JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 113, C07016, doi:10.1029/2007JC004462, 2008
An annual study of inorganic and organic nitrogen and phosphorus
and silicic acid in the southeastern Beaufort Sea
Kyle G. Simpson,1 Jean-Éric Tremblay,1,2 Yves Gratton,3 and Neil M. Price1
Received 24 July 2007; revised 5 February 2008; accepted 5 March 2008; published 12 July 2008.
[1] Water column samples from the Mackenzie Shelf, the Beaufort Sea, and the
Amundsen Gulf were obtained during 2003–2004 to investigate nutrient dynamics in an
Arctic ecosystem influenced by a large river and flaw lead polynya. Nutrient inventories in
the upper water column showed a significant seasonal drawdown of nitrate (NO
3 ) and
silicic acid (Si(OH)4) (1:1.75 mol/mol) and subsequent accumulation of nitrite (NO
2)
and ammonium (NH+4 ). Dissolved organic nitrogen and phosphorus concentrations
(DON and DOP) were elevated in surface waters during the time of peak phytoplankton
growth (bloom and postbloom phases) and covaried in a 17:1 molar ratio. Vertical profiles
3
showed the typical middepth maxima in NO
3 , PO4 , and Si(OH)4 at salinity 32–33.1.
The concentration of DOP (0.76 ± 0.27 mmol P L1) was also 2 times higher in this
layer compared to the surface average and inversely correlated with the water mass
3
tracer N* [N* = (NO
3 16 PO4 + 2.9) 0.87], suggesting that Pacific-derived
waters were a source of the enrichment. Below the mixed layer, DON was generally
constant with depth, although averaged profiles (like those of urea) suggested the
presence of subsurface maxima at 50 m and between 250 and 300 m. Regeneration
ratios varied with depth and were approximately 9.0 mol O2/mol NO
3 and 122.5 mol
in shallow Pacific-derived waters (halocline layer) and 17.4 and
O2/mol PO3
4
193.5 in deep Atlantic waters (Atlantic layer), respectively. Deep waters of the
3
1
L1,
Amundsen Gulf contained an excess of 1.7 mmol NO
3 L , 0.12 mmol PO4
1
and 6.2 mmol Si(OH)4 L and a deficit in O2, relative to waters of similar density
in the Beaufort Sea, in proportions consistent with the remineralization ratios
derived from oxygen-nutrient regressions. Enhanced export of particulate matter from
the overlying polynya and subsequent remineralization at depth are hypothesized to
create this nutrient enrichment.
Citation: Simpson, K. G., J.-É. Tremblay, Y. Gratton, and N. M. Price (2008), An annual study of inorganic and organic nitrogen and
phosphorus and silicic acid in the southeastern Beaufort Sea, J. Geophys. Res., 113, C07016, doi:10.1029/2007JC004462.
1. Introduction
[2] Uncertainties in ocean-atmosphere coupled feedbacks
remain a complex problem in understanding the impact of
future climate on ocean ecosystems [Soden and Held,
2006]. Of particular interest from a biogeochemical perspective is how changes in ocean circulation and mixing
affect nutrient distribution and consequently planktonic
production and regeneration. In the Arctic Ocean, for
example, some oceanographic changes have either already
taken place [McLaughlin et al., 1996; Dickson et al., 2000;
Serreze et al., 2003; Barber and Hanesiak, 2004; Wu et al.,
2004] or are very likely to occur given the complexity of
Arctic water mass structure and the decline in sea ice cover.
1
Department of Biology, McGill University, Montréal, Québec, Canada.
Département de Biologie, Université Laval, Pavillon Alexandre-Vachon,
Québec City, Québec, Canada.
3
Institut national de la recherche scientifique-Eau, Terre et Environnement,
Québec City, Québec, Canada.
2
Copyright 2008 by the American Geophysical Union.
0148-0227/08/2007JC004462
Recent studies identify the sensitivity of the Arctic Ocean to
climate change [Intergovernmental Panel on Climate
Change (IPCC), 2001] and focus on features that set it
apart from other oceans. These include the influence of
freshwater (nutrient) input to the continental shelves
[Carmack et al., 1989; Aagard and Carmack, 1994;
Shiklomanov et al., 2000; Arnell, 2005] and the importance
of polynyas to biological production [Tremblay et al.,
2002; Arrigo and van Dijken, 2004; Tremblay et al.,
2006]. Both are predicted to be greatly affected by rising
temperatures. Meaningful analysis of the impacts of oceanographic change on Arctic ecosystems and validation of
models can only be accomplished if we have historical data
to compare with future measurements. In some of the most
biologically active regions of the Arctic, such baseline data
are still lacking.
[3] The water column of the southeastern Beaufort Sea, in
the western Canadian Arctic, is composed of several layers
whose structure has been thoroughly examined [Carmack et
al., 1989; McLaughlin et al., 1996; Codispoti et al., 2005].
According to Wang et al. [2006] and Codispoti et al. [2005],
C07016
1 of 16
C07016
SIMPSON ET AL.: NUTRIENT CYCLING IN THE BEAUFORT SEA
the surface mixed layer or polar mixed layer (PML) extends
from surface to 25– 50 m and is highly variable in temperature, salinity, and nutrient content. Below it lies the upper
halocline layer (UHL) (50 – 200 m), composed primarily of
Pacific-derived waters from the Bering Sea with salinities
less than 34.4. The lower halocline layer (LHL) below the
UHL [Jones and Anderson, 1986] has a salinity range of
34.2 – 34.6 and originates mainly from the Atlantic [Wang et
al., 2006]. The Atlantic layer (AL) occurs below the LHL
between 250 and 850 m and is warmer and more saline than
the halocline. Canadian basin deep water (CBDW) occupies
the bottom, at depths below 850 m and with salinities of at
least 34.8.
[4] The distribution of dissolved inorganic nutrients in the
Beaufort Sea of the Canadian Basin follows several wellestablished patterns that can be used to define the boundaries of the water masses and understand the processes that
influence nutrient abundance. Nitrate concentrations in the
PML are generally low and in some regions, like the outer
Chukchi Shelf, undetectable and limiting to primary production [Codispoti et al., 2005]. A well-defined nutrient
maximum occurs at salinity ca. 33.1, associated with UHL
water. Part of the maximum originates from the Pacific
Ocean, but the majority of it is created by biological and
physical processes as the water passes through the Chukchi
and East Siberian seas [Bauch et al., 1995, and references
therein]. Lower halocline layer water can be distinguished
by a minimum in the tracer NO [Broecker, 1974], while the
core of Atlantic water resides below the halocline between
300 and 800 m and is characterized by a temperature
maximum [Bauch et al., 1995, and references therein].
Little is known of the abundance and water mass distribution of dissolved organic nutrients in the Arctic. However,
one study [Wheeler et al., 1997] measured low concentrations of DON across the Arctic (lowest over the Chukchi
Shelf) and suggested that rivers, ice melt, and in situ
production were the major sources.
[5] The Mackenzie Shelf in the southeastern Beaufort Sea
is small compared to the extensive shelves of the Russian
Arctic. It is bounded by the Mackenzie Canyon and the
Amundsen Gulf in the east and west, and by the Beaufort
Sea and Mackenzie River to the north and south. Biological
processes on the shelf are greatly affected by seasonal
changes in freshwater input, ice coverage, and irradiance.
Data on the nutrient chemistry and distribution on
the Mackenzie Shelf is sparse [Macdonald et al., 1987;
Carmack et al., 1989; Macdonald et al., 1989; Moore et al.,
1992] and generally limited to observations of dissolved
inorganic nutrients made in the mid 1970s and 1980s. In
close proximity to the Mackenzie Shelf is the Cape Bathurst
Polynya in the Amundsen Gulf, part of the Arctic Ocean’s
circumpolar flaw lead polynya system [Arrigo and van
Dijken, 2004]. To our knowledge, the only biogeochemical
data for this region come from 18 vertical profiles of
inorganic nutrients obtained during August – September
1977 [Macdonald et al., 1978].
[6] From September 2003 to August 2004, a multidisciplinary research project was conducted aboard the CCGS
Amundsen in the Southern Beaufort Sea, the Mackenzie
Shelf and the Amundsen Gulf. Here we present the results
of measurements of the concentrations and distributions of
C07016
3
inorganic (NH +4 , NO 3 , PO 4 , Si(OH) 4 ) and organic
nutrients (DON, urea, and DOP) throughout the region.
2. Materials and Methods
2.1. Study Site
[7] Hydrographic measurements were made aboard the
Canadian Coast Guard icebreaker CCGS Amundsen, during
four open water legs of the 2003 – 2004 Canadian Arctic
Shelf Exchange Study (CASES) program. A total of 120
nutrient profiles were obtained during Leg 1: 13 September
to 14 October 2003; Leg 2: 15 October to 25 November
2003; Leg 7: 13 May to 24 June 2004; and Leg 8/9: 25 June
to 31 August 2004, at sites located throughout the Mackenzie Shelf, Amundsen Gulf and southern Beaufort Sea
(Figure 1).
2.2. Sampling
[8] Discrete water samples for nutrients were collected
using 12-L Niskin bottles mounted on a Seabird carousel
rosette system equipped with Seabird 911 CTD (Sea-Bird
Electronics, Incorporated). Samples were filtered through
5.0 mm polycarbonate filters (Poretics Corporation) and
collected into acid-cleaned (10% HCl) and sample-rinsed
polypropylene tubes (Sarstedt, Incorporated). One sample
+
3
from each depth was analyzed for NO
3 , NO2 , NH4 , PO4 ,
and Si(OH)4 concentration within one hour of collection. A
replicate sample from each depth was stored frozen
(20°C) and analyzed later for total dissolved nitrogen
and phosphorus (TDN and TDP), and urea. Temperature,
salinity, and oxygen were obtained using a Sea-Bird
911 CTD. Oxygen calibration was performed using Winkler
titration with a Mettler DL22.
2.3. Inorganic Nutrient Analyses
+
3
[9] Concentrations of NO
3 , NO2 , NH 4 , PO 4 , and
Si(OH)4 were measured using standard colorimetric methods
[Grasshoff, 1999] adapted for use on a 4-channel (multitest
18 and 19) continuous flow Bran + Luebbe Auto-Analyzer
3 (AA3). Nitrite concentrations were low and generally
represented <10% of the combined NO
3 + NO2 concentration. Nitrate concentrations reported here include NO
2
unless otherwise noted.
[10] The accuracies of the measured concentrations
(reported as percent relative deviation or coefficients
3
of variation) of NO
3 , PO4 , and Si(OH)4 were ±0.9% at
1
15 mmol L , ±1.5% at 1.0 mmol L1, and ±0.8% at
10 mmol L1, and the detection limits were 0.012 mmol L1,
0.016 mmol L1, and 0.020 mmol L1, respectively. Sampling error was assessed by analyzing the nutrient concentrations in five water samples collected at the same time
from the same sampling depth. The coefficients of variation
3
of NO
3 , PO4 , and Si(OH)4 concentrations in these samples were 0.45, 0.56, and 0.57%, respectively. Standard
curves were run with each batch of samples using freshly
prepared standards that spanned the range of concentrations
in the samples. The r2 values of all the standard curves,
including the analyses of the nutrients described below,
were greater than or equal to 0.99.
[11] Salinity varied from 5 to 35 in the study area because
of regional differences in ice melt, river input, and water
mass stratification. A correction factor was thus applied to
2 of 16
C07016
SIMPSON ET AL.: NUTRIENT CYCLING IN THE BEAUFORT SEA
C07016
Figure 1. Location of the Canadian Arctic Shelf Exchange Study (CASES) sites during 2003 – 2004 in
the southeastern Beaufort Sea, Amundsen Gulf, and over the Mackenzie Shelf. Stations are marked by
filled circles. Depth contours are in meters.
calculate nutrient concentrations in these samples to compensate for the salt matrix effect as described [Grasshoff,
1999].
2.4. Total Dissolved and Organic Nutrient Analyses
[12] Total dissolved nitrogen (TDN) and phosphorus
(TDP) concentrations were determined using persulphate
oxidation [Valderrama, 1981]. The efficiency of this method
was calculated to be 98% on the basis of the conversion of a
number of different amino acids and phosphate monoesters
3
to NO
3 and PO4 , respectively. Frozen samples were
thawed in a tepid water bath upon return to the laboratory
and centrifuged at 3500 g for 20 min to remove particulate
material. The supernatant was transferred directly to precombusted (500°C for 24 h) and acid-cleaned (10% HCl)
borosilicate reaction vials with Teflon-lined screw-cap lids
for hydrolysis. Concentrations of DON and DOP were
calculated by subtracting the ambient concentrations of
+
3
NO
3 , NO2 , and NH4 , and PO4 from the TDN and TDP
concentrations, respectively. A subset of samples was filtered through precombusted GF/F filters prior to hydrolysis.
The DON and DOP concentrations in these samples were
not significantly different from the centrifuged samples. The
coefficients of variation of replicate samples of TDN and
TDP were ±1.2% at 16 mmol L1 and ±2.7% at 1.5 mmol L1,
respectively. Errors in the calculations of DON and DOP
concentrations were computed using error propagation to be
±2.1% at 15 mmol L1 and ±3.0% at 1.5 mmol L1,
respectively.
[13] Urea concentration was measured using the room
temperature diacetylmonozime method of Goeyens et al.
[1998], with an analytical detection limit of 0.2 mmol N
L1. Urea-N concentration was also subtracted from the
DON concentration for one set of analysis to calculate the
DON minus urea concentration.
2.5. Water Mass Characterization
[14] Data were grouped spatially and temporally. For the
purpose of regional analysis, samples obtained east of
128.35°W (approximately the tip of Cape Bathurst) were
considered to be from the Amundsen Gulf (East stations)
and samples obtained west of 128.35°W were considered to
be from the Mackenzie Shelf or Beaufort Sea (West stations) (see Figure 1). Stations in the western region of the
CASES site were further divided according to bottom depth:
Beaufort Sea (250 m) and Mackenzie Shelf (249 m).
[15] Depth strata corresponding to unique water masses in
the CASES study region (Canadian Basin – Beaufort Sea)
were identified from temperature-salinity plots, which were
analyzed in a piecewise manner (data not shown) [Carmack
et al., 1989]. This approach was expanded to include
separation of the water masses on the basis of propertyproperty plots of nutrient concentrations. The water masses
were largely restricted to the following depths: 0 – 50 m;
50– 125 m; 125– 200 m; 200– 275 m; 275 m to bottom, as
previously described [Carmack and Kulikov, 1998;
McLaughlin et al., 2002; Codispoti et al., 2005; Wang et
al., 2006]. They corresponded to: the polar mixed layer
(0– 50 m); the upper halocline (50 – 200 m), which was
further subdivided according to nutrient content; the lower
halocline (200– 275 m); and the Atlantic layer (>275 m).
2.6. Temporal Characterization
[16] The data were partitioned into four temporal phases
on the basis of seasonal changes in phytoplankton biomass
(chlorophyll a) (S. Brugel, personal communication, 2007),
3 of 16
C07016
SIMPSON ET AL.: NUTRIENT CYCLING IN THE BEAUFORT SEA
phytoplankton 15NO
3 uptake rates (K. G. Simpson et al.,
Nutrient dynamics in the Mackenzie Shelf, Amundsen Gulf
and Cape Bathurst Polynya: 2. Stable isotope tracer estimates of new and regenerated production, manuscript in
preparation, 2008), and surface-layer-integrated nutrient
concentrations. These phases were labeled prebloom,
bloom, postbloom and autumnal and were categorized
according to day of year (DOY): 136 – 156; 157 – 182;
183 –220; 273 – 330, respectively. Low levels of photosynthetically active radiation (PAR) were present during the
prebloom (136 – 156) and autumnal phases (273 – 330).
These periods were characterized by slow rates of nitrogen
assimilation by plankton (Simpson et al., manuscript in
preparation, 2008). Bloom and postbloom phases had similarly high PAR availability (24 h day length with open
water) and fast rates of nitrogen assimilation by plankton.
Day 182 marked the end of the bloom phase on the basis of
the cessation of the apparent sea-surface nutrient drawdown.
2.7. Statistical and Data Analyses
[17] Statistical analyses were performed using Systat
v11.00.01. For predictive purposes, the linear (model 1),
least squares regression analysis was employed. Direct
stoichiometric comparisons utilized the geometric mean
(model 2) regression [Sokal and Rohlf, 1995]. Samples
1
) or low PO3
containing low NO 3 (<1.5 mmol L
4
1
(<0.5 mmol L ) were excluded from regression analyses
to avoid potential bias caused by non-Redfield uptake
kinetics due to nutrient exhaustion and greater analytical
error at low concentrations. Ammonium concentrations
measured between 6 and 27 November 2003 were unreliable because of a methodological artifact and were thus
excluded from the analyses.
[18] Water column nutrient inventories were calculated
using standard trapezoidal integration. All comparative
analyses that utilized derived variables employed propagation of error in the calculation of standard deviations.
Spatial differences between the Beaufort Sea and Amundsen
Gulf deep water (200–500 m) were calculated from weighted,
average, vertical profiles. Seasonal differences in concentrations were assessed using 1-way ANOVA and Tukey post
hoc comparison at a significance level of p 0.05.
[19] We used the N* parameter as a water mass tracer to
identify modified Bering Sea water in the upper halocline
layer at the CASES study site. N* calculates the deficit or
3
excess of NO
3 relative to PO4 assuming that regeneration
3
produces NO3 and PO4 in a 16:1 molar ratio, viz. N* =
3
+ 2.9) 0.87], [Gruber and
[(NO 3 16 PO 4
Sarmiento, 1997]. Denitrification decreases the NO
3 concentration in the water column of the Bering Sea so that seasurface values of N* are between 7 to 3 mmol L1
[Lehmann et al., 2005]. As the Bering water flows over the
continental shelf into the Arctic Ocean it undergoes further
modification that reduces N* to 15 to 10 mmol L1
[Codispoti et al., 2005]. Vertical profiles in the Canadian
Basin show highly negative N* in the upper halocline that
represents water that originated from the Bering Sea
[Wheeler et al., 1997; Tremblay et al., 2002].
2.8. Terminology
[20] For simplicity, we used the terms phosphate (PO3
4 )
and silicic acid (Si(OH)4) to describe soluble, molybdate-
C07016
reactive phosphorus (orthophosphate) and soluble, molybdatereactive silicate (silicic acid), respectively. All dates
described are in day of year (DOY) format beginning day
1 on 1 January.
3. Results
3.1. Nitrate, Phosphate, and Silicic Acid – Salinity
Relationships
[21] Property-salinity plots were used to characterize the
dominant water masses in the study area and their depth
distribution (Figure 2). A nutrient-rich subsurface layer with
salinity of ca 33.1 and sigma-t (st) of 26.6 occupied depths
between 125 and 200 m. On the basis of its salinity, this was
Pacific-derived water from the Bering Sea [Jones and
Anderson, 1986; Carmack et al., 1989; McLaughlin et al.,
1
1996]. It contained 1.9 mmol PO3
4 L , 17.3 mmol NO3
1
1
L and 33.5 mmol Si(OH)4 L and was thus relatively
enriched in PO3
(N:P = 9.1) compared to Redfield pro4
portions. The Si:N ratio was 1.9, slightly lower than the
Bering Sea outflow but similar to waters composing the 50
to 250 m depth layer on the Chukchi slope [Cooper et al.,
1997]. There was little regional difference in the depth and
density of the water containing these nutrient maxima
(Table 1). In the Amundsen Gulf, however, the peak NO
3
concentration occurred at roughly 200 m depth (at a salinity
of 34.2), well below the PO3
4 and Si(OH)4 maxima in this
region and below the NO
3 maximum observed in the
Beaufort Sea/Mackenzie Shelf. Mixing of the Bering water
with deeper Atlantic-derived waters (salinity 34.6, depth
275 m) and with surface Arctic water (salinity 30, depth
< 50 m) yielded two well-defined mixing lines (Figure 2,
insets). The deep Atlantic water contained less PO3
4 and
Si(OH)4 than the Bering layer but an equivalent concentration of NO
3 , so that its N:P ratio was considerably greater
than observed at the nutrient maximum. A low salinity
surface lens occupied the upper 50 m, and was generated
primarily by freshwater dilution by the Mackenzie River
and to a lesser extent by ice melt. Extrapolation of the
mixing line from 30 to zero salinity showed that the
Mackenzie River contributed roughly 40 mmol Si(OH)4
1
L1 and 4 – 5 mmol NO
to the Shelf waters. As
3 L
to the Mackenzie Shelf
expected, river supply of PO3
4
was very small. Independent measurements at sites near
the Mackenzie River discharge yielded concentrations of
1
56 mmol Si(OH)4 L1, 4 mmol NO
3 L , and <0.1 mmol
3
1
PO4 L . High variability in nutrient concentration of the
surface waters (Figure 2) reflected seasonality in freshwater
input, mixing, and phytoplankton consumption.
3.2. Seasonal Patterns of Inorganic and Organic
Nutrients
[22] Seasonal trends in dissolved nutrient concentrations
were evaluated by dividing the data according to the timing
of the peak and decline in phytoplankton growth. We used
changes in the integrated surface NO
3 concentration to
determine the onset and duration of the bloom and thereby
defined four distinct phases during the study: prebloom,
bloom, postbloom and autumnal (K. G. Simpson et al.,
Nutrient dynamics in the Amundsen Gulf and Cape Bathurst Polynya: 1. New production in spring inferred from
nutrient drawdown, submitted to Marine Ecology Progress
4 of 16
C07016
SIMPSON ET AL.: NUTRIENT CYCLING IN THE BEAUFORT SEA
Series, 2007). The timing of these phases was independently
confirmed by measurements of phytoplankton biomass
(chlorophyll a) (S. Brugel, personal communication,
2007). Temporal and spatial differences in phytoplankton
growth in each phase likely contributed to the variation in
C07016
nutrient concentration regionally and consequently introduced noise to the data.
[23] Nitrate and Si(OH)4 inventories were significantly
lower during the bloom and postbloom phases than during
the prebloom, reflecting net phytoplankton consumption
(Table 2). Silicic acid drawdown in the Amundsen Gulf
stations east of 128.35°W was most pronounced in the
postbloom period after the majority of the NO
3 had been
consumed. Seasonal changes in PO3
4 inventory were not
apparent. Ammonium concentration declined during the
bloom and then rose steadily from the spring to fall periods.
Nitrite concentration varied greatly, increasing threefold
from low prebloom levels to the end of the study. Significant changes were observed in DON concentrations east of
128.35°W and in DOP concentrations west of 128.35°W.
Elsewhere in the study area, average concentrations of DON
and DOP were generally higher during the period of active
algal growth than during the prebloom and autumnal
phases. There was no variation in urea concentration at this
coarse level of analysis.
3.3. Regional Variation in Nutrient Concentration
[24] Regional differences in nutrient content were apparent
in vertical profiles obtained during leg 8/9 when the spring
phytoplankton growth had subsided (Figure 3). Depthaveraged vertical profiles showed that the waters overlying
the Amundsen Gulf were depleted in NO
3 to approximately
and Si(OH)4 concentrations
25 m depth, whereas PO3
4
were detectable. Surface NO
3 concentrations were higher
and more variable in the Mackenzie Shelf region than in the
Gulf and the nitracline was shallower. These differences
were likely due to the input of NO
3 from the Mackenzie
River and high turbidity that restricted phytoplankton
growth to near the sea surface. Silicic acid was also
influenced by input from the river and was greatest during
the time when freshwater discharge was at or near its
content of the river water was
maximum. The low PO3
4
also evident by comparing the differences in the profiles in
the upper 20 m (Figure 3). Below the nutrient maxima, the
3
Amundsen Gulf samples were more enriched in NO
3 , PO4
and Si(OH) 4 than those from the Beaufort Sea and
deep Mackenzie Shelf stations by roughly 1.7 mmol L1,
0.12 mmol L1, and 6.2 mmol L1, respectively.
3.4. Regenerated N and DON
[25] The concentrations of NH+4 and urea were generally
higher in the surface waters than at depth (Figure 4). Most
samples contained less than 0.5 mmol NH+4 -N L1 and less
than 1 mmol urea-N L1. Average concentrations were 0.38
and 0.94 mmol N L1, respectively. Urea concentration
was the most variable, ranging from undetectable to
9.6 mmol N L1. High concentrations of NH+4 were also
present between 50 and 125 m depth and urea was at times
Figure 2. Nutrient concentration plotted against salinity
for all water samples obtained during legs 1, 2, 7, and 8/9 of
the CASES 2003– 2004 program. (a) Nitrate, (b) phosphate,
and (c) silicic acid. Color coding indicates depth strata:
green, 0 – 50 m; yellow, 50– 125 m; red, 125 – 200 m; gray,
200– 275 m; and blue, 275+ m. Insets show nutrient-salinity
relationships over an expanded salinity scale between 31.5
and 35.
5 of 16
SIMPSON ET AL.: NUTRIENT CYCLING IN THE BEAUFORT SEA
C07016
C07016
Table 1. Water Density, Salinity, and Concentrations of Nitrate, Phosphate, and Silicic Acid in the Nutrient Maxima of the Beaufort Sea/
Mackenzie Shelf and the Amundsen Gulf
Region
Nutrient
Concentration (mmol L1)
Depth (m)
Salinity
Sigma-T (st)
Gulf
Shelf
Gulf
Shelf
Gulf
Shelf
NO3
NO3
PO43
PO43
17.3
17.3
1.9
1.8
33.5
36.8
197.8
152.5
131.2
133.3
146.5
151.1
34.21
33.23
33.08
33.04
33.12
33.10
27.49
26.73
26.61
26.56
26.68
26.60
Si(OH)4
Si(OH)4
abundant at all depths except in the most saline, deep water.
The NH+4 concentration-salinity plot showed a significant
decrease in NH+4 concentration as salinity increased from 22
to 28 (p = 0.0182).
[26] Dissolved organic nitrogen concentration was largely
independent of salinity although it decreased significantly
from salinity 22 to 28 (p = 0.0029), as observed for NH+4
(Figure 5). Because urea was detected as part of the DON
pool during analysis, we subtracted its concentration from
the total to see if it influenced the DON-salinity relationship. The resulting plot differed little from that shown in
Figure 5, although the decrease in DON concentration as
salinity increased from 22 to 28 was more striking. As
reported below, urea concentration was proportional to the
DON concentration. At salinities less than 22, DON concentrations remained constant at about 5 mmol N L1,
suggesting a substantial contribution by the Mackenzie
River. This contribution was confirmed by measurements
Table 2. Average Integrated Nutrient Concentration ± Standard Deviation of Surface Waters With Salinity 31.6 or Lessa
Day of Year and Bloom Phase Designation
Nutrient
136 – 156 Prebloom
157 – 182 Bloom
183 – 220 Postbloom
273 – 330 Autumnal
NO3E
NO3W
NO3W *
108.6 ± 20.4 a
44.9 ± 20.5 b
48.1 ± 25.5
28.7 ± 40.9
31.9 ± 17.8 b
54.7 ± 30.8
55.2 ± 23.6
39.9 ± 19.6 b
70.4 ± 31.1
73.6 ± 39.9
NO2E
NO2W
NO2W *
1.8 ± 0.5 a
3.4 ± 1.0 b
2.5 ± 1.4 a
1.8 ± 1.5 a
3.7 ± 1.0 b
3.5 ± 1.0 a
3.7 ± 1.2 a
5.6 ± 1.18 c
6.0 ± 2.7 b
4.9 ± 2.1 b
PO43E
PO43W
PO43W *
28.5 ± 4.3
31.8 ± 7.0
27.9 ± 13.1
21.7 ± 14.2
28.3 ± 7.0
26.0 ± 6.8
24.6 ± 7.7
28.2 ± 6.6
33.3 ± 9.7
31.2 ± 10.2
Si(OH)4
Si(OH)4
Si(OH)4
321.3 ± 51.2 a
251.7 ± 79.4 a
163.4 ± 97.9 a
143.8 ± 122.3
179.2 ± 46.1 b
252.8 ± 115.4 ab
301.5 ± 193.2
172.0 ± 78.5 b
298.5 ± 109.7 b
295.2 ± 90.9
18.1 ± 11.7 a
4.3 ± 2.72 b
2.7 ± 1.84 a
3.0 ± 2.6 #
7.4 ± 3.7 b
12.4 ± 5.5 ab
11.7 ± 2.3
14.6 ± 5.1 a
16.8 ± 11.9 b
20.7 ± 13.0
182.7 ± 8.1 #
240.2 ± 55.6
251.2 ± 75.0
227.7 ± 51.2
241.3 ± 83.1
260.9 ± 77.9
293.4 ± 84.8
187.7 ± 84.5
252.3 ± 90.5
230.3 ± 58.7
117.2 ± 10.4 ab#
201.1 ± 42.8 a
216.4 ± 50.4
201.8 ± 32.4
200.8 ± 66.6 a
195.8 ± 30.7
201.8 ± 59.4
102.4 ± 32.7 b
213.9 ± 96.4
157.7 ± 74.1
26.2 ± 0.2 #
43.6 ± 18.5
56.9 ± 17.7 a
39.9 ± 31.1
46.2 ± 11.2
59.2 ± 19.7 a
57.5 ± 22.6
31.2 ± 6.6
29.0 ± 11.2 b
25.7 ± 8.6
7.5 ± 0.2 #
18.6 ± 6.7
23.0 ± 7.7 a
17.1 ± 15.5
18.9 ± 6.7
24.4 ± 13.8 a
20.0 ± 9.0
11.3 ± 6.8
9.1 ± 5.2 b
8.0 ± 4.6
25.7 ± 13.5 #
23.6 ± 13.7
48.8 ± 24.0
39.1 ± 17.4
21.9 ± 17.9
26.0 ± 15.1
23.2 ± 12.8
15.0 ± 10.5
52.8 ± 33.4
35.1 ± 24.5
NH4+
NH4+
NH4+
TDN
TDN
TDN
DON
DON
DON
TDP
TDP
TDP
DOP
DOP
DOP
E
W
W
*
E
W
W
*
E
W
W
*
E
W
W
*
E
W
W
*
E
W
W
Urea-N
Urea-N
Urea-N
*
E
W
W
*
Units are mmol m2. Data obtained during the study were pooled according to time of collection relative to the spring bloom (see text for details).
Asterisk indicates nutrient inventory of stations with bottom depth of 250 m (Beaufort Sea stations removed). Different letters (a, b, c) indicate significant
differences (ANOVA and Tukey p 0.05). Pound sign (#) indicates samples are limited, n = 2. Subscripts E and W denote the sampling location: E, east of
128.35°W (Amundsen Gulf); W, west of 128.35°W (Beaufort Sea/Mackenzie Shelf).
a
6 of 16
C07016
SIMPSON ET AL.: NUTRIENT CYCLING IN THE BEAUFORT SEA
C07016
taken near the eastern and western outflows, which
contained 2 –4 and 5 – 10 mmol DON N L1, respectively. Dissolved organic nitrogen concentrations were
variable but generally highest in surface waters and lower
at depth. Concentrations ranged from 0.2 – 35.8 mmol
DON N L1, and averaged 4.6 ± 3.7 mmol DON N
L1. The highest values (>10 mmol DON N L1) were
found at the shelf sites and within the Amundsen Gulf.
Waters deeper than 200 m contained 3.6 ± 2.64 mmol
DON N L1.
3.5. DOP Concentrations
[27] Vertical profiles of DOP concentration showed a
pronounced middepth enrichment at around 150– 200 m
that was absent in the DON plot (Figure 5). The high DOP
concentrations (averaging 0.76 ± 0.27 mmol DOP-P L1)
were associated with waters of salinity ca. 33.1 and were
roughly twofold greater than average surface levels. Below
the DOP maximum, concentrations declined to 0.65 mmol
DOP-P L1. Mixing lines between Bering Sea water and
deep Atlantic-derived water, and between Bering Sea water
and Arctic surface water were evident, but the data showed
considerable scatter (Figure 5b, inset).
[28] High variability in the low salinity samples made it
difficult to accurately estimate the river contribution of
DOP; however, regression analysis yielded an intercept at
zero salinity of 1 mmol DOP-P L1. This concentration
was higher than we measured in the Mackenzie River, but
consistent with values observed from the Mackenzie Delta
at salinity 15 – 25 [Emmerton, 2006]. Dissolved organic
phosphorus concentration ranged from detection limit to
2.3 mmol L1 and averaged 0.65 ± 0.38 mmol P L1.
Figure 3. Depth-averaged nutrient concentration and
standard deviation (SD) of (a) nitrate, (b) phosphate, and
(c) silicic acid. Results are for data from 19 stations in the
Amundsen Gulf region and 28 stations in the Beaufort Sea/
Mackenzie Shelf collected during 23 June to 5 August
2004. Standard deviation of depths greater than 800 m were
derived by combining all samples taken during CASES
2003 – 2004 (n = 7).
3.6. Water Mass Distribution of DOP
[29] We evaluated the contribution of the Bering water to
the subsurface DOP enrichment by examining the relationship between DOP concentration and N* in samples of 30–
33.1 salinity. Modified Bering Sea water contributes to the
upper halocline in the Arctic Ocean and is characterized by
a highly negative N* [Codispoti et al., 1991] that can
be used to trace its flow. A highly significant relationship
(p < 0.0001) existed between DOP concentration and N*
(Figure 6). The concentration of DOP was highest in the
most NO
3 -depleted (most negative N*) seawater samples
and lowest in the least NO
3 -depleted (least negative N*)
samples. Thus, Bering-derived water was more enriched in
DOP than either Arctic or deep Atlantic water (Figure 6).
The DON:DOP ratio also decreased with decreasing N*, as
expected from the DOP N* relationship and the apparent
constancy of DON concentration with depth (Figure 6). In
water of salinities between 33.1 and 34, DOP and DON
concentrations did not change and were independent of N*
(data not shown).
3.7. Vertical Profiles of DON and DOP
[30] Depth-averaged vertical profiles of urea and DON
were similar during all seasons (Figure 7). In these plots, the
DON concentration has been corrected to remove the
contribution of urea. As illustrated in Figures 4 and 5, both
N-containing compounds were abundant in the surface
waters and then declined with depth. Vertical profiles of
the average values showed a remarkable correspondence
7 of 16
C07016
SIMPSON ET AL.: NUTRIENT CYCLING IN THE BEAUFORT SEA
C07016
Figure 4. Property-salinity plots and depth profiles of (a, c) ammonium and (b, d) urea concentrations.
Color coding indicates depth strata: green, 0 – 50 m; yellow, 50– 125 m; red, 125– 200 m; gray, 200– 275 m;
and blue 275+ m.
between both nitrogen forms and revealed some subsurface
maxima, notably at 50 m and between 250 and 300 m depth
in the Beaufort Sea/Mackenzie Shelf and the Amundsen
Gulf (Figure 7). Regression analysis of urea-N against DON
showed a highly significant correlation (p < 0.0001) and a
slope of 0.31 mol urea N: mol DON N.
[31] Total dissolved nitrogen to total dissolved phosphorus ratio (TDN:TDP) increased from near 5 in surface
waters to 10 at 400 m depth (Figure 8). The low surface
value was consistent with a shallower concentration maxithan of NO
mum of PO3
4
3 and limited vertical mixing
during fall and winter. Surface water variability was attributed to regional differences in nutrient inventories and to
seasonal drawdown by phytoplankton that reduced DIN
concentration to zero. The DON:DOP ratio ranged from 2
to 200 in surface waters and averaged 4.9:1 at depth. In
surface waters (0 – 50 m), linear regression analysis of DON
on DOP concentration showed a highly significant relationship (p < 0.001) and a DON:DOP production/utilization
ratio of 17:1, r2 = 0.38 (data not shown). No discernable
trend between DON and DOP concentrations was found in
waters deeper than 50 m (but see Figure 6). Nitrate:phosphate ratios were typically low in the surface layer, between
0 and 8, gradually increased to 15 at 300 m, and remained
relatively constant below this depth.
3.8. Nutrient Consumption and Regeneration
[32] Nutrient-nutrient concentration plots were used to
infer consumption and regeneration ratios of the inorganic
nutrients. Samples were pooled according to depth strata
chosen on the basis of their temperature-salinity characteristics to correspond to different water masses (Figure 9).
Ratios (mol mol1) determined from the slopes of regression analysis (model 2) were significantly different from
one another (ANCOVA, p < 0.001). In surface waters (0 –
3
50 m), NO
3 and PO4 , and Si(OH)4 and NO3 were
consumed in the following proportions: 13.3 NO 3:
2
2
1 PO3
4 , r = 0.78 and 1.75 Si(OH)4:1 NO3 , r = 0.87,
while samples from depths of 50 to 125 m had ratios of
15.7:1, r2 = 0.49, and 2.11:1, r2 = 0.82, respectively. In the
3
deep water, below 275 m, the NO
3 : PO4 ratio was 7.6:1,
2
r = 0.33 and the Si(OH)4:NO3 ratio was 4.6:1, r2 = 0.60.
8 of 16
SIMPSON ET AL.: NUTRIENT CYCLING IN THE BEAUFORT SEA
C07016
C07016
Figure 5. Property-salinity plots and depth profiles of (a, c) dissolved organic nitrogen (DON) and
(b, d) dissolved organic phosphorus (DOP) concentrations. Insets in the property-salinity plots show the
DON and DOP-salinity relationships with an expanded salinity scale. Color coding indicates depth strata:
green, 0 – 50 m; yellow, 50– 125 m; red, 125– 200 m; gray, 200 –275 m; and blue, 275 + m.
[33] Remineralization ratios derived from regression analysis of nutrient and oxygen concentrations reveal two significantly distinct relationships (Figure 10). The relatively
shallow Pacific-derived water mass (50 – 200 m) had
3
remineralization ratios of 9.0
O2/NO
3 and O2/PO4
and 122.5, respectively. Deeper waters (275 m) had
3
ratios were 17.4
higher ratios (O2/NO
3 and O2/PO4
and 193.5, respectively) indicating that more oxygen was
utilized per mol of nutrient regenerated (Figure 10).
4. Discussion
4.1. Salinity-Nutrient Distributions
[34] The salinity-nutrient relationships reported here
follow well established patterns described elsewhere in the
Canadian Basin and Beaufort Sea [Jones and Anderson,
1986; Macdonald et al., 1987; Moore et al., 1992; McLaughlin
et al., 1996; Codispoti et al., 2005; Wang et al., 2006]. They
show nutrient maxima in the upper halocline that was
composed of water derived from the Pacific Ocean. Phos-
phate concentration peaked at salinity 33.04– 33.08 at roughly 130 m depth, shallower than the NO
3 and Si(OH)4 maxima
(Table 1). In the Amundsen Gulf, the NO
3 maximum
occurred in deeper more saline water than in the Mackenzie
Shelf and Beaufort Sea. As described below, we believe this
feature was created by remineralization of diatom production
exported from the overlying polynya.
[35] A number of mixing lines between the different
water masses were clearly evident. Of particular note was
the transition between the upper and lower halocline, which
concentrashowed a sharp decline in Si(OH)4 and PO3
4
tions, but not in NO
3 , as salinity increased. On the basis of
3
the NO
3 : PO4 and NO3 : Si(OH)4 ratios of these two water
masses it appears that the upper halocline has lost NO
3
through denitrification [Lehmann et al., 2005] so that it now
contains roughly the same concentration of NO
3 as the
lower halocline.
[36] The surface layer near the Mackenzie Shelf showed
strong evidence of Si(OH)4 input from the Mackenzie River
but little input of NO
3 . Salinity plots demonstrated a
9 of 16
C07016
SIMPSON ET AL.: NUTRIENT CYCLING IN THE BEAUFORT SEA
C07016
Figure 6. (a) Dissolved organic phosphorus (DOP) concentration and (b) the ratio of dissolved organic nitrogen to
dissolved organic phosphorus (DON/DOP) concentrations
regressed against N* in samples of salinity between 30 and
33.1. The equations of the linear regressions (model I) were
(Figure 6a) DOP = 0.097 N* 0.15 r2 = 0.16, p < 0.0001
and (Figure 6b) DON/DOP = 1.39 N* + 19.43 r2 = 0.17 p <
0.0001.
significant decrease in NH+4 concentration during the transition from river-influenced sites to full seawater between
salinity 22 to 28. A decrease in concentration of DON was
also observed over the same salinity range (Figure 5). Both
trends suggest the presence of a frontal zone where changes
in resource availability may cause the phytoplankton community to shift from phosphorus (or light) to nitrogen
limitation. Phosphate limitation is widespread in freshwater
ecosystems and within estuaries gradually gives way to
nitrogen limitation, which is typical of marine systems [Yin
et al., 2001]. We observed that surface waters of the Mackenzie
Shelf contained higher concentrations of NO
3 than the
Amundsen Gulf following the phytoplankton bloom. As N
became more limiting to phytoplankton growth in this
transitional zone the concentrations of NH+4 and DON would
be expected to decline as NO
3 became less available seaward. Similar salinity-nutrient gradients have been noted for
other Arctic river systems [Cauwet and Sidorov, 1996] and to
co-occur with dramatic changes in the plankton community
[Garneau et al., 2006; Parsons et al., 1988, 1989].
Figure 7. Depth-averaged nutrient concentrations and
standard deviations of dissolved organic nitrogen (DON)
and urea for (a) the Beaufort Sea/Mackenzie Shelf derived
from 37 vertical profiles (below 600 m, n = 3) and (b) the
Amundsen Gulf region derived from 55 vertical profiles.
Regression analysis (model I) of DON concentration versus
urea concentration (c) shows a strong positive relationship:
urea = 0.31 [DON-urea] 0.337, r2 = 0.65 p < 0.0001.
An outlier (labeled with an asterisk) was identified by Systat
and not included in the regression analysis.
10 of 16
C07016
SIMPSON ET AL.: NUTRIENT CYCLING IN THE BEAUFORT SEA
C07016
1996; Gordeev et al., 1996] and is considerably higher than
the typical 3 – 4 mmol N L1 that prevails in the deep ocean
[Berman and Bronk, 2003].
[38] High and variable concentration of urea reported
here is common in Arctic regions [Harrison et al., 1985;
but see Codispoti et al., 2005]. Concentrations are known to
range from undetectable to 13.0 mmol N L1 and often to
exceed those of NH+4 . Zooplankton excretion is thought to be
a major source of urea to marine environments and is
dependent on food quality [Bidigare, 1983]. In the Arctic,
urea has been hypothesized to replace NH+4 as a waste
byproduct possibly because it is less toxic than NH+4 to
producing organisms that reside in confined habitats within
sea ice [Conover et al., 1999]. Microbial oxidation of DON
is also known to contribute urea to the environment [Antia
et al., 1991; Berman et al., 1999; Conover and Gustavson,
1999; Conover et al., 1999; Berg and Jørgensen, 2006].
Our observations point to a close association between DON
and urea (Figure 7) and suggest that these compounds
Figure 8. Molar ratios of (a) total dissolved nitrogen and
phosphorus (TDN/TDP), (b) dissolved organic nitrogen and
phosphorus (DON/DOP), and (c) nitrate and phosphate
(NO3/PO43) plotted as a function of depth.
4.2. Urea and DON
[37] Dissolved organic nitrogen concentration averaged
about 5 mmol N L1 throughout the study area, similar to
values measured elsewhere in the Arctic [Wheeler et al.,
1997]. We observed much greater variability in DON
concentration than the earlier measurements, likely because
of the proximity of our stations to the river outflow, which is
expected to be a significant source of organic nitrogen to the
nearshore. Indeed, in other Arctic rivers DON concentration
averages 27.0 ± 5.0 mmol N L1 [Cauwet and Sidorov,
Figure 9. Nutrient-nutrient concentration relationships
delineated by depth: (a) phosphate against nitrate and
(b) silicic acid against nitrate. Color coding indicates depth
strata: green, 0 – 50 m; yellow, 50– 125 m; red, 125– 200 m;
gray, 200 – 275 m; and blue, 275+ m. Samples with low
NO3 or low PO43 concentrations (<1.5 mmol N L1 and
<0.5 mmol P L1) were excluded from analysis as were
samples with anomalous salinity signatures for the depth of
collection. Lines drawn through the data represent model II
geometric mean regressions: solid line, 0 – 50 m; dashed
line, 50– 125 m; dotted line, 275+ m.
11 of 16
C07016
SIMPSON ET AL.: NUTRIENT CYCLING IN THE BEAUFORT SEA
Figure 10. Oxygen-nutrient concentration relationships
delineated by depth: (a) oxygen against nitrate; (b) oxygen
against phosphate. Color coding indicates depth strata:
green, 0 – 50 m; yellow, 50– 125 m; red, 125– 200 m; gray,
200 –275 m; and blue, 275+ m. Lines drawn though the data
represent model II, geometric mean regressions. Shallow
samples (0 –50 m) were excluded from regression analysis.
The slopes of the regression in shallow waters (solid lines)
are O2/NO3 = 9.0 and O2/PO43 = 122.5. In deep
water (dashed lines), the slopes are O2/NO3 = 17.4 and
O2/PO43 = 193.5.
share a common source or that urea is produced from
DON. Indeed, some zooplankton species are known to
excrete between 5 and 10% of their DON waste as urea
[e.g., Mitamura and Saijo, 1980]. Numerous subsurface
peaks in concentration of both nitrogen pools were consistently observed in the Amundsen Gulf and the Beaufort
Sea/Mackenzie Shelf region, most significantly between
250 –300 m depth. Such features may represent persistent
aggregations of producing organisms, such as zooplankton
or fish.
4.3. Dissolved Organic Phosphorus
[39] The DOP concentrations reported here are the first
determined for the Arctic Ocean. In the near surface, values
were variable, ranging from undetectable to 2.3 mmol
DOP-P L1, and highest at the river-affected sites. Gener-
C07016
ally, DOP concentrations were comparable to those of PO3
4
and within the range reported for other ocean regions [Karl
and Björkman, 2002], including the Mackenzie River Delta
[Emmerton, 2006]. Analysis of water from the lower
reaches of the Lena River showed summer values of 1.2–
1.3 mmol DOP L1 and roughly 0.2 mmol DOP L1 within
parts of the Laptev Sea (chlorinity < 20) [Cauwet and
Sidorov, 1996]. The relationship between DOP and salinity
showed features similar to those seen with PO3
4 , particularly over the salinity range of 26– 32 where the average
concentration of both increased (p < 0.01). As argued
above, this increase in DOP concentration could reflect
the transition from phosphorus (or light) limited primary
production in the river-affected sites to nitrogen-limited
primary production in the marine-dominated sites off the
shelf.
[40] The maximum DOP concentration in the upper
halocline suggests that Pacific-derived waters may be a
source of the enrichment. Indeed, inverse correlation (p <
0.0001) between DOP and N* provides some support for
this hypothesis, because low N* is a signature of Beringderived water. The large amount of scatter in the relationship may be caused by local processes that modify the DOP
concentration and N* signature of the water as it moves
from the Bering Sea to the Amundsen Gulf. Supply and
decomposition of particulate matter could vary regionally
which would add DOP and DON to the Bering water
thereby modifying its composition. Mixing with water
above and below would also erode the N* signature.
[41] The question that naturally arises is how this water
may have acquired a high DOP concentration? One possible
answer is by phytoplankton production and release during
nutrient-limited growth. Phytoplankton in the Bering Sea
and over Arctic shelves are known to be limited by iron
[Peers et al., 2005] and nitrogen [Codispoti et al., 2005;
Tremblay et al., 2006] and may take up and assimilate
excess amounts of nonlimiting resources including PO3
4 .
Organic phosphorus may thus accumulate intracellularly
and be released either directly or indirectly by grazing and
decomposition. Dissolved organic nitrogen would not be
expected to be produced in Redfield proportions by Fe or
N-limited phytoplankton, consistent with the low DON/
DOP ratio in the water of the 33.1 salinity layer (Figure 6).
Regeneration of DOP from sinking particulate matter could
also produce the DOP enrichment.
4.4. Seasonal Cycle of Nutrients
[42] Measurements of the prebloom integrated nutrient
concentrations were confined to the eastern portion of the
study site where the ship overwintered (70°02.74 N
126°18.08 W). These measurements are likely representative of the entire study area, judging from the constancy of
calculated NO and PO tracers [Broecker, 1974] (data not
shown) integrated over the upper 50 m in the Amundsen
Gulf and Mackenzie Shelf/Beaufort Sea region. The prebloom nutrient inventories measured by Macdonald on the
Mackenzie Shelf [Macdonald et al., 1987] were higher than
we report, but they likely included measurements of high
nutrient concentrations in the nitracline, which were not
always included in our analysis. Recharging of the surface
nutrient inventory was limited by the strength of the
halocline and the rapid onset of ice that minimized wind
12 of 16
C07016
SIMPSON ET AL.: NUTRIENT CYCLING IN THE BEAUFORT SEA
mixing. The resulting surface values (0– 15 m) of NO
3
reached a maximum of 3.3 mmol L1 [Tremblay et al.,
2008] roughly 1/5th of the deepwater concentration.
[43] The concentrations of NO
3 and Si(OH)4 decreased
significantly following ice breakup and the onset of phytoplankton growth. Phosphate concentrations did not vary,
likely because dissolved values were well above those
required to sustain the biomass of phytoplankton produced.
At both the western and eastern sites, approximately 1/2 of
the NO
3 was consumed and its concentration increased (at
least in the west) as autumnal mixing ensued. The decline in
Si(OH)4 concentration appeared to be more rapid in the
western side, and like NO
3 , its inventory rose in the fall.
Large increases in water column NO
2 from the prebloom to
the autumnal phases were consistent at all sites. The appear
ance of NO
2 was correlated with the drawdown of NO3 in
the surface (Simpson et al., manuscript in preparation, 2008)
and was most pronounced at the depth of the nitracline.
Since NO
3 uptake and assimilation by phytoplankton were
fastest at this depth, we believe that the NO
2 was produced
as a byproduct of assimilatory NO
3 reduction.
[44] The fall and rise of the dissolved inorganic nutrient
concentrations was mirrored by reciprocal changes in the
dissolved organic pools. Dissolved organic nitrogen and
phosphorus concentrations were higher in the bloom and
postbloom phases than at other times, although only significantly so in the Amundsen Gulf. Similar inverse relationships between dissolved inorganic and organic nutrients
have been shown in short-term incubation experiments
[Bronk and Glibert, 1994; Bronk et al., 1994], and in time
series analysis of water column concentrations [Butler et al.,
1979; Maita and Yanada, 1990; Libby and Wheeler, 1997;
Bronk et al., 1998]. The latter studies suggest the seasonality in DON and DOP production is related to grazing or
secretion by phytoplankton and its breakdown to heterotrophic microbial reactions, consistent with the interpretation
presented here. We note that not all ocean regions, however,
exhibit this type of seasonal cycle [Hansell et al., 1993;
Hansell and Carlson, 2001].
[45] To examine the inverse relationship between inorganic and organic nutrients more quantitatively we compared the absolute changes in concentration between the
winter and summer periods. Given the broad spatial coverage in the data collection, this calculation can only be made
in a rough way considering the entire study site. The mean
integrated increase in DON concentration was roughly
84.6 ± 43.9 mmol m2 and in fair agreement with the
+
decline in NO
3 and NH4 during the same time (77.5 ±
2
31.3 mmol m ). Thus, to a first approximation, the
production of DON can account for the entire DIN consumed. Net incorporation of DIN into phytoplankton to
produce particulate organic nitrogen (PON) must be small if
this interpretation is correct. Phytoplankton growth
decreases DIN and increases DON and PON whereas
grazing decreases PON and releases DON. As might be
expected the system is strongly regenerating. We hypothesize that microbial deamination and oxidation reduces the
concentration of the labile DON during fall and winter to
the spring time values. The breakdown of DON may explain
the high concentrations of NH+4 in surface waters shortly
before ice break up (Simpson et al., manuscript in preparation, 2008).
C07016
[ 46 ] A similar mass balance of phosphorus cannot
be obtained from our data. We calculate that approximately
4.8 ± 1.9 mmol P m2 would be consumed by phytoplankton during the bloom on the basis of the decrease in
3
integrated NO
3 concentration and the NO3 :PO4 consump3
tion ratio. This decrease in PO4 concentration is too small
for us to detect given the error in our seasonal measurements (Table 2). The measured increase in DOP concentration, however, shows some seasonality, particularly in the
west where values were significantly greater during the
spring and summer than in the fall. A similar trend is
evident in the eastern sites as well; that is, the mean values
are greater during the active phytoplankton growth phase
than before and after. A rough calculation shows a seasonal
change in DOP of about 10 mmol P m2 throughout the
region. Although subsurface waters contain higher DOP
concentrations, they are unlikely to contribute to the seasonal surface enrichment because surface concentrations are
the lowest during the fall when mixing is greatest. At
present, we are unable to adequately explain this seasonal
increase in DOP.
4.5. Nutrient Consumption Ratios
1
[47] Nitrate and PO3
4 drawdown was 13.3:1 mol mol
in surface waters (upper 50 m) and lower than Redfield
stoichiometry, as reported in other Arctic regions influenced
by Pacific-derived water [Jones et al., 1998; Codispoti et
al., 2005]. A model 2 (geometric mean) regression of PO3
4
3
on NO
intercept
3 concentration yielded a positive PO4
(0.69 mmol L1). Surface Si(OH)4 concentration also de1
clined linearly with NO
3 , and roughly 3.8 mmol L
remained when the entire NO 3 reservoir had been
exhausted. Thus, NO
3 was the biomass-limiting nutrient,
consistent with the observations from the vertical profiles
obtained during leg 8/9 that showed complete consumption
of NO
3 in the surface layer (Figure 8).
[48] The decrease in Si(OH)4 concentration during the
bloom phase suggests that diatoms made up a portion of the
phytoplankton community. The slope of the sea-surface
Si(OH)4 to NO
3 regression, the utilization ratio, was
1.75 mol Si:1 mol N (Figure 9) within the range of values
seen in pure diatom cultures (range 0.28– 4.38; mean 0.95 ±
0.23) [Brzezinski, 1985] and roughly twice that observed in
North Water Polynya [Tremblay et al., 2002]. Diatoms are
well known to be important producers in Arctic habitats
[Booth and Horner, 1997; Lovejoy et al., 2002] that
dominate ice and pelagic environments during bloom
phases. They are a major export flux to the ocean interior
in other Arctic polynyas settling as intact cells, empty
frustules, resting spores and in the feces of zooplankton
[Michel et al., 2002].
4.6. Excess Nutrients in the Amundsen Gulf
[49] Deep waters of the Amundsen Gulf bear the signature of diatom export production. Elevated nutrient concentrations are found in this region compared to source waters
in the southern Beaufort Sea (Figure 3) and along the
western side of Banks Island [McLaughlin et al., 2004]
and are most
R pronounced for Si(OH)4. The Rintegrated
difference ( ð200500mÞ [Amundsen gulf] ð200500mÞ
[Beaufort Sea]) in nutrient inventories shows that NO
3 is
enriched by 512.02 ± 8.65 mmol m2, PO3
4 by 35.24 ±
13 of 16
C07016
SIMPSON ET AL.: NUTRIENT CYCLING IN THE BEAUFORT SEA
19.66 mmol m2, and Si(OH)4 by 1865.72 ± 254.15 mmol
m2. The ratios of the excess nutrient inventories vary like
3
the surface drawdown ratios for NO 3 and PO 4
(14.5 mols:1 mol), but are twofold higher for Si(OH)4 and
NO
3 (3.6 mols:1 mol). They are consistent with enhanced
export of primary production in the Amundsen Gulf and its
subsequent remineralization at depth. We note that the
higher Si:N ratio in the deep than in the near surface waters
implies shallower regeneration of N, an observation made
previously in the North Water Polynya by Michel et al.
[2002]. Sediment trap collections at the North Water site
contained particulates with Si:N ratios of 4.6:1, similar to
the deepwater enrichment described here and to the remineralization ratios calculated from the slope of the dissolved
nutrients in this layer (4.6:1) (Figure 9). Although we have
limited measurements, our data set also shows that some of
the Si(OH)4 remineralization occurs in or near the sediments
as the Si(OH)4 concentration is consistently highest in the
deepest samples.
[50] Oxygen concentrations in the Amundsen Gulf also
differed as expected if there has been more remineralization:
they were depleted relative to the Beaufort Sea by 7664 ±
830 mmol O2 m2. The deep nutrient excess suggests that
the Cape Bathurst Polynya, like the North Water Polynya,
may preferentially export Si(OH)4 to depth [Tremblay et al.,
2002]. The nutrient excess can also be expressed in terms of
C, assuming its assimilation in surface waters followed the
C:N ratio of 6.6 – 8.9 [Redfield et al., 1963; Daly et al.,
1999]. Although the timescale of deepwater replacement is
unknown and so a rate of export cannot be derived, the
2
NO
3 enrichment is equivalent to 40– 50 g C m , roughly
the amount of annual new production in the overlying
polynya (Simpson et al., manuscript in preparation, 2008).
4.7. Remineralization Ratios
[51] The stoichiometry of remineralization deduced from
balancing the equation of organic matter production using
traditional Redfield ratios shows a Corg/N/P/O2 of 106/16/
1/138 (O2/N = 9) [Anderson and Sarmiento, 1994]. The
equation is thought to generally represent organic matter
decomposition in the world ocean although recent analyses
document its latitudinal variations [Li and Peng, 2002] and
suggested changes with depth [Boulahdid and Minster,
1989; Shaffer et al., 1999]. Analysis of our data shows that
organic matter is remineralized as predicted with O2/N and
O2/P of 9.0 and 122.5, respectively. The slightly lower than
expected O2/P ratio in our data set (depth 50 – 200 m)
suggests a shallower PO3
4 maximum and preferential remineralization of phosphorus compared to NO
3 , as observed
(Figure 3 and Table 1). At depth (>250 m), the O2/N and
O2/P ratios increased remarkably to 17.4 and 193.5. We are
less confident in the accuracy of these ratios; however,
because they were derived from a smaller data set than the
shallow samples and the type-2 regression explained less of
the variation [r2 = 0.44 (O2/N) and r2 = 0.74 (O2/P)].
[52] An alternative approach to calculate the deepwater
remineralization is to use the integrated O2 deficit and the
nutrient excess in the Amundsen Gulf deep waters. The
O2/P and O2/N ratios obtained in this way (217.5 ±
123.6 and 15.0 ± 1.64) are roughly equal to the ratios
calculated from the oxygen-nutrient regressions suggesting
C07016
that the high values are correct. They may reflect an unusual
composition of the sinking POM, possibly caused by the
high lipid content [Smith et al., 1997] or C:N ratio [Daly et
al., 1999] of Arctic phytoplankton.
5. Conclusions
3
[53] Salinity-nutrient relationships showed NO
3 , PO4
and Si(OH)4 maxima in the upper halocline layer at salinity
ca. 33.1 throughout the CASES study area. Concentrations
of all nutrients were most variable in the surface layer,
reflecting the seasonal drawdown by phytoplankton and
discharge from the Mackenzie River. Over the salinity range
of 22– 28, ammonium and DON concentrations decreased
significantly possibly indicating a transitional zone where
phytoplankton became progressively more N-limited. Dissolved organic phosphorus and phosphate concentrations in
surface waters increased over a slightly higher salinity
gradient (26 –32).
[54] Seasonal drawdown of NO
3 and Si(OH)4 concentrations were detected on the Mackenzie Shelf and in the
Amundsen Gulf and coincided with the onset of phytoplankton growth. Ammonium concentration also declined
during the phytoplankton bloom and then increased during
+
the remainder of the year. The decrease in NO
3 and NH4
concentrations was accompanied by a roughly equivalent
increase in DON concentration. Water column integrated
DOP concentration was also higher during the bloom and
postbloom periods compared to the early spring and fall.
[55] Dissolved organic phosphorus concentration showed
a middepth enrichment in upper halocline waters of salinity
33.1. Within this layer the concentration of DOP was
inversely correlated with N*, suggesting that water originating from the Bering Sea may be an important source of
DOP to the Arctic.
[56] In surface waters, nutrients were consumed in near
Redfield proportions with N:P = 13.1:1 and Si:N = 1.75:1
indicative of diatom production. Urea concentrations were
typically high and covaried with the concentration of DON,
suggesting a common source or that urea was derived from
the breakdown of the DON pool.
[57] Vertical profiles of nutrients showed regional vari3
ability. Higher concentrations of NO
3 , PO4 and Si(OH)4
(and lower concentration of O2) were present in deep water
of the Amundsen Gulf compared to waters of similar depth
in the Beaufort Sea. Our analysis suggests that the residence
time of waters in the Amundsen Gulf is sufficiently long to
allow the accumulation of nutrients derived from remineralization of exported diatom material. Regeneration ratios
appeared to vary with depth and were approximately 9 mol
3
O2/mol NO
3 and 122.5 mol O2/mol PO4 in shallow
Pacific-derived waters (halocline layer) and 17.4 and 193.5
in deep Atlantic waters (Atlantic layer), respectively.
[58] Acknowledgments. We thank the officers and crew of the CCGS
Amundsen (Canadian Coast Guard) for their support and assistance. Two
reviewers provided useful comments that improved the manuscript. D. Raab
helped collect and analyze nutrient samples. This study was supported by
grants to NMP from the Natural Science and Engineering Council of
Canada, and is a contribution to the Canadian Arctic Shelf Exchange Study
(CASES). Figure 1 was produced with the Ocean-Data-View Software
(R. Schlitzer; http://www.awi-bremerhaven.de/GEO/ODV).
14 of 16
C07016
SIMPSON ET AL.: NUTRIENT CYCLING IN THE BEAUFORT SEA
References
Aagard, K., and E. C. Carmack (1994), The Arctic Ocean and climate: A
perspective, in The Polar Oceans and Their Role in Shaping the Global
Environment, Geophys. Monogr. Ser., vol. 85, edited by O. M. Johannessen,
R. D. Muench, and J. E. Overland, pp. 4 – 20, AGU, Washington, D. C.
Anderson, L. A., and J. L. Sarmiento (1994), Redfield ratios of remineralization determined by nutrient data analysis, Global Biogeochem. Cycles,
8, 65 – 80, doi:10.1029/93GB03318.
Antia, N. J., P. J. Harrison, and L. Oliveira (1991), The role of dissolved
organic nitrogen in phytoplankton nutrition, cell biology and ecology,
Phycologia, 30, 1 – 89.
Arnell, N. W. (2005), Implications of climate change for freshwater inflows
to the Arctic Ocean, J. Geophys. Res., 110, D07105, doi:10.1029/
2004JD005348.
Arrigo, K. R., and G. L. van Dijken (2004), Annual cycles of sea ice and
phytoplankton in Cape Bathurst Polynya, southeastern Beaufort Sea,
Canadian Arctic, Geophys. Res. Lett., 31, L08304, doi:10.1029/
2003GL018978.
Barber, D. G., and J. M. Hanesiak (2004), Meteorological forcing of sea ice
concentrations in the southern Beaufort Sea over the period 1979 to 2000,
J. Geophys. Res., 109, C06014, doi:10.1029/2003JC002027.
Bauch, D., P. Schlosser, and R. G. Fairbanks (1995), Freshwater balance
and the sources of deep and bottom waters in the Arctic Ocean inferred
from the distribution of H18
2 O, Prog. Oceanogr., 35, 53 – 80, doi:10.1016/
0079-6611(95)00005-2.
Berg, G. M., and N. O. G. Jørgensen (2006), Purine and pyrimidine metabolism by estuarine bacteria, Aquat. Microbial Ecol., 42, 215 – 226,
doi:10.3354/ame042215.
Berman, T., and D. A. Bronk (2003), Dissolved organic nitrogen: a
dynamic participant in aquatic ecosystems, Aquat. Microbial Ecol., 31,
279 – 305, doi:10.3354/ame031279.
Berman, T., C. Bechemin, and S. Y. Maestrini (1999), Release of ammonium and urea from dissolved organic nitrogen in aquatic ecosystems,
Aquat. Microbial Ecol., 16, 295 – 302, doi:10.3354/ame016295.
Bidigare, R. R. (1983), Nitrogen excretion by marine zooplankton, in
Nitrogen in the Marine Environment, edited by E. J. Carpenter and
D. G. Capone, pp. 385 – 409, Acad. Press, New York.
Booth, B. C., and R. A. Horner (1997), Microalgae on the Arctic Ocean
Section, 1994: Species abundance and biomass, Deep Sea Res., Part II,
44, 1607 – 1622, doi:10.1016/S0967-0645(97)00057-X.
Boulahdid, M., and J. F. Minster (1989), Oxygen consumption and nutrient
regeneration ratios along isopycnal horizons in the Pacific Ocean, Mar.
Chem., 26, 133 – 153, doi:10.1016/0304-4203(89)90057-1.
Broecker, W. S. (1974), ‘‘NO’’, a conservative water-mass tracer, Earth
Planet. Sci. Lett., 23, 100 – 107, doi:10.1016/0012-821X(74)90036-3.
Bronk, D. A., and P. M. Glibert (1994), The fate of the missing 15N differs
among marine systems, Limnol. Oceanogr., 39, 189 – 195.
Bronk, D. A., P. M. Glibert, and B. B. Ward (1994), Nitrogen uptake,
dissolved organic nitrogen release, and new production, Science, 265,
1843 – 1846, doi:10.1126/science.265.5180.1843.
Bronk, D. A., P. M. Glibert, T. C. Malone, S. Banahan, and E. Sahlsten
(1998), Inorganic and organic nitrogen cycling in Chesapeake Bay: Autotrophic versus heterotrophic processes and relationships to carbon flux,
Aquat. Microbial Ecol., 15, 177 – 189, doi:10.3354/ame015177.
Brzezinski, M. A. (1985), The Si:C:N ratio of marine diatoms: Interspecific
variability and the effect of some environmental variables, J. Phycol., 21,
347 – 357.
Butler, E. I., S. Knox, and M. I. Liddicoat (1979), The relationship between
inorganic and organic nutrients in sea water, J. Mar. Biol. Assoc. U. K.,
59, 239 – 250.
Carmack, E. C., and E. A. Kulikov (1998), Wind-forced upwelling and
internal Kelvin wave generation in Mackenzie Canyon, Beaufort Sea,
J. Geophys. Res., 103, 18,447 – 18,458, doi:10.1029/98JC00113.
Carmack, E. C., R. W. Macdonald, and J. E. Papadakis (1989), Water mass
structure and boundaries in the Mackenzie Shelf estuary, J. Geophys.
Res., 94, 18,043 – 18,055, doi:10.1029/JC094iC12p18043.
Cauwet, G., and I. Sidorov (1996), The biogeochemistry of the Lena River:
Organic carbon and nutrients distribution, Mar. Chem., 53, 211 – 227,
doi:10.1016/0304-4203(95)00090-9.
Codispoti, L. A., G. E. Friederich, C. M. Sakamoto, and L. I. Gordon
(1991), Nutrient cycling and primary production in the marine systems
of the Arctic and Antarctic, J. Mar. Syst., 2, 359 – 384, doi:10.1016/09247963(91)90042-S.
Codispoti, L. A., C. Flagg, V. Kelly, and J. H. Swift (2005), Hydrographic
conditions during the 2002 SBI process experiments, Deep Sea Res., Part
II, 52, 3199 – 3226, doi:10.1016/j.dsr2.2005.10.007.
Conover, R. J., and K. R. Gustavson (1999), Sources of urea in Arctic seas:
Zooplankton metabolism, Mar. Ecol. Prog. Ser., 179, 41 – 54,
doi:10.3354/meps179041.
C07016
Conover, R. J., N. Mumm, P. Bruecker, and S. MacKenzie (1999), Sources
of urea in arctic seas: Seasonal fast ice?, Mar. Ecol. Prog. Ser., 179, 55 –
69, doi:10.3354/meps179055.
Cooper, L. W., T. E. Whitledge, J. M. Grebmeier, and T. J. Weingartner
(1997), The nutrient, salinity and stable oxygen isotope composition of
Bering and Chukchi seas waters in and near the Bering Strait, J. Geophys.
Res., 102, 12,563 – 12,573, doi:10.1029/97JC00015.
Daly, K., D. W. R. Wallace, W. Smith Jr., A. Skoog, R. J. Lara, M. Gosselin,
E. Falck, and P. L. Yager (1999), Non-Redfield carbon and nitrogen cycling
in the Arctic: Effects of ecosystem structure and dynamics, J. Geophys.
Res., 104, 3185 – 3199, doi:10.1029/1998JC900071.
Dickson, R. R., T. J. Osborn, J. W. Hurrell, J. Meincke, J. Blindheim,
B. Adlandsvik, T. Vinje, G. Alekseev, and W. Maslowski (2000), The
Arctic Ocean response to the north Atlantic Oscillation, J. Clim., 13,
2671 – 2696, doi:10.1175/1520-0442(2000)013<2671:TAORTT>
2.0.CO;2.
Emmerton, C. A. (2006), Downstream nutrient changes through the Mackenzie River Delta and estuary, western Canadian Arctic, M. Sci. thesis.
181 pp., Simon Fraser Univ., B. C., Canada.
Garneau, M.-E., W. F. Vincent, L. Alonso-Saez, Y. Gratton, and C. Lovejoy
(2006), Prokaryotic community structure and heterotrophic production in
a river-influenced coastal Arctic ecosystem, Aquat. Microbial Ecol., 42,
27 – 40, doi:10.3354/ame042027.
Goeyens, L., N. Kindermans, M. Abu Yusuf, and M. Elskens (1998), A
room temperature procedure for the manual determination of urea in
seawater, Estuar. Coast. Shelf Sci., 47, 415 – 418, doi:10.1006/
ecss.1998.0357.
Gordeev, V. V., J. M. Martin, I. S. Sidorov, and M. V. Sidorova (1996), A
reassessment of the Eurasian river input of water, sediment, major elements and nutrients to the Arctic Ocean, Am. J. Sci., 296, 664 – 691.
Grasshoff, K. (1999), Methods of Seawater Analysis, 3rd ed., pp. 159 – 228,
Wiley-VCH, Weinheim, New York.
Gruber, N., and J. L. Sarmiento (1997), Global patterns of marine nitrogen
fixation and denitrification, Global Biogeochem. Cycles, 11, 235 – 266,
doi:10.1029/97GB00077.
Hansell, D. A., and C. A. Carlson (2001), Biogeochemistry of total organic
carbon and nitrogen in the Sargasso Sea: Control by convective overturn,
Deep Sea Res., Part I, 48, 1649 – 1667, doi:10.1016/S0967-0645(00)
00153-3.
Hansell, D. A., P. M. Williams, and B. B. Ward (1993), Measurements of
DOC and DON in the southern California Bight using oxidation by hightemperature combustion, Deep Sea Res., Part I, 40, 219 – 234,
doi:10.1016/0967-0637(93)90001-J.
Harrison, W. G., E. J. Head, R. J. Conover, A. R. Longhurst, and D. D.
Sameoto (1985), The distribution and metabolism of urea in the eastern
Canadian Arctic, Deep Sea Res., Part A, 32, 23 – 42, doi:10.1016/01980149(85)90015-9.
Intergovernmental Panel on Climate Change (IPCC) (2001), Climate
Change 2001: The Scientific Basis. Contributions of Working Group I
to the Third Assessment Report of the Intergovernmental Panel on
Climate Change (IPCC), edited by J. T. Houghton et al., 944 pp.,
Cambridge Univ. Press, UK.
Jones, E. P., and L. G. Anderson (1986), On the origin of the chemical
properties of the Arctic Ocean halocline, J. Geophys. Res., 91, 10,759 –
10,767, doi:10.1029/JC091iC09p10759.
Jones, E. P., L. G. Anderson, and J. H. Swift (1998), Distribution of Atlantic and Pacific waters in the upper Arctic ocean: Implications for circulation, Geophys. Res. Lett., 25, 765 – 768, doi:10.1029/98GL00464.
Karl, D. M., and K. M. Björkman (2002), Dynamics of DOP, in Biogeochemistry of Marine Dissolved Organic Matter, edited by D. A. Hansell
and C. A. Carlson, pp. 249 – 366, Acad. Press, New York.
Lehmann, M. F., D. M. Sigman, D. C. McCorkle, B. G. Brunelle,
S. Hoffmann, M. Kienast, G. Cane, and J. Clement (2005), Origin of
the deep Bering Sea nitrate deficit: Constraints from the nitrogen and
oxygen isotopic composition of water column nitrate and benthic nitrate
fluxes, Global Biogeochem. Cycles, 19, GB4005, doi:10.1029/
2005GB002508.
Li, Y.-H., and T.-H. Peng (2002), Latitudinal change of remineralization
ratios in the oceans and its implications for nutrient cycles, Global Biogeochem. Cycles, 16(4), 1130, doi:10.1029/2001GB001828.
Libby, P. S., and P. A. Wheeler (1997), Particulate and dissolved organic
nitrogen in the central and eastern equatorial Pacific, Deep Sea Res., Part
I, 44, 345 – 361, doi:10.1016/S0967-0637(96)00089-1.
Lovejoy, C., L. Legendre, and N. M. Price (2002), Prolonged diatom
blooms and microbial food web dynamics: Experimental results from
an Arctic polynya, Aquat. Microbial Ecol., 29, 267 – 278, doi:10.3354/
ame029267.
Macdonald, R. W., M. E. McFarland, S. J. De Mora, D. M. Macdonald, and
W. K. Johnson (1978), Oceanographic Data Report, Amundsen Gulf,
Pac. Mar. Sci. Rep. 78-10, Inst. of Ocean Sci., Sidney, Canada.
15 of 16
C07016
SIMPSON ET AL.: NUTRIENT CYCLING IN THE BEAUFORT SEA
Macdonald, R. W., C. S. Wong, and P. E. Erickson (1987), The distribution
of nutrients in the southeastern Beaufort Sea: Implications for water
circulation and primary production, J. Geophys. Res., 92, 2939 – 2952,
doi:10.1029/JC092iC03p02939.
Macdonald, R. W., E. C. Carmack, F. A. McLaughlin, K. Iseki, D. M.
Macdonald, and M. C. O’Brien (1989), Composition and modification
of water masses in the Mackenzie Shelf Estuary, J. Geophys. Res., 94,
18,057 – 18,070, doi:10.1029/JC094iC12p18057.
Maita, Y., and M. Yanada (1990), Vertical distribution of total dissolved
nitrogen and dissolved organic nitrogen in seawater, Geochem. J., 24,
245 – 254.
McLaughlin, F. A., E. C. Carmack, R. W. Macdonald, and J. K. Bishop
(1996), Physical and geochemical properties across the Atlantic/Pacific
water mass front in the southern Canadian Basin, J. Geophys. Res., 101,
1183 – 1197, doi:10.1029/95JC02634.
McLaughlin, F. A., E. C. Carmack, and R. W. Macdonald (2002), The Canada
Basin, 1989 – 1995: Upstream events and far-field effects of the Barents
Sea, J. Geophys. Res., 107(C7), 3082, doi:10.1029/2001JC000904.
McLaughlin, F. A., E. C. Carmack, R. W. Macdonald, H. Melling, J. H.
Swift, P. A. Wheeler, B. F. Sherr, and E. B. Sherr (2004), The joint roles
of Pacific and Atlantic-origin waters in the Canada Basin, 1997 – 1998,
Deep Sea Res., Part I, 51, 107 – 128, doi:10.1016/j.dsr.2003.09.010.
Michel, C., M. Gosselin, and C. Nozais (2002), Preferential sinking export
of biogenic silica during the spring and summer in the North Water
Polynya (northern Baffin Bay): Temperature or biological control?,
J. Geophys. Res., 107(C7), 3064, doi:10.1029/2000JC000408.
Mitamura, O., and Y. Saijo (1980), Urea supply from decomposition and
excretion of zooplankton, J. Oceanogr. Soc. Jpn., 36, 121 – 125,
doi:10.1007/BF02312097.
Moore, R. M., H. Melling, and K. R. Thompson (1992), A description of
water types on the Mackenzie Shelf of the Beaufort Sea during winter,
J. Geophys. Res., 97, 12,607 – 12,618, doi:10.1029/92JC00842.
Parsons, T. R., D. G. Webb, H. Dovey, R. Haigh, M. Lawrence, and G. E.
Hopky (1988), Production studies in the Mackenzie River – Beaufort Sea
estuary, Polar Biol., 8, 235 – 239, doi:10.1007/BF00443455.
Parsons, T. R., D. G. Webb, B. E. Rokeby, M. Lawrence, G. E. Hopky, and
D. B. Chiperzak (1989), Autotrophic and heterotrophic production in the
Mackenzie River/Beaufort Sea Estuary, Polar Biol., 9, 261 – 266,
doi:10.1007/BF00263774.
Peers, G., S.-A. Quesnel, and N. M. Price (2005), Copper requirements for
iron acquisition and growth of coastal and oceanic diatoms, Limnol.
Oceanogr., 50, 1149 – 1158.
Redfield, A. C., B. H. Ketchum, and F. A. Richards (1963), The influence
of organisms on the composition of sea-water, in The Sea, vol. 2, edited
by M. N. Hill, pp. 26 – 77, Intersci., New York.
Serreze, M. C., J. A. Maslanik, T. A. Scambos, F. Fetterer, J. Stroeve,
K. Knowles, C. Fowler, S. Drobot, R. G. Barry, and T. M. Haran
(2003), A record minimum Arctic sea ice extent and area in 2002, Geophys. Res. Lett., 30(3), 1110, doi:10.1029/2002GL016406.
Shaffer, G., J. Bendtsen, and O. Ulloa (1999), Fractionation during remineralization of organic matter in the ocean, Deep Sea Res., Part I, 46,
185 – 204, doi:10.1016/S0967-0637(98)00061-2.
C07016
Shiklomanov, A. I., R. B. Lammers, B. J. Peterson, and C. J. Vörösmarty
(2000), The dynamics of river water inflow to the Arctic Ocean, in The
Freshwater Budget of the Arctic Ocean: Proceedings of the NATO
Advanced Research Workshop, edited by E. L. Lewis et al., pp. 281 –
296, Kluwer, Dortrecht.
Smith, R. E. H., M. Gosselin, G. Kattner, L. Legendre, and S. Pesant
(1997), Biosynthesis of macromolecular and lipid classes by phytoplankton in the Northeast Water Polynya, Mar. Ecol. Prog. Ser., 147, 231 – 242,
doi:10.3354/meps147231.
Soden, B. J., and I. M. Held (2006), An assessment of climate feedbacks in
coupled ocean-atmosphere models, J. Clim., 19, 3354 – 3360,
doi:10.1175/JCLI3799.1.
Sokal, R. R., and F. J. Rohlf (1995), Biometry, the Principles and Practice
of Statistics in Biological Research, 3rd ed., 887 pp., W. H. Freeman,
New York.
Tremblay, J.-É., Y. Gratton, E. C. Carmack, C. D. Payne, and N. M. Price
(2002), Impact of the large-scale Arctic circulation and the North Water
Polynya on nutrient inventories in Baffin Bay, J. Geophys. Res., 107(C8),
3112, doi:10.1029/2000JC000595.
Tremblay, J.-É., C. Michel, K. A. Hobson, M. Gosselin, and N. M. Price
(2006), Bloom dynamics in early opening waters of the Arctic Ocean,
Limnol. Oceanogr., 51, 900 – 912.
Tremblay, J.-É., K. G. Simpson, J. Martin, L. Miller, Y. Gratton, D. G.
Barber, and N. M. Price (2008), Vertical stability and the annual
dynamics of nutrients and chlorophyll fluorescence in the coastal, southeast Beaufort Sea, J. Geophys. Res., doi:10.1029/2007JC004547, in press.
Valderrama, J. C. (1981), The simultaneous analysis of total nitrogen and
total phosphorus in natural waters, Mar. Chem., 10, 109 – 122,
doi:10.1016/0304-4203(81)90027-X.
Wang, D., S. M. Henrichs, and L. Guo (2006), Distributions of nutrients,
dissolved organic carbon and carbohydrates in the western Arctic Ocean,
Cont. Shelf Res., 26, 1654 – 1667, doi:10.1016/j.csr.2006.05.001.
Wheeler, P. A., J. M. Watkins, and R. L. Hansing (1997), Nutrients, organic
carbon, and organic nitrogen in the upper water column of the Arctic
Ocean: implication for the sources of dissolved organic carbon, Deep Sea
Res., Part II, 44, 1571 – 1592, doi:10.1016/S0967-0645(97)00051-9.
Wu, B., J. Wang, and R. Zhang (2004), Effects of intraseasonal variations of
the Arctic Oscillation on the Barents Sea, Polar Meteorol. Glaciol., 18,
82 – 95.
Yin, K., P.-Y. Qian, M. Wu, J. C. Chen, L. Huang, X. Song, and W. Jian
(2001), Shift from P to N limitation of phytoplankton growth across the
Pearl River estuarine plume during summer, Mar. Ecol. Prog. Ser., 221,
17 – 28, doi:10.3354/meps221017.
Y. Gratton, Institut national de la recherche scientifique-Eau, Terre et
Environnement, 490 de la Couronne, Québec City, QC G1K 9A9, Canada.
N. M. Price and K. G. Simpson, Department of Biology, McGill
University, 1205 Avenue Docteur Penfield, Montréal, QC H3A 1B1,
Canada. ([email protected])
J-É. Tremblay, Département de Biologie, Université Laval, Pavillon
Alexandre-Vachon, Québec City, QC G1K 7P4, Canada.
16 of 16