Zooplankton respiration and the export of carbon at depth in the Amundsen Gulf (Arctic Ocean)

Zooplankton respiration and the export of carbon at depthin the Amundsen Gulf (Arctic Ocean)

Gérald Darnis1 and Louis Fortier1

Received 9 June 2011; revised 6 February 2012; accepted 12 February 2012; published 6 April 2012.

[1] In arctic seas, lipids accumulated by zooplankton migrants in the surface layer inspring-summer are respired at depth during the winter. The resulting active downwardtransport of carbon by the 200–1000 and >1000 mm mesozooplankton fractions wasquantified based on 41 biomass and respiration profiles from October 2007 to July 2008 inthe Amundsen Gulf (Canadian Arctic Ocean). The small fraction, dominated by CII-CIIICalanus glacialis, represented on average 12% of the overall zooplankton biomass andcontributed little to the active transport of carbon by respiration. From April to July, totalzooplankton ingested 17–28% of the estimated gross primary production (GPP) in thesurface 100 m, and 36–59% of GPP over the entire water column. The large fraction,comprised mainly of CIV, CV and adults Calanus hyperboreus and C. glacialis thataccumulate large lipid reserves, was responsible for 89% of grazing. The downwardmigration of large zooplankton in late summer coincided with a sharp decline in specificrespiration rates signaling the start of diapause and the endogenous fuelling of metabolism.From October to April, Calanus migration-respiration actively transported 3.1 g C m�2

beyond 100 m, a flux that represented 85 to 132% of the gravitational POC fluxes at 100 mfrom October to July. Our results stress the importance of including active transport bylarge zooplankton migrants in carbon budgets of the Arctic Ocean.

Citation: Darnis, G., and L. Fortier (2012), Zooplankton respiration and the export of carbon at depth in the Amundsen Gulf(Arctic Ocean), J. Geophys. Res., 117, C04013, doi:10.1029/2011JC007374.

1. Introduction

[2] In the ocean, several processes mediated by the zoo-plankton can either enhance or limit vertical carbon fluxesand, therefore, regulate the efficiency of the biological CO2

pump [Hernández-León and Ikeda, 2005; Kobari et al.,2008; Steinberg et al., 2008; Tremblay et al., 2006]. In theepipelagic layer of arctic seas in spring-summer, the largeherbivorous copepods Calanus hyperboreus and C. glacialisexert heavy grazing pressure on the diatom-dominated phy-toplankton bloom [Tremblay et al., 2006]. Some of theassimilated photosynthetic carbon is immediately reminer-alized by respiration in the surface layer. But the energystored in the large lipid reserves of Calanus will be respiredlater at depth, several hundred meters below the surfacewhere the copepods overwinter for up to 10 months in aresting stage [Hirche, 1997].[3] In the temperate and tropical ocean, the diel vertical

migration (DVM) of the zooplankton is primarily responsi-ble for this active export of carbon linked to respiration[Longhurst and Williams, 1992; Hernández-León et al.,2001]. For instance, Hernández-León et al. [2001] calcu-lated that the DVM respiratory flux SW of Gran Canaria and

west of Tenerife represented 16–45% of the passive partic-ulate export flux. By contrast, Longhurst and Williams[1992] concluded that the active transport of carbon todepth by the seasonal migration of zooplankton (mainlyCalanus finmarchicus) was trivial relative to the gravita-tional flux of carbon. In arctic waters, the limited DVM ofzooplankton [Blachowiak-Samolyk et al., 2006; Daase et al.,2008] likely contributes little to the downward respiratoryflux. Previous studies suggest instead that the seasonalmigration of large herbivorous copepods such as Calanushyperboreus and C. glacialis could generate a significantdownward respiratory flux by exporting to depth the lipidsaccumulated in spring-summer [Auel et al., 2003; Hirche,1997].[4] Small omnivorous and detritivorous copepods that

numerically dominate arctic zooplankton are not known toundergo extensive ontogenetic vertical migrations [e.g.,Fortier et al., 2001]. Given their high turnover rates, con-tinuous activity, and capacity to graze small particles, smallcopepods may contribute substantially to recycling organiccarbon in the surface layer and to the attenuation of the ver-tical particulate organic carbon (POC) flux [Hopcroft et al.,2005]. As opportunistic feeders that accumulate little if anylipid reserves, small arctic zooplankton, in particular cope-pods, are believed to remain active under the ice in the wintermonths [Sampei et al., 2009]. For instance, the predominanceof small fecal pellets in the vertical POC flux in fall andwinter [Forest et al., 2008; Lalande et al., 2009] suggests

1Québec-Océan, Département de biologie, Université Laval, Québec,Québec, Canada.

Copyright 2012 by the American Geophysical Union.0148-0227/12/2011JC007374

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 117, C04013, doi:10.1029/2011JC007374, 2012

C04013 1 of 12

that, as large herbivorous copepods migrate to depth tooverwinter, these small copepods become, together withmicrozooplankton, the main grazers in the surface layer fromthe end of summer to the onset of microalgal production inspring/summer [Svensen et al., 2011].[5] The role of zooplankton grazing and respiration in

mediating the vertical transport of carbon remains poorlyresolved for arctic waters, particularly for the winter months.During the international Circumpolar Flaw Lead systemstudy (CFL), the research icebreaker CCGS Amundsenoverwintered in the southeastern Beaufort Sea, remainingmobile in the flaw lead that separates the drifting centralpack and the landfast ice. The 10-month expedition providedan unprecedented opportunity to monitor zooplankton bio-mass and respiration from October 2007 to July 2008 in theAmundsen Gulf. In this study, we track the vertical migra-tion of the 200–1000 mm and >1000 mm zooplankton sizefractions from October 2007 to July 2008. In particular, wemeasure the transport and remineralization at depth of car-bon that result from the annual vertical migration of the

mesozooplankton, and assess its importance relative to othervertical carbon fluxes.

2. Material and Methods

2.1. Study Area and Oceanographic Setting

[6] The Amundsen Gulf (60000 km2) connects the Beau-fort Sea and Mackenzie Shelf to the Canadian ArcticArchipelago (Figure 1). The western sector of the Gulf isbounded by Franklin and Darnley bays to the south andBanks Island to the north. The maximum depth of 600 moccurs in a small central basin between the southern tip ofBanks Island and Darnley Bay. Water masses in the regioncomprise the Polar-Mixed Layer (salinity < 31.6; 0–50 m),the Pacific Halocline (32.4 < S < 33.1; 50–200 m), and thewarmer Atlantic Waters (S > 34; >200 m) [Carmack andMacdonald, 2002]. The surface circulation is dominated bythe anticyclonic Beaufort Gyre that entrains the pack ice andsurface waters westward toward the Canada Basin [Carmack

Figure 1. Bathymetry of the southeastern Beaufort Sea with the location of zooplankton sampling sta-tions. Stations are numbered in chronological order of sampling. The stars indicate the location of themooring stations CA-08 (to the south) and CA-16 (to the north).

DARNIS AND FORTIER: ZOOPLANKTON RESPIRATION AND CARBON FLUX C04013C04013

2 of 12

and Macdonald, 2002; Ingram et al., 2008]. Large inter-annual variability occurs in this exchange between theBeaufort Sea and Amundsen Gulf [Barber et al., 2010].Below �80 m depth, the Beaufort shelf-break jet carrieswater of Pacific origin along the continental margin and intoAmundsen Gulf [Carmack and Macdonald, 2002; Pickart,2004]. The deep circulation (>200 m) is weak and highlyvariable [Ingram et al., 2008]. Sea-ice starts to form inOctober at the coastal boundaries of the gulf and consoli-dation is completed by December [Galley et al., 2008]. ByMay–June, breakup begins and the flaw lead polynyaenlarges to form the Cape Bathurst Polynya complex at theentrance of Amundsen Gulf. Satellite data indicate largeinter-annual variability in the extent and persistence of openwater [Arrigo and van Dijken, 2004].

2.2. Sampling

[7] Zooplankton sampling was conducted from theresearch icebreaker CCGS Amundsen between 24 October2007 and 30 July 2008 at 44 stations in the southeasternBeaufort Sea (Figure 1). The majority of stations (35) werelocated in western Amundsen Gulf, and the remaining siteswere sampled along the northwest coast of Banks Island, theMackenzie Shelf, Darnley Bay and Franklin Bay.[8] A SBE911 plus® CTD and a Seapoint® fluorometer

attached to the rosette sampler provided profiles of salin-ity, temperature and in situ fluorescence. The chlorophyll a(chl a) concentrations determined for water samples takenat discrete depths were used to calibrate the in situ fluo-rescence signal. Weekly ice coverage of Amundsen Gulffor 2007–2008 was extracted from the Canadian Ice Ser-vice archives using the IceGraph Version 1.03 application(http://ice-glaces.ec.gc.ca/IceGraph103/page1.jsf).[9] Depth-stratified samples for the assessment of zoo-

plankton biomass and respiration were obtained by deploy-ing a 0.50-m2 square aperture Hydrobios® multinet samplercarrying nine nets of 200 mm mesh size each fitted with a 2-L rigid codend with filtration apertures located at the top ofthe cylinder. The sampler was hauled vertically at 0.5 m s�1

from 10 m above the seafloor to the surface. Three 20-mdepth strata were sampled from 10 m above the bottomupward and three from 60 m to the surface. The remaininginterval from 70 m above the bottom to 60 m below thesurface was divided in three equal sampling strata, thethickness of which depended on station depth. For example,the nine depth strata sampled in succession from 10 m abovebottom to the surface at a 430 m deep station were: 420–400,400–380, 380–360, 360–260, 260–160, 160–60, 60–40, 40–20, 20–0 m.[10] Upon retrieval, macrozooplankton (e.g., large amphi-

pods, euphausiids, medusae and fish larvae) were removedfrom the samples, as they are not sampled quantitatively bythe Hydrobios. The samples were subdivided in quartersusing a Motoda splitting box in a cold room (0�C). Twoquarters from each codend were allocated to respiration andbiomass measurements respectively. The remaining half waspreserved in a borax-buffered seawater solution of 4%formaldehyde for later taxonomic analysis. The subsamplewas diluted in filtered seawater at 0�C and kept at 0�C untilthe measurement of respiration based on enzymatic assay.[11] To calibrate the enzymatic assays, additional live

zooplankton was collected for the direct measurements of

respiration using a polarographic electrode at 17 stations inthe Amundsen Gulf from late February to late July 2008. A1-m2 square aperture vertical sampler with a 200-mm meshnet and a 2-L rigid codend was towed vertically from 10 mabove the bottom to the surface at 0.5 m s�1. The samplerwas lowered codend-first to avoid filtering on the way down.Upon recovery, the content of the codend was diluted infiltered seawater and cnidarians were removed to reducezooplankton mortality. The sample was kept at 0�C untilrespiration was measured both by the polarographic elec-trode method and the ETS assay (see below).

2.3. Respiration

[12] The enzymatic activity of the Electron-Transfer-Sys-tem (ETS), an index of respiration, was determined using theETS assay method developed by Båmstedt [2000]. Foreach assay, live zooplankton samples from the nine depthlayers were fractionated into 200–1000 and >1000 mmsize classes, and then homogenized directly with an INT(p-iodonitrotetrazolium violet) reagent following preciselythe protocol of Båmstedt [2000]. The homogenates wereincubated for one hour at 40�C after which the reaction wasstopped with a quench solution (50% formaldehyde and50% phosphoric acid). A blank of INT reagent withoutbiological material received the same treatment. After adding1 mL of chloroform/methanol (2:1 by volume), the samplewas mixed before centrifugation at 3000 rpm for four minutes.The lower phase was completed to 3 mLwith methanol beforea second centrifugation. The reaction color was measured at475 nm against the blank. Standards for the transformation ofmeasured ETS activity into respiration (mg O2 h�1) wereprepared according to Båmstedt [2000] for each batch ofreagents used over the quasi-annual duration of the study.[13] The relatively quick and easy ETS assay enabled us to

estimate relative respiration on board for the 362 samplesmaking up the 41 vertical profiles. To validate the ETSmeasures and transform enzymatic activity into actual res-piration, live zooplankton was incubated for the directmeasure of respiration using a polarographic electrode. At17 stations from late February to late July, a fraction of thelive sample was poured into a funnel equipped with a1000 mm sieve and a gate valve. The volume of water abovethe sieve was maintained constant and at maximum level byadding filtered seawater at 0�C. The large size class wasretained in this volume of water in the top part of the devicewhile the small size class was slowly evacuated through thesieve with the filtered seawater, recovered delicately in acontainer and stored at 0�C prior to incubation. Each of thetwo size fractions was gently split into three equal triplicatesand, after verifying visually that the animals were in goodcondition, each triplicate was introduced into a separateglass bottle (473 mL capacity) that was then filled to thebrim with filtered seawater at 0�C and capped. Simulta-neously, control bottles without zooplankton were preparedin triplicates. The triplicates for the two size classes and thecontrol were incubated for 24–48 h in the dark at 0�C. Dis-solved oxygen concentration in the control bottles wasmeasured with a polarographic electrode before and afterincubation to verify for possible changes. At the end of theexperiment, the difference in oxygen concentration betweenanimal-free control bottles and incubation bottles wasassumed to be the zooplankton respiration, divided by the

DARNIS AND FORTIER: ZOOPLANKTON RESPIRATION AND CARBON FLUX C04013C04013

3 of 12

duration of the incubation to obtain the hourly (ml O2 h�1)

or daily (ml O2 d�1) oxygen consumption RO. To avoid apotential effect of oxygen depletion on respiration, only thetriplicate batches where average saturation exceeded 70% atthe end of the incubation were retained. At the end of theincubation, ETS activity was determined in each tripli-cate, following the assay described above. The equationof the regression of RO on ETS for the 200–1000 mmand >1000 mm mesozooplankton fractions was used to con-vert ETS profiles into daily oxygen consumption profiles.[14] Daily oxygen consumption RO (ml O2 d

�1) was con-verted to daily respiratory carbon loss RC (mg C d�1) using:

Rc ¼ RO � RQ� 12=22:4; ð1Þ

where RQ (respiratory quotient) is the molar ratio of carbonproduced to oxygen consumed, 12 is the molecular weight ofcarbon and 22.4 is the molar volume of carbon dioxide atstandard temperature and pressure [Ikeda et al., 2000].Assuming a metabolism primarily based on proteins, a RQ of0.97 [Gnaiger, 1983] was used at all times for the smallfraction and during the season of high pelagic primary pro-duction for the large fraction. A RQ of 0.75 was applied tothe large zooplankton during the overwintering period whenmetabolism is fuelled predominantly by lipid reserves[Ingvarsdóttir et al., 1999].

2.4. Biomass and Taxonomy

[15] At sea, the aliquot for total zooplankton biomass wassize fractionated on a 1000-mm sieve and rinsed thoroughlywith filtered seawater. The 200–1000 and >1000 mm size

classes were placed on Nitex screens and carefully blottedwith absorbing material before freezing at �20�C. On land,the samples were transferred to pre-weighed plastic cups andoven-dried at 60�C for 48 h prior to dry weight measurementon a high precision (�1 mg) microbalance. For a given sta-tion, the weighted mean depth of the zooplankton biomasswas calculated as S(widizi)/S(dizi), where wi is the dryweight of zooplankton within depth interval i, di is thethickness of depth interval i, and zi is the midpoint depth ofthe interval.[16] In the laboratory, the formalin-preserved subsamples

for taxonomy were size fractionated on a 1000-mm sieve andre-suspended in distilled water. Known aliquots were takenfrom the small and large fractions using a Henson-Stempelpipette and a Motoda splitting box, respectively, to obtainapproximately 300 animals that were counted, measured andidentified to species or to the lowest taxonomic resolutionattainable. Carbon content for each species was estimatedusing published length-mass relationships for arctic zoo-plankton following Hopcroft et al. [2005], Mumm [1991],and Forest et al. [2011a]. Zooplankton community structureand the vertical distribution and population dynamics of keyspecies are the focus of other work in preparation. Only thegeneral taxonomy and biomass of the two size fractions ofzooplankton are presented here.

3. Results

3.1. Sea Ice, Temperature and Chlorophyll

[17] Sea-ice extent in SE Beaufort Sea in September 2007was the lowest on record since 1979, and the study area wasessentially ice-free at the end of summer. Starting in earlyNovember, young and new ice formed rapidly, filling mostof the eastern Amundsen Gulf. Consolidation into >80%first-year ice was completed by December (Figure 2a). Theice cover remained dynamic throughout winter and iceconcentration seldom reached 100% while substantialamounts (up to 28%) of new ice formed in winter flaw leads.Sea ice started to break-up in early May, about one monthearlier than average [Barber et al., 2010], followed by adecline in ice cover extent from 90% in early May to lessthan 5% in early July (Figure 2a).[18] The three water masses typical of the highly stratified

Beaufort Sea were detected in the temperature profiles of allstations along the track of the ship except the shallowest(Figure 2b). Atmospheric forcing strongly influenced thePolar-Mixed Layer (PML) that extended from the surface to�50 m depth. At the onset of ice formation in October,surface temperature was close to freezing point andremained so until ice break-up in May. Then the PMLwarmed rapidly to reach temperatures above 8�C in the 0–20 m layer in July. The steep Pacific Halocline (PH)extended from the PML to about 200 m with an inversetemperature gradient from �1.6 to 0�C at 200 m. Thewarmer Atlantic Waters with temperatures >0�C extendedfrom 200 m to the bottom.[19] In late October, the depth of the chlorophyll maxi-

mum (ZCM) was located around 25 m, with concentrationsup to 1 mg chl a L�1 (Figure 2c). As the ice cover formed inNovember, fluorescence declined below the detection limitsof the probe before increasing again in late April whenconcentrations above 0.1 mg chl a L�1 were measured in the

Figure 2. (a) Time series of ice concentration in AmundsenGulf and (b) corresponding time-depth sections of tempera-ture and (c) chl a concentration along the track of the shipfrom October 2007 to August 2008.

DARNIS AND FORTIER: ZOOPLANKTON RESPIRATION AND CARBON FLUX C04013C04013

4 of 12

top meters of the water column, presumably coinciding withthe sloughing off of ice algae as ice melt started. Microalgalbiomass increased in May (9.8 mg chl a L�1 peak at a ZCM of15 m) as the ice cover rapidly receded. After this first bloom,fluorescence values decreased to <1 mg chl a L�1 for mostof June then bounced back in July, forming a subsurface chla maximum (SCM) at about 55 m depth (Figure 2c).

3.2. Zooplankton Composition, Biomass, and DielVertical Migration

[20] The composition of the mesozooplankton was typicalof Arctic waters with a strong dominance of copepods(Table 1). On average over the year, the small fraction (200–1000 mm) was 7.4 times more numerous than the largefraction (>1000 mm), but its biomass was 7.1 times less.Calanus glacialis copepodites CIII and young Metridialonga (CI-CIV) represented 48 and 30% respectively of thebiomass in the small fraction. Calanus hyperboreus was themain contributor (51%) to biomass in the large fraction,followed by C. glacialis (19%) and Metridia longa (13%).[21] The comparison of day and night profiles of biomass

in different seasons failed to show any extensive diel verticalmigration by either the small or the large fraction of zoo-plankton (Figure 3). In all seasons, the biomass of the smallfraction was distributed relatively uniformly over the watercolumn and the position of the weighted mean depthchanged little between day and night. The biomass of thelarge zooplankton fraction was concentrated at depth in allseasons except summer (June) and vertical distributionchanged little between day and night. The small differences

in weighted mean depth between daytime and nighttimeprofiles reflected differences in station depth rather than anydiel migration (Figure 3).

3.3. ETS Activity and Respiration

[22] Ro, the respiration measured directly by the polar-ographic electrode method was linearly correlated to ETS,the enzymatic activity of the Electron-Transfer-System(Figure 4). The slope of the regression did not differ sig-nificantly between the two size classes (t = 0.44, P =0.666). However, the intercept was significantly higher

Figure 3. Nighttime (closed histograms) and daytime(open histograms) vertical distribution of zooplankton bio-mass for the (left) 200–1000 mm and (right) >1000 mm frac-tions at selected stations in Amundsen Gulf. Open circlescorrespond to the weighted mean depth of the biomassdistribution.

Table 1. Abundance and Biomass of the Main ZooplanktonTaxa Integrated Over the Water Column and Averaged Overthe Period October 2007 to July 2008 in Amundsen Gulfa

Abundance(Number m�2)

Biomass(g C m�2)

200–1000 mm FractionCalanus glacialis (CI-CIII) 7 676 � 6338 0.280 � 0.214Metridia longa (CI-CIV) 11 160 � 8445 0.174 � 0.148Pseudocalanus spp. (CI-CV) 10 772 � 6974 0.025 � 0.015Oithona similis (CIV-F) 35 314 � 18904 0.022 � 0.012Triconia borealis (CV-F) 18 772 � 11394 0.022 � 0.014Microcalanus spp. (CIII-F) 9 451 � 4429 0.012 � 0.006Othersb 9 013 � 5384 0.044 � 0.054Total 102 158 � 45705 0.579 � 0.340

>1000 mm FractionCalanus hyperboreus (CIII-F) 1 564 � 869 2.104 � 1.231Calanus glacialis (CIV-F) 4 898 � 3100 0.785 � 0.500Metridia longa (CV-F) 5 033 � 4623 0.531 � 0.491Carnivoresc 1 164 � 1058 0.537 � 0.551Mesopelagic and deep copepodsd 593 � 526 0.071 � 0.068Otherse 641 � 725 0.084 � 0.092Total 13 893 � 8167 4.112 � 2.171

aMeans and standard deviations are given.bOthers (small fraction) include primarily Oncaea sp., Bivalves,

Limacina helicina, Clione limacina, Scolecithricella minor by decreasingorder of abundance.

cCarnivores include primarily Paraeuchaeta glacialis, Eukronhiahamata, Dimophyes arctica, Parasagitta elegans, Aglantha digitale bydecreasing order of abundance.

dMesopelagic and deep copepods include primarily Aetideopsis rostrata,Gaetanus tenuispinus, Gaidius spp. by decreasing order of abundance.

eOthers (large fraction) include primarily Boroecia maxima, Limacinahelicina by decreasing order of abundance.

DARNIS AND FORTIER: ZOOPLANKTON RESPIRATION AND CARBON FLUX C04013C04013

5 of 12

(t = �2.27, P = 0.031) in the small fraction than in the large(Figure 4). Phytoplankton, in particular chain-forming dia-toms, was retained on the 200-mm sieve, but not on the1000-mm sieve. As well, the ratio R/ETS was weakly cor-related to chl a concentration in the 0–60 m depth strata forthe 200–1000 mm zooplankton fraction (r2 = 0.32, n = 14,P = 0.036), but not for the >1000 mm fraction. Hence, thehigher intercept of the regression of Ro on ETS in the smallfraction likely reflected a slight overestimation of Ro due tophytoplankton respiration during incubation in the dark.Given the difference in intercept, the data for the two frac-tions were not pooled and ETS was transformed into respi-ration using the separate regression for each fraction.

3.4. Seasonal Variability in Specific Respiration

[23] On average over the quasi-annual cycle, the produc-tion of respiratory carbon per unit biomass was higher in thesmall zooplankton fraction (14.4 � 5.7 mg C mg DW�1 d�1)than in the large fraction (8.4 � 4 mg C mg DW�1 d�1)(Figure 5). Specific respiratory carbon loss was variable insmall zooplankton with values generally below the meanfrom January to May, except for a peak in early April(Figure 5a). The seasonal pattern was clearer in large zoo-plankton, with specific carbon production remaining lowfrom October to mid-March (5.2� 0.9 mg C mg DW�1 d�1),and then nearly doubling from mid-March to mid-July(11.3 � 3.2 mg C mg DW�1 d�1), before reverting in lateJuly to low values comparable to the fall-winter rates(Figure 5b).

3.5. Spatiotemporal Distribution of ZooplanktonBiomass and Community Respiration

[24] As expected, zooplankton community respiration (Rc,as C loss) was strongly related to biomass (B) for the smallfraction (lnRc = �2.388 + 0.563 lnB, n = 317, r2 = 0.671,P < 0.0001) and the large fraction (lnRc = �4.370 + 0.909

lnB, n = 328, r2 = 0.728, P < 0.0001). Spatially, thedepth-integrated community respiration of the small frac-tion varied little across the study area (Figure 6a). Varia-tions were weakly but significantly correlated to depth (r2 =0.16, P = 0.0097). The much higher community respiration ofthe large fraction was also correlated to depth (r2 = 0.27, P =0.0006) and reached maximum values in the deep centralchannel of Amundsen Gulf (Figure 6b).[25] Temporally, the depth-integrated community respira-

tion Rc was minimum from December to mid-January for thetwo size fractions of zooplankton (Figure 7). In the smallfraction, Rc increased rapidly from late January to Marchand then declined progressively until the end of summer(Figure 7a). High values occurred in October. In the largefraction, depth-integrated community respiration increasedirregularly from late January to July (Figure 7b). Relativelyhigh values were also recorded in October.[26] Temporally and vertically, the distribution of com-

munity respiration was closely related to the distribution ofbiomass (Figure 8 and 9). From December to March, thesmall biomass of the small zooplankton fraction (200–1000 mm) tended to increase and was distributed relativelyuniformly over depth (Figure 8a). As a result, respirationincreased uniformly in the water column over the wintermonths (Figure 8b). From mid-April to early November, thesmall zooplankton fraction invaded the 0–50 m surface layer(Figure 8a), and community respiration for this size classtook place primarily near the surface (Figure 8b).[27] Large zooplankton were concentrated in the deep

basins of Amundsen Gulf from December to late April, thencongregated in the surface layer (0–50 m) from early May toearly July, and started their downward migration in July(Figure 9a). By October, the large zooplankton fraction was

Figure 4. Regression of respiration (Ro, measured bypolarographic electrode) on Electron-Transfer-Systemactivity (ETS) for the 200–1000 mm (open symbols) and>1000 mm fraction (closed symbols) of zooplankton. Eachdata point is the average of triplicates with horizontal andvertical range bars corresponding to one standard deviationfor the two variables respectively.

Figure 5. Time series of daily specific respiratory carbonloss for the (a) small and (b) large fractions of zooplanktoncollected in the southeastern Beaufort Sea from October2007 to July 2008. Horizontal lines correspond to the meanover the 10-month time series. The open symbols in Figure 5bindicate values at stations outside the Amundsen Gulf.

DARNIS AND FORTIER: ZOOPLANKTON RESPIRATION AND CARBON FLUX C04013C04013

6 of 12

Figure 6. Spatial distribution of the vertically integrated community respiration (Rc, as respiratory car-bon loss) for the (a) 200–1000 mm and (b) >1000 mm zooplankton fractions in the southeastern BeaufortSea from 24 October 2007 to 30 July 2008.

DARNIS AND FORTIER: ZOOPLANKTON RESPIRATION AND CARBON FLUX C04013C04013

7 of 12

again concentrated at depth. The time-depth distribution ofcommunity respiration Rc followed essentially the sameseasonal pattern as biomass (Figure 9b).

4. Discussion

4.1. Zooplankton Community Respiration in Polar Seas

[28] Measures of zooplankton community respiration arefew for the Arctic Ocean. Welch et al. [1997] used largerespiration chambers to incubate non-fractionated live zoo-plankton assemblages >202 mm in repeated experimentsover an annual cycle in shallow Resolute Passage (Cana-dian Arctic Archipelago). Specific respiratory carbon lossvaried from 5.5 to 23.4 mg C mg DW�1 d�1 over theyear, a range commensurate with that reported here for the200–1000 mm (5.9–30.3 mg C mg DW�1 d�1) and>1000 mm (4.1–17.1 mg C mg DW�1 d�1) zooplanktonfractions in Amundsen Gulf. Our estimates of specific res-piration also fall within the range 4–80 mg C mg DW�1 d�1

measured by Alcaraz et al. [2010] for the whole zoo-plankton assemblage collected along a transect from FramStrait to Svalbard in July 2007. In the Southern Ocean,mesozooplankton size fractions of 200–500, 500–1000and >1000 mm off the Antarctic Peninsula respired ataverage rates of 12.6 � 6.0, 26.2 � 20.1 and 24.3 �15.5 mg C mg DW�1 d�1 respectively in December

Figure 7. Time series of depth-integrated daily respiratorycarbon loss (Rc) for the (a) small and (b) large fractions ofzooplankton collected in the southeastern Beaufort Sea fromOctober 2007 to July 2008. The open symbols indicate sta-tions outside the Amundsen Gulf.

Figure 8. (a) Time-depth section of biomass and (b) community respiration as carbon loss for the200–1000 mm zooplankton size fraction in the southeastern Beaufort Sea from October 2007 to July2008. Vertical lines correspond to the date of biomass and respiration profiles. White surfaces indicatemissing data. Numbers on top refer to stations in Figure 1.

DARNIS AND FORTIER: ZOOPLANKTON RESPIRATION AND CARBON FLUX C04013C04013

8 of 12

(southern hemisphere summer) [Hernández-León et al.,1999]. These values compare well with rates of 18.1 �5.7 and 12.7 � 2.8 mg C mg DW�1 d�1 for the 200–1000 and >1000 mm zooplankton fractions respectivelywhen averaged over the corresponding summer period in theAmundsen Gulf. Based on the tight regression between ETSactivity and direct respiration measurements, and the goodagreement between our estimates of specific respiration andthose reported for other polar seas, we conclude that ourrespiration estimates based on this new ETS enzymaticapproach are reliable. As expected, biomass was a goodpredictor of zooplankton community respiration. Hence, therespiration-biomass linear relationships presented here rep-resent a potentially useful tool to estimate zooplankton res-piration from dry weight.

4.2. Zooplankton Control of Pelagic PrimaryProduction

[29] In the Beaufort Sea, relatively large diatoms dominateboth the ice algal production in spring and the spring-sum-mer phytoplankton bloom [Forest et al., 2008; Riedel et al.,2008]. To quantify the impact of zooplankton grazing onprimary production, we first translated respiration (R) into arough estimate of ingestion (I), assuming an assimilation anda gross growth efficiency of 70% and 30% respectively: I =100R/(70–30) = 2.5R [Ikeda and Motoda, 1978]. Estimated

ingestion confirms that the large calanoid herbivores makingup the bulk of the >1000 mm fraction were responsible formost (89%) of total zooplankton ingestion in the top 100 min spring and summer. Large calanoid herbivores ascendedfrom their deep overwintering habitat in late April, a monthahead of the ice break-up (defined as 50% ice cover) in lateMay. In the interval between the vernal recolonization ofthe surface layer and the ice break-up that triggered thephytoplankton bloom, the mesozooplankton presumablyfuel carbon demand from several sources including the lowconcentration of ice algae sloughed from the melting sea ice,a precocious under-ice phytoplankton production [Forest etal., 2011b], and microzooplankton [Seuthe et al., 2007].[30] Based on nutrient drawdown from February to July

2008, Forest et al. [2011b] estimated gross primary pro-duction (GPP) at 40–65 g C m�2 in the euphotic layer ofareas deeper than 250 m in the Amundsen Gulf. Assuming24-h grazing in the 0–100 m layer of the same sector, zoo-plankton consumption between May and July accounted onaverage for 11.3 � 4.0 g C m�2 or 17–28% of GPP. Thisestimate of the fraction of GPP intercepted by zooplanktonin the euphotic zone of Amundsen Gulf in spring-summeroverlaps the 22–44% derived from fecal pellet carbon flux innorthern Barents Sea [Wexels Riser et al., 2008]. By com-parison, in Baffin Bay, copepods and appendiculariansgrazed an estimated 52% of GPP between April and July

Figure 9. (a) Time-depth section of biomass and (b) community respiration as carbon loss for the>1000 mm zooplankton size fraction in the southeastern Beaufort Sea from October 2007 to July 2008.Vertical lines correspond to the date of biomass and respiration profiles. White surfaces indicate missingdata. Numbers on top refer to stations in Figure 1.

DARNIS AND FORTIER: ZOOPLANKTON RESPIRATION AND CARBON FLUX C04013C04013

9 of 12

1998 in the surface layer of the highly productive easternNorth Water [Tremblay et al., 2006], while the Calanuscomplex (C. hyperboreus, C. glacialis, and C. finmarchicus)intercepted between 13 and 93% of primary productionduring the phytoplankton bloom from May to early June inDisko Bay [Madsen et al., 2001].[31] By contrast to the spring-summer period, algal cells

<5 mm typically dominate the subsurface chlorophyllmaximum in the southeastern Beaufort Sea in late fall[Brugel et al., 2009; Martin et al., 2010; Tremblay et al.,2009]. In October 2007, picophytoplankton <5 mm madeup 75% of the microalgal assemblage (M. Ardyna, unpub-lished data, 2011). A dominance of small phytoplankton cellsfavor grazing by microzooplankton and small mesozoo-plankton to the detriment of large calanoid herbivores thatspecialize on large cells [Campbell et al., 2009]. Based onrespiration, the 200–1000 mm zooplankton fraction wasresponsible for 64% of the POC grazed by the overallmesozooplankton in the surface 100 m in the fall. Primaryproduction was not measured in the fall of 2007, but Brugelet al. [2009] provide an estimate of 20–40 mg C m�2 d�1

for late October–early November 2003 in Amundsen Gulf.Assuming a similar primary production in the two years, adaily GPP of 20–40 mg C m�2 d�1 would have satisfied thecarbon demand of the 200–1000 mm zooplankton fraction(20.2 � 5.2 mg C m�2 d�1, or 51–101% of daily GPP) in2007, but could not sustain the metabolic carbon demand(105 � 51 mg C m�2 d�1) of the abundant assemblage oflarge calanoid copepods that had already left the surface layerand migrated to depth. Small pellets largely dominate thevertical fecal POC flux in fall and early winter in the BeaufortSea [Forest et al., 2007; Forest et al., 2008], an observationthat confirms that small mesozooplankton are the main gra-zers of phytoplankton biomass during this period.[32] Total mesozooplankton ingestion (200–1000 +

>1000 mm fractions) integrated over the entire water columnamounted to 23.5 � 7.5 g C m�2 or 36–59% of theestimated 40–65 g C m�2 GPP integrated over spring-summer. For comparison, total ingestion by micro-zooplankton (5–200 mm) over the same period wasestimated at 15.6 g C m�2 or approximately 30% of GPP byForest et al. [2011b]. Therefore, based on respiration, weconclude that the large (>1000 mm) Calanus herbivores andMetridia longa active in the surface layer in spring andsummer are the major consumers of primary production overthe annual cycle of the pelagic food web of Amundsen Gulf,as reported for other arctic regions [Rysgaard et al., 1999;Tremblay et al., 2006]. In the fall, the 200–1000 mm meso-zooplankton fraction, made up at 42% of juvenile CIII of C.glacialis, was primarily responsible for the grazing of theweak primary production in the surface layer.

4.3. Active Carbon Export by Calanus

[33] Not all respiration at depth by zooplankton actuallycontributes to the active transport of carbon from the surfacelayer to depth. For instance, respiration at depth by non-migrant omnivores feeding on sinking particles or by non-migrant carnivores does not accelerate the export to depth ofthe POC produced in the surface layer. In offshore arcticseas, large ontogenetic migrants such as late copepoditestages and adult Calanus that build up large lipid reservesare presumably responsible for most of the active export of

carbon to depth [Auel et al., 2003; Hirche, 1997]. Hence, thecombined respiration at depth of C. hyperboreus and C.glacialis >1000 mm is likely a conservative estimate ofactive carbon transport. Given the tight correlation betweenrespiration and biomass for the >1000 mm fraction, respi-ration by Calanus below 100 m can be calculated by pro-rating overall zooplankton respiration by the biomass ofCalanus. This approach first assumes that Calanus do notfeed at depth in winter. Stable isotope analyses and grazingexperiments have confirmed the true herbivorous nature ofC. hyperboreus [Campbell et al., 2009; Forest et al., 2011a],and winter feeding at depth is unlikely for this species.C. glacialis sometimes preys on heterotrophs [e.g., Campbellet al., 2009], but the low specific respiration rate of largezooplankton at depth from October to mid-March supportsthe notion that basal metabolism and eggmaturation in winterwere fuelled essentially by endogenous reserves. Pro-ratingrespiration by biomass also assumes that specific respirationrates of Calanus are similar to those of other large zoo-plankton. From February to early April 2004, the specificrespiration of C. hyperboreus and C. glacialis under thelandfast ice of Franklin Bay was 4.1 � 0.04 and 6.2 �0.9 mg C mg DW�1 d�1, respectively (G. Darnis, unpub-lished results based on the separate incubation of the twospecies, 2004). These values are close to the mean specificrespiration rate of the large size class in early winter 2007–2008 (5.2 � 0.9 mg C mg DW�1 d�1), which supports ourassumption that the two Calanus species respire at ratessimilar to those of other large zooplankton in winter.[34] Based on the above, mean active vertical carbon

transport by Calanus was estimated at 3.1 � 2.1 and 2.5 �1.7 g C m�2 below 100 m and 200 m respectively for theperiod 24 October–17 April in regions of Amundsen Gulfdeeper than 250 m. Active transport of carbon beyond 100 mrepresented approximately 85% of the POC passive flux at100 m recorded from 24 October to 28 July at Station CA-08(M. Sampei, unpublished data, 2010) and 132% of the pas-sive flux for the same period at Station CA-16 [Sampei et al.,2012]. Our estimate of active transport of carbon beyond100 m by Calanus is close to the lower limit of the range ofannual gravitational POC fluxes (3.3–6.0 g C m�2 yr�1) atStation CA-08 for 2004, 2005 and 2006 [Forest et al.,2010]. No POC flux was recorded at 200 m in 2007–2008. However, our estimate of active carbon transportbeyond this depth falls in the range of annual gravitationalPOC fluxes recorded at 200 m in the deepest region ofAmundsen Gulf (1.9–3.8 g C m�2 yr�1) and at StationCA-08 (1.3–2.3 g C m�2 yr�1) for 2004, 2005 and 2006[Lalande et al., 2009]. Overall, the active downwardtransport of carbon linked to the seasonal migration ofCalanus species in Amundsen Gulf was of the samemagnitude as the annual passive POC flux, which stressesthe importance of the seasonal migration of arctic zoo-plankton in the downward transport of organic carbon.

5. Conclusion

[35] Longhurst and Williams [1992] calculated that, at thescale of the North Atlantic, the active transport of carbon bythe ontogenetic migration of C. finmarchicus was only 0.1%of the gravitational flux at 200 m, and concluded that thistransport could therefore be neglected in carbon budgets. At

DARNIS AND FORTIER: ZOOPLANKTON RESPIRATION AND CARBON FLUX C04013C04013

10 of 12

odds with their results for the North Atlantic, our analysis ofthe time-depth distribution of size-fractionated mesozoo-plankton respiration over an annual cycle in Amundsen Gulfconfirms the importance of the seasonal ontogenetic migra-tions of Calanus in actively transporting organic carbonfrom the surface layer to depth in Arctic seas [Auel et al.,2003; Hirche, 1997]. Conservatively, active carbon trans-port by the seasonal migration of zooplankton was of thesame magnitude as the gravitational flux of POC. Thereforeit should be included in future synthesis of carbon fluxes inthe Arctic Ocean. The divergence between the conclusionsof Longhurst and Williams [1992] and ours certainly stemsin part from (1) the large gravitational fluxes prevailing inthe Atlantic Ocean (10–58 g C m�2 yr�1 at 150 m [Morales,1999]) relative to the Arctic Ocean (1.6–5.9 g C m�2 yr�1 at200 m [Lalande et al., 2009]), and (2) the large body sizeand lipid reserves of C. hyperboreus and C. glacialis com-pared to C. finmarchicus [e.g., Scott et al., 2000]. We sus-pect however that the difference in the importance of therespiratory carbon fluxes generated by the Calanus complexin the two regions is ultimately linked to the different lightregimes prevailing at high and temperate latitudes. Byfeeding near the surface at night and descending at depth indaytime, the zooplankton of temperate and tropical oceansactively export some carbon at depth on a daily basis[Longhurst and Williams, 1992; Hernández-León et al.,2001]. In Arctic seas, the diel cycle of night and day doesnot occur outside brief periods around the equinoxes. It isreplaced by the annual cycle of polar night and midnightsun, which is responsible for the extreme seasonality ofarctic primary production. Obviously, seasonal ontogeneticmigrations and the remarkable capacity of arctic zooplank-ton (in particular Calanus hyperboreus) to delay the remi-neralization of organic matter by storing it into large lipidreserves are adaptations forced by the pulsed nature of arcticmicroalgal production [e.g., Falk-Petersen et al., 1990]. Theobserved strong downward flux of carbon is a by-product ofthe combination of widespread annual migrations andexceptionally large lipid reserves in arctic zooplankton. Weconclude that the importance of the active transport of car-bon by zooplankton relative to the gravitational flux reportedhere for the Amundsen Gulf is ultimately rooted in theannual photoperiod (as opposed to diel) prevailing in theArctic.

[36] Acknowledgments. We thank the officers and crews of theCCGS Amundsen for their support at sea. This work would not have beenpossible without the dedication of several colleagues and technicians whocontributed to the sampling and laboratory analyses during the 10-monthCFL expedition. Special thanks to M. Sampei and M. Ardyna for long-termsediment trap data and information on phytoplankton size classes, respec-tively. J. Martin provided the calibration factors of the fluorometer probeand Y. Gratton and his team processed the CTD data. We thank two anon-ymous reviewers for insightful comments that helped improve the submit-ted manuscript. The CFL program and this study were funded by theCanadian International Polar Year (IPY) program office, the NaturalSciences and Engineering Research Council of Canada (NSERC), theCanada Research Chairs (CRC) Program, the Network of Centres of Excel-lence ArcticNet, and the Canada Foundation for Innovation (CFI). This is acontribution to Québec-Océan at Université Laval and the Canada ResearchChair on the response of marine arctic ecosystems to climate warming.

ReferencesAlcaraz, M., R. Almeda, A. Calbet, E. Saiz, C. M. Duarte, S. Lasternas,S. Agusti, R. Santiago, J. Movilla, and A. Alonso (2010), The role of

arctic zooplankton in biogeochemical cycles: Respiration and excretion ofammonia and phosphate during summer, Polar Biol., 33(12), 1719–1731,doi:10.1007/s00300-010-0789-9.

Arrigo, K. R., and G. L. van Dijken (2004), Annual cycles of sea ice andphytoplankton in Cape Bathurst polynya, southeastern Beaufort Sea,Canadian Arctic, Geophys. Res. Lett., 31, L08304, doi:10.1029/2003GL018978.

Auel, H., M. Klages, and I. Werner (2003), Respiration and lipid content ofthe Arctic copepod Calanus hyperboreus overwintering 1 m above theseafloor at 2,300 m water depth in the Fram Strait, Mar. Biol. Berlin,143(2), 275–282, doi:10.1007/s00227-003-1061-4.

Båmstedt, U. (2000), A new method to estimate respiration rate of biologi-cal material based on the reduction of tetrazolium violet, J. Exp. Mar.Biol. Ecol., 251(2), 239–263, doi:10.1016/S0022-0981(00)00217-3.

Barber, D. G., M. G. Asplin, Y. Gratton, J. Lukovich, R. J. Galley, R. L.Raddatz, and D. Leitch (2010), The International Polar Year (IPY) Cir-cumpolar Flaw Lead (CFL) system study: Overview and the physical sys-tem, Atmos. Ocean, 48(4), 225–243.

Blachowiak-Samolyk, K., S. Kwasniewski, K. Richardson, K. Dmoch,E. Hansen, H. Hop, S. Falk-Petersen, and L. Thybo Mouritsen (2006),Arctic zooplankton do not perform diel vertical migration (DVM) duringperiods of midnight sun, Mar. Ecol. Prog. Ser., 308, 101–116,doi:10.3354/meps308101.

Brugel, S., C. Nozais, M. Poulin, J. E. Tremblay, L. A.Miller, K. G. Simpson,Y. Gratton, and S. Demers (2009), Phytoplankton biomass and produc-tion in the southeastern Beaufort Sea in autumn 2002 and 2003, Mar.Ecol. Prog. Ser., 377, 63–77, doi:10.3354/meps07808.

Campbell, R. G., E. B. Sherr, C. J. Ashjian, S. Plourde, B. F. Sherr, V. Hill,and D. A. Stockwell (2009), Mesozooplankton prey preference and grazingimpact in the western Arctic Ocean, Deep Sea Res., Part II, 56, 1274–1289,doi:10.1016/j.dsr2.2008.10.027.

Carmack, E. C., and R. W. Macdonald (2002), Oceanography of theCanadian Shelf of the Beaufort Sea: A setting for marine life, Arctic,55(1), 29–45.

Daase, M., K. Eiane, D. L. Aksnes, and D. Vogedes (2008), Vertical distri-bution of Calanus spp. and Metridia longa at four Arctic locations, Mar.Biol. Res., 4(3), 193–207, doi:10.1080/17451000801907948.

Falk-Petersen, S., C. C. E. Hopkins, and J. R. Sargent (1990), Trophicrelationships in the pelagic, Arctic food web, in Trophic Relationshipsin the Marine Environment, edited by M. Barnes and R. N. Gibson,pp. 315–333, Aberdeen Univ. Press, Aberdeen, U. K.

Forest, A., M. Sampei, H. Hattori, R. Makabe, H. Sasaki, M. Fukuchi,P. Wassmann, and L. Fortier (2007), Particulate organic carbon fluxeson the slope of the Mackenzie Shelf (Beaufort Sea): Physical and biolog-ical forcing of shelf-basin exchanges, J. Mar. Syst., 68(1–2), 39–54,doi:10.1016/j.jmarsys.2006.10.008.

Forest, A., M. Sampei, R. Makabe, H. Sasaki, D. G. Barber, Y. Gratton,P. Wassmann, and L. Fortier (2008), The annual cycle of particulateorganic carbon export in Franklin Bay (Canadian Arctic): Environmentalcontrol and food web implications, J. Geophys. Res., 113, C03S05,doi:10.1029/2007JC004262.

Forest, A., S. Belanger, M. Sampei, H. Sasaki, C. Lalande, and L. Fortier(2010), Three-year assessment of particulate organic carbon fluxes inAmundsen Gulf (Beaufort Sea): Satellite observations and sediment trapmeasurements, Deep Sea Res., Part I, 57(1), 125–142, doi:10.1016/j.dsr.2009.10.002.

Forest, A., V. Galindo, G. Darnis, C. Lalande, S. Pineault, J. E. Tremblay,and L. Fortier (2011a), Carbon biomass, elemental ratios (C:N) andstable isotopic composition (13C, 15N) of dominant calanoid cope-pods during the winter-summer transition in the Amundsen Gulf(Arctic Ocean), J. Plankton Res., 33(1), 161–178, doi:10.1093/plankt/fbq103.

Forest, A., et al. (2011b), Biogenic carbon flow pathways in the planktonicfood web of the Amundsen Gulf (Arctic Ocean): A synthesis of fieldmeasurements and inverse modeling analyses, Prog. Oceanogr., 91,410–436, doi:10.1016/j.pocean.2011.05.002.

Fortier, M., L. Fortier, H. Hattori, H. Saito, and L. Legendre (2001),Visual predators and the diel vertical migration of copepods under Arc-tic sea ice during the midnight sun, J. Plankton Res., 23(11), 1263–1278, doi:10.1093/plankt/23.11.1263.

Galley, R. J., E. Key, D. G. Barber, B. J. Hwang, and J. K. Ehn (2008),Spatial and temporal variability of sea ice in the southern Beaufort Seaand Amundsen Gulf: 1980–2004, J. Geophys. Res., 113, C05S95,doi:10.1029/2007JC004553.

Gnaiger, E. (1983), Calculation of energetic and biochemical equivalentsof respiratory oxygen consumption, in Polarographic Oxygen Sensors:Aquatic and Physiological Applications, edited by E. Gnaiger and H.Forstner, pp. 337–345, Springer, New York, doi:10.1007/978-3-642-81863-9_30.

DARNIS AND FORTIER: ZOOPLANKTON RESPIRATION AND CARBON FLUX C04013C04013

11 of 12

Hernández-León, S., and T. Ikeda (2005), A global assessment of mesozoo-plankton respiration in the ocean, J. Plankton Res., 27(2), 153–158.

Hernández-León, S., S. Torres, M. Gomez, I. Montero, and C. Almeida(1999), Biomass and metabolism of zooplankton in the Bransfield Strait(Antarctic Peninsula) during austral spring, Polar Biol., 21(4), 214–219,doi:10.1007/s003000050355.

Hernández-León, S., M. Gomez, M. Pagazaurtundua, A. Portillo-Hahnefeld,I. Montero, and C. Almeida (2001), Vertical distribution of zooplankton inCanary Island waters: Implications for export flux, Deep Sea Res., Part I,48(4), 1071–1092, doi:10.1016/S0967-0637(00)00074-1.

Hirche, H.-J. (1997), Life cycle of the copepod Calanus hyperboreus in theGreenland Sea, Mar. Biol. Berlin, 128(4), 607–618, doi:10.1007/s002270050127.

Hopcroft, R. R., C. Clarke, R. J. Nelson, and K. A. Raskoff (2005), Zoo-plankton communities of the Arctic’s Canada Basin: The contributionby smaller taxa, Polar Biol., 28(3), 198–206, doi:10.1007/s00300-004-0680-7.

Ikeda, T., and S. Motoda (1978), Estimated zooplankton productionand their ammonia excretion in Kuroshio and adjacent seas, Fish. Bull.,76(2), 357–367.

Ikeda, T., J. J. Torres, S. Hernández-León, and S. P. Geiger (2000), Metab-olism, in ICES Zooplankton Methodology Manual, edited by R. P. Harriset al., pp. 455–532, Academic, San Diego, Calif., doi:10.1016/B978-012327645-2/50011-6

Ingram, R. G., W. J. Williams, B. Van Hardenberg, J. T. Dawe, and E. C.Carmack (2008), Seasonal Circulation over the Canadian Beaufort Shelf,in On Thin Ice: A synthesis of the Canadian Arctic Shelf Exchange Study(CASES), edited by L. Fortier, D. G. Barber, and J. Michaud, pp. 13–36,Aboriginal Issues Press, Winnipeg.

Ingvarsdóttir, A., D. F. Houlihan, M. R. Heath, and S. J. Hay (1999), Sea-sonal changes in respiration rates of copepodite stage V Calanus fin-marchicus (Gunnerus), Fish. Oceanogr., 8(1), 73–83, doi:10.1046/j.1365-2419.1999.00002.x.

Kobari, T., D. K. Steinberg, A. Ueda, A. Tsuda, M. W. Silver, and M.Kitamura (2008), Impacts of ontogenetically migrating copepods ondownward carbon flux in the western subarctic Pacific Ocean, Deep SeaRes., Part II, 55(14–15), 1648–1660, doi:10.1016/j.dsr2.2008.04.016.

Lalande, C., A. Forest, D. G. Barber, Y. Gratton, and L. Fortier (2009), Var-iability in the annual cycle of vertical particulate organic carbon export onArctic shelves: Contrasting the Laptev Sea, Northern Baffin Bay and theBeaufort Sea, Cont. Shelf Res., 29(17), 2157–2165, doi:10.1016/j.csr.2009.08.009.

Longhurst, A., and R. Williams (1992), Carbon Flux by Seasonal Verti-cal Migrant Copepods Is a Small Number, J. Plankton Res., 14(11),1495–1509, doi:10.1093/plankt/14.11.1495.

Madsen, S. D., T. G. Nielsen, and B. W. Hansen (2001), Annual populationdevelopment and production by Calanus finmarchicus, C. glacialis andC. hyperboreus in Disko Bay, western Greenland, Mar. Biol. Berlin,139(1), 75–93, doi:10.1007/s002270100552.

Martin, J., J. E. Tremblay, J. Gagnon, G. Tremblay, A. Lapoussiere, C. Jose,M. Poulin, M. Gosselin, Y. Gratton, and C. Michel (2010), Prevalence,structure and properties of subsurface chlorophyll maxima in CanadianArctic waters, Mar. Ecol. Prog. Ser., 412, 69–84, doi:10.3354/meps08666.

Morales, C. (1999), Short communication. Carbon and nitrogen fluxes inthe oceans: The contribution by zooplankton migrants to active transportin the North Atlantic during the Joint Global Ocean Flux Study, J. Plank-ton Res., 21(9), 1799–1808, doi:10.1093/plankt/21.9.1799.

Mumm, N. (1991), On the summerly distribution of mesozooplankton inthe Nansen Basin, Arctic Ocean (in German), Ph.D. thesis, Naturwiss.Fak., Kiel Univ., Kiel, Germany.

Pickart, R. S. (2004), Shelfbreak circulation in the Alaskan Beaufort Sea:Mean structure and variability, J. Geophys. Res., 109, C04024,doi:10.1029/2003JC001912.

Riedel, A., C. Michel, M. Gosselin, and B. LeBlanc (2008), Winter-springdynamics in sea-ice carbon cycling in the coastal Arctic Ocean, J. Mar.Syst., 74(3–4), 918–932, doi:10.1016/j.jmarsys.2008.01.003.

Rysgaard, S., T. G. Nielsen, and B. W. Hansen (1999), Seasonal variationin nutrients, pelagic primary production and grazing in a high-Arcticcoastal marine ecosystem, Young Sound, northeast Greenland, Mar.Ecol. Prog. Ser., 179, 13–25, doi:10.3354/meps179013.

Sampei, M., A. Forest, H. Sasaki, H. Hattori, R. Makabe, M. Fukuchi, andL. Fortier (2009), Attenuation of the vertical flux of copepod fecal pelletsunder Arctic sea ice: Evidence for an active detrital food web in winter,Polar Biol., 32(2), 225–232, doi:10.1007/s00300-008-0523-z.

Sampei, M., H. Sasaki, A. Forest, and L. Fortier (2012), A substantialexport flux of particulate organic carbon linked to sinking dead cope-pods during winter 2007–2008 in the Amundsen Gulf (southeasternBeaufort Sea, Arctic Ocean), Limnol. Oceanogr., 57, 90–96, doi:10.4319/lo.2012.57.1.0090.

Scott, C. L. S., S. Kwasniewski, S. Falk-Petersen, and J. R. Sargent (2000),Lipids and life strategies of Calanus finmarchicus, Calanus glacialis andCalanus hyperboreus in late autumn, Kongsfjorden, Svalbard, PolarBiol., 23(7), 510–516, doi:10.1007/s003000000114.

Seuthe, L., G. Darnis, C. W. Riser, P. Wassmann, and L. Fortier (2007),Winter-spring feeding and metabolism of Arctic copepods: Insights fromfaecal pellet production and respiration measurements in the southeasternBeaufort Sea, Polar Biol., 30(4), 427–436, doi:10.1007/s00300-006-0199-1.

Steinberg, D. K., B. A. S. Van Mooy, K. O. Buesseler, P. W. Boyd, T.Kobari, and D. M. Karl (2008), Bacterial vs. zooplankton control of sink-ing particle flux in the ocean’s twilight zone, Limnol. Oceanogr., 53(4),1327–1338, doi:10.4319/lo.2008.53.4.1327.

Svensen, C., L. Seuthe, Y. Vasilyeva, A. Pasternak, and E. Hansen (2011),Zooplankton distribution across Fram Strait in autumn: Are small copepodsand protozooplankton important?, Prog. Oceanogr., 91(4), 534–544,doi:10.1016/j.pocean.2011.08.001.

Tremblay, J. E., H. Hattori, C. Michel, M. Ringuette, Z. P. Mei, C. Lovejoy,L. Fortier, K. A. Hobson, D. Amiel, and K. Cochran (2006), Trophicstructure and pathways of biogenic carbon flow in the eastern NorthWater Polynya, Prog. Oceanogr., 71(2–4), 402–425, doi:10.1016/j.pocean.2006.10.006.

Tremblay, G., C. Belzile, M. Gosselin, M. Poulin, S. Roy, and J. E. Tremblay(2009), Late summer phytoplankton distribution along a 3500 km transectin Canadian Arctic waters: Strong numerical dominance by picoeukar-yotes, Aquat. Microb. Ecol., 54(1), 55–70, doi:10.3354/ame01257.

Welch, H. E., T. D. Siferd, and P. Bruecker (1997), Marine zooplanktonicand benthic community respiration rates at Resolute, Canadian high Arc-tic, Can. J. Fish. Aquat. Sci., 54(5), 995–1005, doi:10.1139/f97-006.

Wexels Riser, C., P. Wassmann, M. Reigstad, and L. Seuthe (2008), Verti-cal flux regulation by zooplankton in the northern Barents Sea duringArctic spring, Deep Sea Res., Part II, 55(20–21), 2320–2329,doi:10.1016/j.dsr2.2008.05.006.

G. Darnis and L. Fortier, Québec-Océan, Département de biologie,Université Laval, Québec, QC, G1V 0A6 Canada. ([email protected]; [email protected])

DARNIS AND FORTIER: ZOOPLANKTON RESPIRATION AND CARBON FLUX C04013C04013

12 of 12