Journal of Geophysical Research: Oceans

Impact of biogeochemical processes and environmental factors on the calcium carbonate saturation state in the Circumpolar Flaw Lead in the Amundsen Gulf, Arctic Ocean



[1] We report on measurements across an annual cycle of carbon dioxide system parameters in the polar mixed layer (PML) of the circumpolar flaw lead in the Amundsen Gulf, Arctic Ocean. From these and other properties (nitrate, S, T) of the PML, we found that biological processes (photosynthesis and respiration) accounted for about 50% of the monthly variations in the carbonate ion concentration, [CO32−] and Ω, the saturation state of these waters with respect to calcite (ΩCa) and aragonite (ΩAr). Vertical mixing and salinity changes had equal impacts over the annual cycle. The impact of sea ice meltwater resulted in decreasing Ω values in summer, but most of this change was offset by the Ω increase as a result of CO2 drawdown during biological photosynthesis.

1. Introduction

[2] Because of the relatively cold and fresh waters, the Arctic Ocean is particularly sensitive to increased atmospheric carbon dioxide (CO2) levels and consequent ocean acidification. The latter may have deleterious effects on marine organisms that form calcium carbonate shells and skeletons and may result in marine ecosystem changes [Doney et al., 2009, and references therein]. In turn, this may impact carbon dynamics and transport in the water column. Chierici and Fransson [2009]reported the presence of aragonite-undersaturated waters on the freshwater-influenced shelves of the western Arctic Ocean in summer 2005, substantially sooner than predicted by recent dynamic models [Orr et al., 2005; Steinacher et al., 2009]. Depressed saturation states were also found in Barrow Strait in the eastern Canadian Arctic Archipelago, likely caused by the intrusion of Pacific water into the Arctic [Azetsu-Scott et al., 2010]. Given the potential ecological consequences, studies of processes affecting the natural variability of the calcium-carbonate saturation levels in the Arctic Ocean are of great importance in predicting the impact of increased atmospheric CO2 levels on the vulnerable Arctic Ocean.

[3] In this study, we report the results and interpretations of an 11-month time series of water column property measurements, carried out from October 2007 to August 2008 in the Amundsen Gulf of the western Canadian Arctic Archipelago. These measurements were used to characterize the CO2 system and the influence of freshwater additions (river runoff and sea ice melt), biological processes (photosynthesis and respiration), temperature, and physical mixing on the seasonal variability of carbonate ion concentrations (CO32−) and the calcium carbonate (CaCO3) saturation state in the polar mixed layer (PML). The relative contributions of sea ice melt and river runoff are estimated based on the relationships between salinity, and total alkalinity (AT) of the PML.

2. Study Area

[4] This study was conducted in the Amundsen Gulf of the western Canadian Arctic Archipelago (Figure 1). This area is influenced by an annually recurring polynya (i.e., opening in ice-covered waters) which forms due to wind-induced ice divergence and upwelling of warmer subsurface waters, maintaining an ice-free area for much of the year [Barber and Massom, 2007]. The Cape Bathurst Polynya is part of the circumpolar flaw lead that extends along the entire coastal Arctic Ocean [Barber and Massom, 2007] and a site of active sea ice formation, concomitant brine rejection and, hence, potential carbon transport to deeper layers.

Figure 1.

Map of study area with all sampling locations shown as black dots. The area within the dashed box is the study region on which the seasonal variability analysis was applied (122°W to 126°W, 70°N to 71.5°N).

[5] Figure 1 shows all stations occupied within Amundsen Gulf during the Circumpolar Flaw Lead Study (CFL) expedition. In this paper, we focused on the seasonal variability in the PML (depths < 50 m) in the region between 122 °W and 126 °W, and from 70 °N to 71.5 °N (dashed box in Figure 1).

2.1. Hydrography in the Amundsen Gulf

[6] In simple terms, the water column in this area is typically dominated by three water masses [Macdonald et al., 1989, 2002]: the relatively fresh polar mixed layer (PML), that approximately occupies the upper 50 m (S < 32), the deeper Upper Halocline Layer (UHL) of Pacific origin with a salinity of ∼33.1; and the bottom, warm and salty, Lower Halocline Layer (LHL), originating from the Atlantic (S > 34.6). Steep salinity gradients generated a strongly stratified water column, limiting exchange between surface and deeper waters. For a detailed description of the inorganic carbon system dynamics in the water column, we refer to Shadwick et al. [2011a]. Despite the strong stratification, episodes of wind and internal wave-driven upwelling [Mucci et al., 2010], as well as convection induced by brine rejection during ice formation [Shadwick et al., 2011a; A. Fransson et al., unpublished manuscript, 2011], have been documented in the study area. In summer, the upper water column (top 20 to 30 m) in the Amundsen Gulf is generally highly stratified by surface freshwater [Carmack and Macdonald, 2002; Mucci et al., 2010]. The surface flow into the Amundsen Gulf from the west is weak and the main circulation is dominated by a subsurface flow toward the Beaufort Sea [Lanos, 2009]. The consequence is that the river water has limited influence on the Amundsen Gulf [Macdonald et al., 2002; Magen et al., 2010] and the main source of surface freshwater in Amundsen Gulf is sea ice melt [Shadwick et al., 2011b; Tremblay et al., 2008].

2.2. Sea Ice Cover in Amundsen Gulf

[7] The direction and strength of the winds during fall as well as the summer ice conditions influences the timing of the freezeup, which generally begins in early to mid-October [Stewart et al., 1998]. We used archived ice chart data from the Canadian Ice Service [2007]to document the seasonal evolution of sea ice cover in the study area. In 2007, sea ice started to form southwest of Banks Island by 1 October, and the area was completely ice covered by early November. In mid-April, new ice was observed over a small part of the area, possibly the first signal of the flaw lead opening. One week later, the area was again completely ice covered (90% to 100%), and this condition prevailed until mid-May when sea ice started to retreat. Not until the first week of July, did the sea ice clear from the area, with 10% sea ice cover remaining in only a small part of the Amundsen Gulf. The seasonal ice summary for 2007 (CIS) indicates that extensive sea ice melt continued through September, leading to record-low ice extent in September 2007 [Comiso et al., 2008]. The freezeup in autumn 2007 was delayed by two to three weeks in the western Arctic.

3. Methods and Data

[8] The field-work took place from October 2007 to August 2008 as part of the International Polar year (IPY) during the Circumpolar Flaw Lead project (CFL) [Barber et al., 2010] onboard the Canadian icebreaker CCGS Amundsen. During this period, about 80 stations were visited (Figure 1), and more than 2300 water samples were collected for the determination of total alkalinity (AT), total inorganic carbon (CT), and nitrate (NO3). Samples were collected from 12-L Niskin bottles mounted on a General Oceanics 24-bottle rosette, equipped with a Conductivity-Temperature-Depth sensor (CTD, Seabird SBE-911 plus). Following water collection using standard protocols [Dickson et al., 2007], CT and AT samples were poisoned with 100 μL of a saturated HgCl2 solution to halt biological activity and stored in the dark, at 4°C, until analysis. Analytical methods for CT and AT determination have been fully described elsewhere [Dickson et al., 2007; Mucci et al., 2010; Shadwick et al., 2011a], but briefly, all CT and AT samples were analyzed onboard by coulometric and potentiometric titration, respectively. All CT and some AT analyses used a Versatile Instrument for the Determination of Titration Alkalinity (VINDTA 3C, Marianda), and some AT was also measured using an automated Radiometer® potentiometric titrator [Mucci et al., 2010]. The average standard deviation, determined from replicate sample analyses, was between 2 and 6 μmol kg−1. Routine analysis of Certified Reference Materials (CRM, provided by A. G. Dickson, Scripps Institution of Oceanography), where the reproducibility was better than ±1 μmol kg−1 and ±2 μmol kg−1, for CT and AT, ensured the accuracy of the measurements. The carbon dioxide system data will be published at the Carbon Dioxide Analysis Information Centre (CDIAC, and in the Polar Data Catalogue (Reference#: 10900, operated by the Canadian Cryospheric Information Network (CCIN,

[9] Nitrate samples were collected in acid-washed 15-mL polypropylene tubes. A 47-mm filter holder, containing a 5.0μm pore size polycarbonate filter, was attached directly to the sampling bottle to remove large particles. Colorimetric determination of NO3- was performed on an Autoanalyzer 3 using routine methods [Grasshoff et al., 1999]. Analytical detection limits were 0.03 μmol L−1 for NO3-. The nutrient data are available in the Polar Data Catalogue (

4. Calculations

4.1. Carbon Dioxide System

[10] We used CT, AT, salinity, temperature and depth for each sample as input parameters in a CO2-chemical speciation model (CO2SYS program [Pierrot et al., 2006]) to calculate the carbon dioxide partial pressure of (pCO2), carbonate ion concentration ([CO32−]), and the saturation state (Ω = [CO32−] * [Ca2+]/K*sp) with respect to aragonite (ΩAr) and calcite (ΩCa). We used the CO2-system dissociation constants (K*1 and K*2) estimated by Roy et al. [1993, 1994], since an internal consistency study showed them to be the most suitable constants for cold and fresher surface waters [Chierici and Fransson, 2009]. Calculations using the Roy et al. [1993, 1994] constants yield slightly higher ΩAr and ΩCa values than those derived from other CO2-system constants, such asMehrbach et al. [1973] (refit by Dickson and Millero [1987]). The calculations were performed on the total hydrogen ion scale (pHT) using the HSO4- dissociation constant of Dickson [1990]. The calcium concentration, [Ca2+], was assumed to be proportional to salinity, 10.28xSp/35 μmol kg−1, where Sp is the practical salinity. The stoichiometric solubility constants for aragonite and calcite (K*sp) were taken from Mucci [1983] and corrected for pressure according to Ingle [1975]. If Ω < 1, solutions are undersaturated, whereas, for Ω > 1, they are supersaturated with respect to the mineral of interest. The CO2SYS calculations were performed without the soluble reactive phosphate (SRP) and silicic acid (Si(OH)4) concentrations, introducing a mean error of approximately 0.8% in the [CO32−] and calcium carbonate saturation state estimates.

4.2. Water Mass Fractions

[11] The relationships between salinity and AT, were used to determine the end-member signatures of the two freshwater sources (meteoric water and sea ice meltwater) and the polar mixed layer (Table 1), according to well-established protocols [Macdonald et al., 1999, 2002; Cooper et al., 2008; Fransson et al., 2001, 2009]. The ATand salinity end-member values for sea ice meltwater were estimated from the AT and salinity measured in melted sea ice samples (ATice and Sice, respectively, N = 166) collected during the CFL project (Fransson et al., unpublished manuscript, 2011). The mean ATand salinity of the sea ice meltwater end-member (ATSIM and SSIM) were estimated to 509 ± 32 μmol kg−1 and 7.4 ± 0.5, respectively.

Table 1. Source Water Salinity and Total Alkalinity (AT) for River Runoff (Meteoric Water, MW), Sea Ice Meltwater (SIM), and the Polar Mixed Layer (PML)
Source WaterSAT (μmol kg−1)References
  • a

    Values obtained from this study.

MW01540 ± 35Cooper et al. [2008]
SIM7.4 ± 0.5509 ± 32Macdonald et al. [2002], Fransson et al. (unpublished manuscript, 2011)
PML winter32.0 ± 0.1a2256 ± 12aMacdonald et al. [1989, 1999]

[12] The relative fractions of PML water (fPML), sea ice meltwater (fSIM), and river runoff (meteoric water, fMW) were then computed from the following three equations:

display math
display math
display math

4.3. Evaluation of Major Controls on [CO32−] and CaCO3 Saturation State

[13] Here we estimate the effect of biological processes, physical mixing, and variations in salinity, and temperature on [CO32−] and the CaCO3 saturation state (ΩCa and ΩAr) in the upper 20 m in the study area. During photosynthesis, CO2 and NO3 are taken up by organisms resulting in an increase in [CO32−] and the CaCO3 saturation state of the waters. Conversely, organic matter respiration by heterotrophic microorganisms releases CO2 and NO3 to the waters, decreasing [CO32−] and CaCO3 saturation state. We use monthly variations (prefix Δ) in CT, AT, [CO32−], ΩAr, ΩCa, S, T, NO3, fSIM and fMW, to quantitatively distinguish the effects of biological primary production (suffix bio), physical mixing (suffix mix), salinity (suffix S) and temperature (suffix temp) on the [CO32−] and the CaCO3 saturation state.

[14] The salinity effects on [CO32−]S, ΔΩArS and ΔΩCaS were derived from the linear correlation between the monthly change in salinity (ΔS) and AT (ΔAT) in the upper 20 m (ΔAT = 50.5ΔS + 0, r2 = 0.95, rmse = ±3 μmol kg−1). The intercept of the regression was null (±1.7 μmol/kg), meaning that salinity changes explained most of the ΔAT, and other processes, such as biological primary production made insignificant contributions. We used the slope and the [CO32−]/AT ratio to estimate the change in [CO32−] due to changes in salinity, similar to the approach adopted by Shadwick et al. [2011a].

[15] The effects of biological CO2 drawdown during photosynthesis on [CO32−], ΩCa, and ΩAr ([CO32−]bio, ΩCabio, and ΩArbio, respectively) were estimated from the correlation between [CO32−] and NO3 during the period of largest NO3 loss, from March to August (see section 5.2 and Figure 5e). To exclude the effect of salinity variations on [CO32−], Ω, and [NO3], their values were normalized to a salinity of 31 ([CO32−]Snorm = 31[CO32−]/S, NO3Snorm = 31[NO3]/S), corresponding to the annual mean value in the upper 20 m. The linear correlation between mean salinity-normalized [CO32−] and [NO3]Snorm for this period, [CO32−]Snorm = −6.20 [NO3]Snorm + 106.7, yielded a coefficient of determination (r2) of 0.91, and rmse on the fit of ±2.3 μmol kg−1. The slope of the fit (∼6 ± 0.7 μmol kg−1) was multiplied by ΔNO3Snorm to estimate the change in carbonate ion concentration due to biological production ([CO32−]bio). The same approach was used to derive the biological effects on ΩAr and ΩCa, using slopes from the linear correlations between ΩAr or ΩCa and [NO3]Snorm, (slopes of −0.09, (rmse = 0.04) and −0.14 (rmse = 0.06), respectively and r2 for both fits of 0.95, Figure 2).

Figure 2.

The linear regression between salinity normalized nitrate (NO3Snorm) and the calcium carbonate saturation state between March and August. The slopes of ΩCa versus NO3SnormCa = −0.1345 × NO3Snorm + 2.6081, r2 = 0.941, black diamonds) and ΩArAr = −0.841 × NO3Snorm + 1.627, r2 = 0.939, open pyramids) are used to derive the biological effects on ΩAr and ΩCa.

[16] The [CO32−]mix, ΩCamix and ΩArmix were estimated using the monthly changes in the mixed layer depth (MLD) and the difference between [CO32−], ΩCa or ΩAr in the subsurface (ssw) and the surface (sw, upper 20 m) waters for each month, as shown for [CO32−], as an example, in equation (4). The values for [CO32−]ssw, ΩCassw, and ΩArssw were annual mean values taken from the depth below the PML layer (∼60 m, S > 32) i.e., 73 ± 12 μmol kg−1, 1.75 ± 0.3, and 1.10 ± 0.2, respectively.

display math

where MLD is the mixed layer depth (in m) at time t (month). Since deepening of the surface mixed layer induces mixing with underlying waters, the function Θ(dMLD/dt) is equal to dMLD/dt when dMLD/dt > 0, and equal to 0 when dMLD/dt ≤ 0 (usually referred as the Heavisied function).

[17] The effect of temperature on [CO32−], ΩCa, and ΩAr, was inferred from the thermodynamics of the carbonate system over the temperature range of −2°C to 7°C as described by the CO2SYS chemical speciation program [Pierrot et al., 2006]. Over this small temperature range, the correlation between [CO32−] and temperature ([CO32−] = 0.542xT + 93.72, r2 = 1, rmse = 0.002), ΩCa and temperature (ΩCa = 0.012xT +2.28, r2 = 0.999, rmse = 0.001), and ΩAr and temperature (ΩAr = 0.009xT + 1.42, r2 = 0.999, rmse = 0.001). Increasing temperature leads to increased [CO32−], ΩCa, and ΩAr, and the slopes of the linear regressions were multiplied with the ΔT to estimate the temperature effects.

[18] To explore the effect of changes in fSIM on the Δ[CO32−], ΩCa, and ΩAr, we used the linear relationship between solely salinity-dependent changes (i.e., ΔCO32−S, ΩCaS, and ΩArS) and the monthly change in fSIMfSIM), (we did not do this analysis for fMW because it is negligible relative to fSIM). The negative correlation between Δ[CO32−]Snorm and ΔfSIM indicates that a 1% increase in fSIM leads to a [CO32−] decrease of 1 μmol kg−1 (Figure 3, r2 = 0.951, rmse = 0.28) and 0.022 and 0.014 changes in ΩCa and ΩAr, respectively (r2 = 0.951, rmseΩCa = 0.007 and rmseΩAr = 0.004).

Figure 3.

The linear regression between the monthly change in salinity normalized carbonate ion concentration, Δ[CO32−]Snorm, and the monthly change in sea ice meltwater fraction, ΔfSIM, (Δ[CO32−]Snorm = −0.901x, ΔfSIM-0.04, r2 = 0.952).

4.3.1. Error Estimates

[19] The magnitude of the errors (Ex) in the evaluation of the major drivers of Δ[CO32−], ΔΩCa, and ΔΩAr were derived from the sum of the rmse values for each of the individual processes n (Ex shown for Δ[CO32−] in equation (5)).

display math

[20] Based on this calculation, image is ±3.8 μmol kg−1, EΔΩCa = ±0.18, and EΔΩAr = ±0.19. From these estimates, relative errors are about 4%, 8%, and 14% of the absolute values of the total change in Δ[CO32−], ΔΩCa, and ΔΩAr, respectively. It is important to realize these error estimates only take into account the rmse of the functions used to estimate the drivers. This uncertainty is responsible for only part of the uncertainties in the calculations. For instance Cbiofor each month is evaluated from the correlation over the period of March to August, which may not apply to the situation in fall. Note also that the estimate of the effect of biological photosynthesis approximates only new primary production, and not total primary production, because it does not account for recycled and imported nutrients. Another limitation of our approach is that it does not consider the coupled effects of physical-biological processes on the carbonate ion change, such as physical upwelling bringing nutrients to the surface, hence affecting the biological CO2 uptake. Moreover, this calculation does not include the measurement variability in the AT and CT values or the error associated with the CO2SYS calculation of [CO32−] and saturation states. Hence, these error estimates should be considered lower limits.

5. Results

5.1. Seasonal Variability of Physical Properties in the PML

[21] The seawater temperature (T, Figure 4a) was below zero for most of the year. Warming began in early June, reaching a maximum temperature of 8.7°C by the end of July in the upper 20 m. The warm period coincided with the lowest surface mixed layer salinity (∼28, Figure 4b), due to the influence of sea ice meltwater (Figure 4b). The highest salinity was observed from the end of February to mid-May (Figure 4b), coinciding with the period of the largest sea ice cover. The relatively high density water formed during this period reflects brine rejection and convection triggered by sea ice formation (Figure 4c) [Shadwick et al., 2011a; Fransson et al., unpublished manuscript, 2011]. By mid-October, the water column temperature was well below zero and near freezing, and salinity was increasing in response to physical mixing and sea ice formation.

Figure 4.

The seasonal variability in (a) temperature (°C), (b) salinity (S), and (c) potential density anomaly in the polar mixed layer (PML) in the boxed region (Figure 1).

5.2. Seasonal Variability of Carbon Dioxide System Parameters and Nitrate

[22] We found large seasonal variability in the top 20 to 40 m of the water column (Figure 5). The seasonal range of surface AT was between 2000 μmol kg−1 and 2270 μmol kg−1, and the seasonal cycle was similar to that of salinity [see also Shadwick et al., 2011a]. Low values (AT < 2150 μmol kg−1) were recorded in the upper 15 m from mid-July to the beginning of September (Figure 5a), whereas the high AT values were observed throughout the polar mixed layer (PML) between April and the end of May (AT ≥ 2225 μmol kg−1, Figure 5a). The seasonal variability in CT was similar to that of AT, with a pronounced minimum (≤1950 μmol kg−1) from July to the beginning of September and high values from April to May. From April to the end of May, the temperature was lowest and the salinity highest (Figures 4a and 4b). The high salinity that extended throughout the PML was likely the result of brine rejection during sea ice formation, ensuing convection affecting AT and CT, as well as, transporting CO2 from the surface to deeper waters (fugacity of CO2, fCO2 in Figure 5c). Evidence of brine rejection and convection was also found by Lanos [2009]. Relatively low CT concentrations were observed at approximately 30 to 50 m depth in the summer period. This subsurface patch was better defined in the fCO2 and pH fields during August 2008 (Figure 5c and 5d), and was likely associated to biological CO2 drawdown by a subsurface phytoplankton bloom. This agreed with an observed fluorescence maximum (data not shown) and low nitrate levels observed over the same depth interval (NO3 < 0.2 μM, Figure 5e). Subsurface blooms have previously been observed in this area [Tremblay et al., 2008].

[23] The [CO32−] varied between 53 and 127 μmol kg−1, also with a large seasonal variability (Figure 5f). The ΩAr (Figure 5g) and ΩCa (Figure 5h) followed the same seasonal cycle as [CO32−], with high values during summer, decreasing at the end of the year. Values generally decreased with depth, although the highest values were located at the same depth as the apparent subsurface bloom (30 to 40 m). The increasing [CO32−] and saturation states observed between February and July were due to biological CO2 drawdown during photosynthesis. In December 2007 and January 2008, we found a patch of water that was undersaturated with respect to aragonite (ΩAr < 1), coinciding with high salinity (Figure 4b), high NO3 concentrations (Figure 5e), and low pH in situ (Figure 5d) below 60 m. This ΩAr undersaturation may have been caused by upwelling of upper halocline waters of Pacific origin characterized by a low saturation state [Azetsu-Scott et al., 2010]. The mean annual ΩAr and ΩCa in the PML were 1.4 ± 0.2 and 2.2 ± 0.4, respectively.

Figure 5.

Seasonal variability in (a) total alkalinity (AT, μmol kg−1), (b) total dissolved inorganic carbon (CT, μmol kg−1), (c) the carbon dioxide fugacity (fCO2, μatm), (d) the pHT at in situ temperature (pHT i-s), (e) the nitrate concentration (NO3, μmol L−1), (f) the carbonate ion concentration ([CO32−], μmol kg−1), and the calcium carbonate saturation of (g) aragonite (ΩAr) and (h) calcite ΩCa, in the polar mixed layer (PML).

Figure 5.


5.3. Seasonal Variability of Freshwater Fractions

[24] There was no clear seasonal variability in the contribution of river runoff to the upper 20 m, and the fMW varied between zero and a maximum of 4% in August (Figure 6). The contribution of sea ice melt, fSIM, had a stronger seasonal cycle (Figure 6). The fSIM values were at a minimum in March and April, increased steeply in early June, and reached maximum levels of 13% in September. Gradually decreasing fSIMcorresponded with the onset of sea ice formation from mid-October and through the winter. Our study agrees with previous findings that sea ice meltwater is the main freshwater source to the Amundsen Gulf, and that river runoff plays a minor role [Magen et al., 2010; Miller et al., 2011; Shadwick et al., 2011b].

Figure 6.

The seasonal variability of freshwater fraction from river runoff (fMW, %) and the freshwater fraction from sea ice meltwater (fSIM, %) in the upper 20 m of the water column.

5.4. Magnitude of the Processes Affecting the [CO32−] and CaCO3 Saturation

[25] In Figure 7a, we summarize the monthly change of Δ[CO32−] due to: biological processes (Δ[CO32−]bio), physical mixing (Δ[CO32−]mix), sea ice melt (ΔfSIM) and variations in salinity (Δ[CO32−]S) and temperature (Δ[CO32−]temp) changes in the upper 20 m of the water column. Figures 7b and 7c show the same components of the monthly variations of ΩCa and ΩAr. Negative values indicate a net loss of [CO32−] and concomitant decrease in CaCO3 saturation states (ΩCa and ΩAr), whereas positive values indicate a net increase in [CO32−], ΩCa and ΩAr from the previous month. For example biological production played a major role in modulating the carbonate chemistry of the upper 20 m from April to June, resulting in a significant increase of [CO32−], ΩCa and ΩAr over this period. The gain was offset by decreases in [CO32−], ΩCa and ΩAr in autumn, in response to cooling and the addition of CO2 by mixing with subsurface waters.

Figure 7.

The monthly change in the upper 20 m in (a) [CO32−] (μmol kg−1), (b) ΩCa, and (c) ΩAr due to biological processes (photosynthesis (plus signs) and respiration (minus signs), Cbio, green striped), physical mixing (Cmix, white), sea ice meltwater (ΔfSIM, blue), and salinity (CS, orange) and temperature (Ctemp, black) changes. C represents either [CO32−], ΩCa, or ΩAr. Negative values represent losses from the surface water, and positive values represent gains.

[26] Table 2 summarizes the relative impact (Crel, %) of each of the considered processes on the annual variability of [CO32−], ΩCa and ΩAr in the top 20 m of the water column in the study area. Biological processes controlled more than 50% of the annual variability in [CO32−], in ΩCa and ΩAr. Physical mixing and the salinity-related changes (combined effect of changes in salinity and ΔfSIM), were in the same order and accounted for 16% to 21%. Temperature played a minor role and accounted for about 10% of the total variations. The error on Crel is about 4%, 8%, and 14% of the absolute values of the total change in ΔCO32−, ΔΩCa, and ΔΩAr, respectively.

Table 2. Relative Impact (Crel, %) of Each of the Considered Processes on the Seasonal Variability of [CO32−], ΩCa, and ΩAra
  • a

    C represents changes in [CO32−], ΩCa, or ΩAr due to variability of biological processes (Cbio), physical mixing (Cmix), salinity related changes (CS and fSIM) and temperature variations (Ctemp). The error on Crel is about 4%, 8%, and 14% of the absolute values of the total change in ΔCO32−, ΔΩCa, and ΔΩAr, respectively.


6. Discussion

[27] We find that biological processes have the most impact on change in [CO32−], ΩAr and ΩCa over the annual cycle. The combined effect of salinity changes and ΔfSIM was of similar magnitude as vertical mixing over the annual cycle. Vertical mixing had a major impact from September to December, whereas most of the salinity change is offset by temperature and biological CO2 drawdown in spring/summer (Figure 7). We summarize the net effect of the major processes on ΩAr during three periods of the year: winter (January to March); spring/summer (April to August); and autumn (September to December) in Figure 8.

Figure 8.

The net effect of the two major processes (Cmix and Cbio) on the ΩAr in the upper 20 m. The values of net changes in the ΩAr for three periods of the year are indicated. We also show a qualitative representation of the growth stages of the pteropod Limacina Helicina from veligers and juveniles (smaller symbols) to adults (largest symbol) in relation to the changes in ΩAr.

[28] The physical mixing in late autumn and winter decreased ΩAr by 0.20, as relatively low pH water was brought to the surface from depth. Below the PML, the influence of Pacific water increases, and its impact on calcium carbonate saturation states was also noted by Azetsu-Scott et al. [2010]. Thus, any increase in the Pacific water inflow to the Arctic Ocean could substantially lower the saturation states.

[29] Our analysis showed that sea ice meltwater has a relatively low impact on the seasonal change of saturation states. However, earlier reports have related aragonite undersaturation and decreasing ΩAr on increasing sea ice meltwater [Yamamoto-Kawai et al., 2009; Chierici and Fransson, 2009]. The reason for the different conclusions between our study and the previous reports may be the due to the fact that we use data from a full annual cycle, whereas their estimates are based on data collected in summer.

[30] Further decreases in ΩAr occur during winter due to net respiration, but the main increase in spring and summer due to net biological production counteracts the winter decrease (Figure 8). Although the biological processes result in no net change over the annual cycle, the large seasonal variations could have significant effects on the life cycles of individual organisms and thereby the overall ecosystem functioning in the area. For example, the pteropod Limacina Helicina, which forms an aragonite skeleton and is ubiquitous throughout the Arctic Ocean, may be particularly vulnerable to the seasonal amplitude of the ΩAr variations in the surface waters. Several studies show that the larvae and juvenile stages are crucial for the development of hard parts for several organisms [Fabry et al., 2008; Dupont and Thorndyke, 2009], although it is not yet clear whether that is the case for L.Helicina. Nonetheless, the youngest specimens L.Helicina are concentrated in the polar mixed layer [Kobayashi, 1974], where we have observed the largest variations in carbonate chemistry and saturation state, and the transition from veligers into juveniles takes place in winter [Gannefors et al., 2005], when we found that the aragonite saturation state drops dramatically. This implies that the main skeleton development occurs in the same period when physical mixing and biological respiration are the most likely to result in an undersaturation (Figure 8).

[31] Changes in the life cycle or the survival of L. Helicina, as well as other calcifying organisms, may modify the efficiency of the biological carbon pump in the Arctic Ocean and in turn, the seasonal cycle we have observed in this study, constituting either a positive or negative feedback in the Arctic Ocean carbon cycle. Such changes will also ultimately affect the efficiency of the biological pump and the net air-sea CO2 exchange. Our observation that the annual biological cycle has a strong influence on CaCO3 saturation states emphasizes the importance of full annual data coverage of the oceanic carbonate system if we are to ultimately understand the impact of ocean acidification in the Polar Ocean.


[32] We are grateful to the captains, officers, and crew of the Canadian Coast Guard Ship (CCGS) Amundsenfor their cooperation in the collection of field data, as well as to the chief scientists throughout the project. We thank Yves Gratton and his team of rosette operators for the collection and distribution of hydrographic data. Many thanks to Friederike Prowe, Stelly Lefort, Constance Guignard, Nes Sutherland, Stephanie Moore, Doris Leong, and Kyle Simpson for their assistance with sample collection and analysis. This work was supported by Swedish Research Council (VR) projects 2008-6226 and 2007-8365, the Royal Society of Arts and Sciences in Sweden, the Canadian Natural Science and Engineering Research Council, ArcticNet, Fisheries and Oceans Canada, and MetOcean DataSystems. This research contributes to the Canadian International Polar Year initiatives, as part of the Circumpolar Flaw Lead System Study project.