the surfacemixed layer. Unlike other environments, there was nospring bloom and chl a never rose beyond 1 μg L−1 inFranklin Bay. Nutrient concentrations in surface waterswere low throughout winter, and our enrichment

incubations provided an experimental test of whethernutrient supply was a limiting factor during the initialphase of the winter–spring transition. The water forthese incubations was collected from the upper watercolumn covered by 1.8 m of sea ice, and the experimentsimulated a fast retreat of the ice with exposure toirradiance levels similar to those found at surface openwaters at that time of the year. The two treatments withwater replacement were designed to simulate differentnutrient flux conditions. The LOW treatment simulatedlow nutrient conditions for the upper mixed layer and atthe top of the halocline. The HIGH treatment simulatedthe entrainment of nutrients into the upper mixed layerfrom below. The N:P ratio of dissolved nutrients in theHIGH replacement waters (13.1:1) was close to theRedfield ratio of 16:1, a common feature of deep waters,while the LOW replacement waters had a lower ratio(2.6:1) indicating that these waters were potentiallymore limiting in nitrate compared to phosphorus(Guildford and Hecky, 2000). The N:Si ratio of bothreplacement waters was below 1, suggesting again thatnitrate would be the nutrient most likely to be limiting(Levasseur and Therriault, 1987). However, althoughnitrate was always potentially limiting, it was neverdepleted over the course of the experiments, even in theBATCH control culture (Table 1), suggesting thatphytoplankton were never strongly nutrient-limited.There were no differences in the total consumption ofnutrients among treatments. Even in the case wheremore nutrients were added in the HIGH treatment, themicrobial community did not respond to this enrich-ment. The ammonium levels were always below thelimit of detection, indicating that any production wasconsumed as fast as it was produced, consistent withammonium as the preferred source of nitrogen (Harrisonet al., 1996).

Despite the different nutrient conditions, no differ-ences were detected among treatments in either biomassproduction or in community structure. A marked shift incommunity composition was evident at the end of theexperiment relative to starting conditions; however, thefinal proportions of the different size fractions weresimilar among treatments (Fig. 4). The initial communityat TS was dominated by heterotrophs (N50% of totalbiomass) and by the end of the experiment, at TS+ 12, allof the microcosms were dominated by phototrophs,which contributed N80% of the total protist biomass. AtTS the phototrophic community biomass was dominatedby small organisms (b10 µm), and diatoms representedonly a small percentage (ca. 20%) of the N10 µmfraction. At the end of the experiment the N10 µmfraction contributed a greater proportion of biomass, and


was composed mostly of diatoms. This shift towards aphototrophic, diatom-rich community was not influ-enced by nutrient conditions.

Other factors could have influenced phytoplanktongrowth during the course of our incubations. Forexample, vitamin B12 is an important micronutrientfor large phytoplankton (Sañudo-Wilhelmy et al., 2006),and it has been suggested that bacteria may be animportant source of this vitamin for phytoplankton(Croft et al., 2005). In our incubations, bacterial biomassincreased towards the end of the experiment and mayhave resulted in increased vitamin B12 availability, ormay have had other stimulatory effects on phytoplank-ton growth.

A genetic fingerprinting analysis, DGGE, confirmedthat the different treatments gave rise to similar protistcommunities in all the incubations. The band patternsthat defined the community composition at the end ofthe incubations (Fig. 5) did not show major changesamong the treatments, with only minor shifts in theb3 µm fraction and undetectable change in N3 µm.Excision and sequencing of the most prominent bandswas attempted, but the majority of sequenced bandswere of poor quality or double sequences. The DGGEtechnique is largely qualitative and clean separation ofphylotypes depends mostly on relative GC content.Given that similar GC contents are found in many smalleukaryotes in Arctic waters (C. Lovejoy & M. Potvinunpublished data), it was impossible to assign a singleband to a single species. Despite this, the strongsimilarity of band patterns among the treatmentssuggested that the initial winter community hadmoved towards a new species structure once exposedto an improved irradiance regime, and that this effectoccurred irrespective of differences in nutrient supply.

Rose and Caron (2007) have suggested that in termsof growth rates, heterotrophic protists in cold waters aremore affected by low temperature than autotrophicprotists. They suggested that this might assist thedevelopment of phytoplankton blooms particularly inthe Polar Regions where there is already a mismatchbetween early diatom production and calanoid copepodabundance. In our incubations the growth rates ofheterotrophic flagellates were lower than those ofautotrophs, consistent with this notion.

Our experimental results contrast with those using asimilar protocol in early June 1998 in the North WaterPolynya (Lovejoy et al., 2004). The major finding ofthose experiments was that advective nutrient inputresulted in the production of a diatom bloom. In our casewe did not induce a diatom bloom, although theproportion of diatoms at the end of the incubations

was much higher than at the beginning. Heterotrophicflagellates, which were typical of the winter community,dominated the starting community of our incubations.Exposing this community to increased irradiance levelspromoted the development of photosynthetic organisms.At the end of the incubations, independent of nutrientconditions, the community was dominated by photo-trophs, but not in the form of a diatom bloom. In the caseof the Lovejoy et al. (2004) study, the initial communitywas already an established spring community, with ahigh proportion of photosynthetic organisms, and theinput of nutrients caused a successional change incommunity structure and biomass. This was not the casein our microbial community, where the transition from awinter to a spring community was not driven bynutrients but by light availability. Similarly, Sherret al. (2003) found an increase of Micromonas sp. andlarge phytoplankton under the ice cover in spring duringthe SHEBA project and attributed this growth toincreasing day length and the melting snow cover.These two factors would also contribute to the increasein phototrophic cells in Franklin Bay in spring. Incontrast, the influence of increasing nutrients may berestricted to open water conditions (Sieracki et al., 1993;Boyd and Harrison, 1999; Klein et al., 2002; Yang et al.,2004).

The winter–spring transition is of special importanceto higher trophic levels. During spring there was anincrease in both phototrophic and heterotrophic biomass(Table 3), which can be a food resource for herbivorousand omnivorous copepods in the Arctic basin (Thibaultet al., 1999). The increase of protist biomass was mostpronounced for the N10 μm cells, which are thepreferred prey of zooplankton (Stevens et al., 2004).Copepods may survive winter on lipid reserves withoutactive feeding (Scott et al., 2000) but in spring, theirfeeding and reproductive strategies are strongly coupledto primary production (Bamstedt et al., 1999; Kosobo-kova, 1999; Lischka and Hagen, 2005). The timing andchanges in microbial community structure during thewinter–spring transition are therefore likely to have adirect influence on the transfer of organic matter andenergy to higher trophic levels.

5. Conclusion

Our results show that a microbial community persiststhroughout winter in the Arctic shelf environment.Heterotrophic organisms dominated the winter commu-nity, but a small community of phototrophs was alsomaintained throughout winter. The relative abundanceof phototrophs increased rapidly during the winter–


spring transition, which would have implications forhigher trophic levels. Our experimental study indicatedthat this transition from a heterotrophic winter commu-nity to a phototrophic spring community was littleinfluenced by nutrient supply. Our experiments underdifferent nutrient regimes did not show major differ-ences in the community dynamics, which in all casesmoved from a heterotrophic dominated community to anphototrophic dominated community in response toincreased irradiance. These results are consistent withthe observations made during a previous study of anArctic winter–spring transition (Sherr et al., 2003) andimply a general response under arctic sea ice toimproved irradiance conditions during the winter–spring transition, with little control by nutrient supplyat that time of year.

Acknowledgements

Financial support for this study was provided by theNatural Sciences and Engineering Research Council ofCanada, the Canada Research Chair program, Fondsquébécois de recherche sur la nature et les technologiesand Indian and Northern Affairs Canada. We thank theofficers and crew of the CCGS Amundsen, and C.Martineau, S. Roy, M. Rautio, K. Simpson, J-E.Tremblay, N.M. Price, A. Riedel, M. P. Lizotte and T.Papakyriakou for field assistance and analysis. Thiswork was conducted within the Canadian Arctic ShelfExchange Study (CASES) led by L. Fortier.

References

Arrigo, K.R., Robinson, D.H., Worthen, D.L., Dunbar, R.B., DiTullio,G.R., VanWoert, M., Lizotte, M.P., 1999. Phytoplankton commu-nity structure and the drawdown of nutrients and CO2 in thesouthern ocean. Science 283 (5400), 365–367.

Auel, H., Hagen, W., 2002. Mesozooplankton community structure,abundance and biomass in the central Arctic Ocean. MarineBiology 140, 1013–1021.

Azam, F., Fenchel, T., Field, J.G., Gray, J.S., Meyer-Reil, L.A.,Thingstad, F., 1983. The ecological role of water-column microbesin the sea. Marine Ecology. Progress Series 10, 257–263.

Azam, F., Smith, D.C., Steward, G.F., Hagström, A., 1993. Bacteria-organic matter coupling and its significance for oceanic carboncycling. Microbial Ecology 28, 167–179.

Bamstedt, U., Nejstgaard, J.C., Solberg, P.T., 1999.Utilisation of small-sized food algae by Calanus finmarchicus (Copepoda, Calanoida)and the significance of feeding history. Sarsia 84, 19–38.

Bell, E.M., Laybourn-Parry, J., 2003. Mixotrophy in the antarcticphytoflagellate, Pyramimonas gelidicola (Chlorophyta: Prasino-phyceae). Journal of Phycology 39, 644–649.

Booth, B.C., Horner, R.A., 1997. Microalgae on the Arctic OceanSection, 1994: species abundance and biomass. Deep-sea Research.Part 2. Topical Studies in Oceanography 44 (8), 1607–1622.

Borsheim, K.Y., Bratbak, G., Heldal, M., 1990. Enumeration andbiomass estimation of planktonic bacteria and viruses bytransmission electron microscopy. Applied and EnvironmentalMicrobiology 56 (2), 352–356.

Boyd, P., Harrison, P.J., 1999. Phytoplankton dynamics in the NEsubarctic Pacific. Deep-sea Research. Part 2. Topical Studies inOceanography 46, 2405–2432.

Bratbak, G., 1985. Bacterial biovolume and biomass stimations.Applied and Environmental Microbiology 49 (6), 1488–1493.

Bratbak, G., Dundas, I., 1984. Bacterial dry matter content and biomassestimations.Applied andEnvironmentalMicrobiology48 (4), 755–757.

Bower, C.E., Holm-Hansen, T., 1980. A salicylate-hypochloride methodfor determining ammonia in seawater. Canadian Journal of Fisheriesand Aquatic Sciences 37, 794–798.

Burn, C., 1996. The Polar Night. Scientific Report, 4. Aurora College,Inuvik, pp. 1–11.

Calbet, A., Landry, M.R., 2004. Phytoplankton growth, microzoo-plankton grazing, and carbon cycling in marine systems. Limnol-ogy and Oceanography 49 (1), 51–57.

Croft,M.T., Lawrence, A.D., Raux-Deery, E.,Warren,M.J., Smith,A.G.,2005. Algae acquire vitamin B12 through a symbiotic relationshipwith bacteria. Nature 438, 90–93.

DeLong, E.F., Karl, D.M., 2005. Genomic perspectives in microbialoceanography. Nature 437, 336–342.

Derelle, E., Ferraz, C., Rombauts, S., Rouze, P., Worden, A.Z., Robbens,S., Partensky, F., Degroeve, S., Echeynie, S., Cooke, R., Saeys, Y.,Wuyts, J., Jabbari, K., Bowler, C., Panaud, O., Piegu, B., Ball, S.G.,Ral, J.-P., Bouget, F.-Y., Piganeau, G., De Baets, B., Picard, A.,Delseny, M., Demaille, J., Van de Peer, Y., Moreau, H., 2006.Genome analysis of the smallest free-living eukaryote Ostreococcustauri unveils many unique features. Proceedings of the NationalAcademy of Sciences 103 (31), 11647–11652.

Díez, B., Pedrós-Alió, C., Marsh, T.L., Massana, R., 2001. Applicationof denaturing gradient gel electrophoresis (DGGE) to study thediversity of marine picoeukaryotic assemblages and comparison ofDGGE with other molecular techniques. Applied and Environ-mental Microbiology 67 (7), 2942–2951.

Díez, B., Massana, R., Estrada, M., Pedrós-Alió, C., 2004. Distributionof eukaryotic picoplankton assemblages across hydrographicfronts in the Southern Ocean, studied by denaturing gradient gelelectrophoresis. Limnology and Oceanography 49 (4), 1022–1034.

Ducklow, H., 2000. Bacterial production and biomass in the oceans. In:Kirchman, D.L. (Ed.), Microbial Ecology of the Oceans.Wiley-Liss,New York.

Fukuda, R., Ogawa, H., Nagata, T., Koike, I., 1998. Direct determinationof carbon and nitrogen contents of natural bacterial assemblages inmarine environments. Applied and Environmental Microbiology 64(9), 3352–3358.

Garneau, M.-E., Roy, S., Lovejoy, C., Gratton, Y., Vincent, W.F., inpress. Seasonal dynamics of bacterial biomass and production in acoastal arctic ecosystem: Franklin Bay, western Canadian. Arctic.Journal of Geophysical Research.

Gasol, J.M., del Giorgio, P.A., Duarte, C.M., 1997. Biomassdistribution in marine planktonic communities. Limnology andOceanography 42 (6), 1353–1363.

Giovannoni, S.J., Sting, U., 2005. Molecular diversity and ecology ofmicrobial plankton. Nature 437, 343–348.

González, J.M., Sherr, B.F., Sherr, E.B., 1993. Digestive enzyme activityas a quantitative measure of protistan grazing: the acid lysozymeassay for bacterivory.Marine Ecology. Progress Series 100, 197–206.

Grasshoff, K., 1999. Methods of Seawater Analyses. New York,Weinheim.


Guildford, S.J., Hecky, R.E., 2000. Total nitrogen, total phosphorus,and nutrient limitation in lakes and oceans: is there a commonrelationship? Limnology and Oceanography 45 (6), 1213–1223.

Hargrave, B.T., Walsh, I.D., Murray, D.W., 2002. Seasonal and spatialpatterns in mass and organic matter sedimentation in the NorthWater Polynya. Deep-sea Research. Part 2. Topical Studies inOceanography 49, 5227–5244.

Harrison, W.G., Harris, L.R., Irwin, B.D., 1996. The kinetics ofnitrogen utilization in the oceanic mixed layer: nitrate andammonium interactions at nanomolar concentrations. Limnologyand Oceanography 41 (1), 16–32.

Hillebrand, H., Duerselen, C.D., Kirschtel, D., Pollingher, U., Zohary,T., 1999. Biovolume calculation for pelagic and benthic micro-algae. Journal of Phycology 35 (2), 403–424.

Hopcroft, R.R., Clarke, C., Nelson, R.J., Raskoff, K.A., 2005.Zooplankton communities of the Arctic's Canada Basin: thecontribution of the smaller taxa. Polar Biology 28, 198–206.

Karl, D.M., 2002. Hidden in a sea of microbes. Nature 415, 590–591.Klein, B., et al., 2002. Phytoplankton biomass, production and

potential export in the North Water. Deep-sea Research. Part 2.Topical Studies in Oceanography 49, 4983–5002.

Kosobokova, K.N., 1999. The reproductive cycle and life history ofthe Arctic copepod Calanus glacialis in the White Sea. PolarBiology 22, 254–263.

Laybourn-Parry, J., Marshall, W.A., Marchant, H.J., 2005. Flagellatenutritional versatility as a key to survival in two contrastingAntarctic saline lakes. Freshwater Biology 50, 830–838.

Lee, S., Fuhrman, J.A., 1987. Relationships between biovolume andbiomass of naturally derived marine bacterioplankton. Applied andEnvironmental Microbiology 53 (6), 1298–1303.

Lee, S.H., Whitledge, T.E., 2005. Primary and new production in thedeep Canada Basin during summer 2002. Polar Biology 28,190–197.

Levasseur, M., Therriault, J.-C., 1987. Phytoplankton biomass andnutrient dynamics in a tidally induced upwelling: the role of theNO3:SiO4 ratio. Marine Ecology. Progress Series 39, 87–97.

Lischka, S., Hagen, W., 2005. Life histories of the copepods Pseu-docalanus minutus, P. acuspes (Calanoida) and Oithona similis(Cyclopoida) in the Arctic Kongsfjorden (Svalbard). Polar Biology28, 910–921.

Lovejoy, C., Carmack, E.C., Legendre, L., Price, N.M., 2002a. Watercolumn interleaving: a new physical mechanism determiningprotist communities and bacterial states. Deep-sea Research. Part2. Topical Studies in Oceanography 47 (6), 1819–1831.

Lovejoy, C., Legendre, L., Price, N.M., 2002b. Prolonged diatomblooms and microbial food web dynamics: experimental resultsfrom an Arctic polynya. Aquatic Microbial Ecology 29, 267–278.

Lovejoy, C., Price, N.M., Legendre, L., 2004. Role of nutrient supply andloss in controlling protist species dominance and microbial food-webs during spring blooms. Aquatic Microbial Ecology 34, 79–92.

Lovejoy, C., Vincent, W.F., Bonilla, S., Roy, S., Martineau, M.J.,Terrado, R., Potvin, M., Massana, R., Pedrós-Alió, C., 2007.Distribution, phylogeny and growth of cold-adapted picoprasino-phytes in arctic seas. Journal of Phycology 43, 78–89. doi:10.1111/j.1529-8817.2006.310.x.

Macdonald, R.W., Carmack, E., McLaughlin, F., Iseki, K., Macdonald,D.M., O'Brien, M.C., 1989. Composition and modification ofwater masses in the Mackenzie shelf estuary. Journal ofGeophysical Research 94, 18057–18070.

Mei, Z.-P., Legendre, L., Gratton, Y., Tremblay, J.-É., LeBlanc, B.,Mundy, C.J., Klein, B., Gosselin, M., Larouche, P., Papakyriakou,T.N., Lovejoy, C., von Quillfeldt, C.H., 2002. Physical control of

spring–summer phytoplankton dynamics in the North water,April–July 1998. Deep-sea Research. Part 2. Topical Studies inOceanography 49, 4959–4982.

Melling, H., 1993. The formation of a haline shelf front in wintertime inan ice-covered arctic sea. Continental Shelf Research 13, 1123–1147.

Melling, H.,Moore, R.M., 1995.Modification of halocline sourcewatersduring freezing on the Beafourt Sea shelf: evidence from oxygenisotopes and dissolved nutrients. Continental Shelf Research 15 (1),89–113.

Menden-Deuer, S., Lessard, E.J., 2000. Carbon to volume relation-ships for dinoflagellates, diatoms, and other protist plankton.Limnology and Oceanography 45 (3), 569–579.

Mundy, C.J., Barber, D.G., Michel, C., 2005. Variability of snow andice thermal, physical and optical properties pertinent to sea icealgae biomass during spring. Journal of Marine Systems 58,107–120.

Mundy, C.J., Ehn, J.K., Barber, D.G., Michel, C., 2007. Influence ofsnow cover and algae on the spectral dependance of transmittedirradiance through Arctic landfast first-year sea ice. Journal ofGeophysical Research 112. doi:10.1029/2006JC003683.

Nagata, T., 1986. Carbon and nitrogen of natural planktonic bacteria.Applied and Environmental Microbiology 52 (1), 28–32.

Nielsen, T.G., Hansen, B., 1995. Plankton community structure andcarbon cycling on the western coast of Greenland during and afterthe sedimentation of a diatom bloom. Marine Ecology. ProgressSeries 125, 239–257.

Nusch, E.A., 1980. Comparison of different methods for chlorophylland phaeopigment determination. Archiv für Hydrobiologie,Ergebnisse der Limnologie 14, 14–36.

Opsahl, S., Benner, R., Amon, R.M., 1999. Major flux of terrigenousdissolved organic matter through the Arctic Ocean. Limnology andOceanography 44 (8), 2017–2023.

Pomeroy, L.R., Deibel, D., 1986. Temperature regulation of bacterialactivity during the spring bloom in Newfoundland coastal waters.Science 233 (4761), 359–361.

Porter, K.G., Feig, Y.S., 1980. The use of DAPI for identifying andcounting aquatic microflora. Limnology and Oceanography 25 (5),943–948.

Psenner, R., 1990. From image analysis to chemical analysis of bacteria:a long-term study? Limnology and Oceanography 35 (1), 234–237.

Riedel, A., Michel, C., Gosselin, M., 2006. Seasonal study of sea-iceexopolymeric substances on the Mackenzie shelf: implications fortransport of sea-ice bacteria and algae. Aquatic Microbial Ecology45, 195–206.

Rose, J.M., Caron, D.A., 2007. Does low temperature constrain thegrowth rates of heterotrophic protists? Evidence and implications foralgal blooms in cold waters. Limnology and Oceanography 52 (2),886–895.

Sañudo-Wilhelmy, S.A., Gobler, C.J., Okbamichael, M., Taylor, G.T.,2006. Regulation of phytoplankton dynamics by vitamin B12.Geophysical Research Letters 33. doi:10.1029/2005GL025046.

Scott, C.L., Kwasniewski, S., Falk-Petersen, S., Sargent, J.R., 2000.Lipids and life strategies of Calanus finmarchicus, Calanusglacialis and Calanus hyperboreus in late autumn, Kongsfjorden,Svalbard. Polar Biology 23, 510–516.

Sherr, E.B., Sherr, B.F., 2002. Significance of predation by protists inaquatic microbial food webs. Antonie Van Leeuwenhoek 81 (1–4),293–308.

Sherr, B.F., Sherr, E.B., 2003. Community respiration/production andbacterial activity in the upper water column of the central ArcticOcean. Deep-sea Research. Part 2. Topical Studies in Oceanography50, 529–542.

http://dx.doi.org/10.1111/j.15292006.310.x

http://dx.doi.org/10.1111/j.15292006.310.x

http://dx.doi.org/10.1029/2006JC003683

http://dx.doi.org/10.1029/2005GL025046


Sherr, E.B., Sherr, B.F., Fessenden, L., 1997. Heterotrophic protists inthe Central Arctic Ocean. Deep-sea Research. Part 2. TopicalStudies in Oceanography 44 (8), 1665–1682.

Sherr, E.B., Sherr, B.F., Wheeler, P.A., Thompson, K., 2003. Temporaland spatial variation in stocks of autotrophic and heterotrophicmiccrobes in the upper water column of the central Arctic Ocean.Deep-sea Research. Part 1. Oceanographic Research Papers 50,557–571.

Sieracki, M.E., Verity, P.G., Stoecker, D.K., 1993. Plankton comunityresponse to sequential silicate and nitrate depletion during the 1989North Atlantic spring bloom. Deep-sea Research. Part 2. TopicalStudies in Oceanography 40, 213–225.

Smith Jr., W.A., Sakshaug, E., 1990. Polar phytoplankton, polaroceanography. Part B: Chemistry, Biology, and Geology. Aca-demic Press, Inc., pp. 477–525.

Smith Jr., W.O., Nelson, D.M., Mathot, S., 1999. Phytoplanktongrowth rates in the Ross Sea, Antarctica, determined byindependent methods: temporal variations. Journal of PlanktonResearch 21 (8), 1519–1536.

Solomon, H., 1973. Wintertime surface layer convection in the ArcticOcean. Deep-Sea Research 20, 269–283.

Stevens, C.J., Deibel, D., Parrish, C.C., 2004. Species-specificdifferences in lipid composition and omnivory indices in Arcticcopepods collected in deep water during autumn (North WaterPolynya). Marine Biology 144, 905–915.

Suttle, C.A., 2005. Viruses in the sea. Nature 437, 356–361.Thibault, D., Head, E.J.H., Wheeler, P.A., 1999. Mesozooplankton in

the Arctic Ocean in summer. Deep-sea Research. Part 1.Oceanographic Research Papers 46, 1391–1415.

Vaqué, D., Guadayol, O., Peters, F., Felipe, J., Angel-Ripoll, J.,Terrado, R., Lovejoy, C., Pedrós-Alió, C., submitted for publica-tion. Seasonal changes in planktonic bacterivory rates under the icein Franklin Bay (Arctic-Canada). Limnology and Oceanography.

Vaqué, D., Guixa-Boixereu, N., Gasol, J.M., Pedrós-Alió, C., 2002.Distribution of microbial biomass and importance of protists inregulating prokaryotic assemblages in three areas close to theAntarctic Peninsula in spring and summer 1995/96. Deep-seaResearch. Part 2. Topical Studies in Oceanography 49, 847–867.

Vincent, W.F., Belzile, C., 2003. Biologicañ UVexposure in the polaroceans: Arctic–Antarctic comparisons. In: Huiskes, A.H.L., et al.(Ed.), Antarctic Biology in a Global Context. Backhuys Publish-ers, Leiden, pp. 176–181.

Watson, S.W., Novitsky, T.J., Quinby, H.L., Valois, F.W., 1977.Determination of bacterial number and biomass in the marineenvironment. Applied and Environmental Microbiology 33 (4),940–946.

Wheeler, P.A., Kirchman, D.L., Landry, M.R., Kokkinakis, S.A., 1989.Diel periodicity in ammonium uptake and regeneration in theoceanic subarctic Pacific: implications for interactions in microbialfood webs. Limnology and Oceanography 34 (6), 1025–1033.

Wheeler, P.A., Watkins, J.M., Hansing, R.L., 1997. Nutrients, organiccarbon and nitrogen in the upper water column of the Arctic Ocean:implications for the sources of dissolved organic carbon. Deep-seaResearch. Part 2. Topical Studies in Oceanography 44, 1571–1592.

Yang, E.Y., Choi, J.K., Hyun, J.-H., 2004. Distribution and structure ofheterotrophic protist communities in the northeast equatorialPacific Ocean. Marine Biology 146 (1), 1–15.

microbial food web responses to light and nutrients ... terrado08.pdf · microbial food web...

Documents