On the importance of freshwater fluxes from the Arctic ...
1
On the importance of freshwater fluxes from the Arctic Ocean into the North Atlantic: the Nordic Seas versus Canadian Arctic Archipelago Wieslaw Maslowski 1 , Jaclyn Clement Kinney 1 , Jaromir Jakacki 2 1 Naval Postgraduate School, Dept. of Oceanography, Monterey, CA, USA 2 Institute of Oceanology, Polish Academy of Sciences, Sopot, Poland ABSTRACT We use a high resolution coupled ice-ocean model of the Pan-Arctic region forced with realistic atmospheric data to investigate the variability of sea ice and liquid freshwater fluxes from the Arctic Ocean into the North Atlantic during 1979-2004. Model results are analyzed to compare the relative contribution of the total combined liquid and solid freshwater flux through the two main pathways: Fram-Denmark Strait (FDS) and the Canadian Arctic Archipelago-Davis-Hudson Strait (CAADHS). Our results suggest the relative importance of the freshwater flux through CAADHS into the Labrador Sea. This implies the need for ocean models to adequately represent mass and property fluxes through the narrow and shallow passages of the Canadian Archipelago and Davis and Hudson Straits. We argue that this requirement must be satisfied to advance studies of the Atlantic Meridional Overturning Circulation (MOC) and especially its variability. Given the recent record sea ice melt in the Arctic Ocean, it is critical that global ocean and climate models improve their skill in simulating and predicting effects of changing exports from the Arctic Ocean into the North Atlantic. This talk will outline a possible approach to satisfy such requirements. Table 1. 24-year mean volume fluxes for various sections Mean Vol. Trans. (Sv) Net Into AO Out of AO Bering Strait 0.649 0.725 -0.076 Davis Strait -1.571 0.934 -2.505 Fram Strait -2.346 4.741 -7.088 Barents Sea Opening 3.264 5.199 -1.934 Sum -0.004 11.599 -11.603 MODEL DESCRIPTION In an effort to obtain past estimates of oceanic fluxes through various passages controlling oceanic flow between the Arctic and the lower latitude oceans, we analyze results from a high-resolution coupled ice-ocean model of the Pan-Arctic region. The model domain extends from ~30 o N in the North Pacific through the Bering Sea, Arctic Ocean, Nordic Seas, into the North Atlantic to ~45 o N. At 9-km and 45-level resolution, the model allows realistic exchanges through important narrow passages. The model has been forced with the atmospheric fields from European Centre for Medium-range Weather Forecast (ECMWF), including 1979-1993 reanalysis and 1994-2004 operational products. Additional details on the model are provided elsewhere (Maslowski et al. 2004; Clement et al. 2005). Volume, heat, salt, and freshwater fluxes are calculated from this integration to allow comparison with historical and future estimates. Artificial North American Channel Canada Alaska Greenland Russia China Europe Bering Strait Fram Strait Smith Strait Lancaster Sound Davis Strait Cape Farewell Denmark Strait Figure 1. Model domain showing the locations of various sections used for frehwater flux analysis. Barents Sea Opening Hudson Bay Mouth Pre-Labrador Labrador Sea Central Labrador ACKNOWLEDGEMENTS We would like to thank the National Science Foundation support of this research. Additonal support has been provided by the Department of Energy (DOE), National Aeronautic and Space Administration (NASA). Computer resources were provided by the Arctic Region Supercomputing Center (ARSC), Army Engineering and Research Development Center (ERDC), and Army High Performance Computing Research Center (AHPCRC) through the Department of Defense High Performance Computer Modernization Program (HPCMP). 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 0 2 4 6 8 10 12 Time Freshwater Transport (mSv) Cape Farewell Freshwater Fluxes 2.342 Reference salinity = 34.8 FW net FW East FW West FW mean 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 5 10 15 20 25 30 35 Time Freshwater Transport (mSv) Denmark Strait Freshwater Fluxes 11.429 Reference salinity = 34.8 FW net FW North FW South FW mean Figure 3. Monthly mean freshwater flux through various sections shown in Figure 1. A thick black line represents a 13-month running mean of net flux. 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 0 5 10 15 20 25 30 35 Time Freshwater Transport (mSv) Fram Strait Freshwater Fluxes 10.638 Reference salinity = 34.8 FW net FW North FW South FW mean 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 10 20 30 40 50 60 70 80 90 100 Time Freshwater Transport (mSv) Pre-Labrador Freshwater Fluxes 61.642 Reference salinity = 34.8 FW net FW North FW South FW mean 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 0 5 10 15 20 25 30 Time Freshwater Transport (mSv) Hudson Bay Mouth Freshwater Fluxes 4.069 Reference salinity = 34.8 FW net FW East FW West FW mean 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 0 20 40 60 80 100 120 Time Freshwater Transport (mSv) Central Labrador Freshwater Fluxes 63.896 Reference salinity = 34.8 FW net FW North FW South FW mean 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 10 20 30 40 50 60 70 80 90 100 Time Freshwater Transport (mSv) Davis Strait Freshwater Fluxes 60.772 Reference salinity = 34.8 FW net FW North FW South FW mean 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 10 20 30 40 50 60 70 80 90 Time Freshwater Transport (mSv) Lancaster Sound Freshwater Fluxes 48.299 Reference salinity = 34.8 FW net FW East FW West FW mean 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 5 10 15 20 Time Freshwater Transport (mSv) Smith Sound Freshwater Fluxes 15.65 Reference salinity = 34.8 FW net FW North FW South FW mean Freshwater Flux 3.971 67.867 REFERENCES Clement, J. L., W. Maslowski, L. Cooper, J. Grebmeier, W. Walczowski, 2005. Ocean circulation and exchanges through the northern Bering Sea - 1979-2001 model results. Deep Sea Research II, 52, 3509-3540, doi: 10.1016/j.dsr2.2005.09.010. Cuny, J., P.B. Rhines, P.P. Niiler, and S. Bacon, 2002. Labrador Sea boundary currents and the fate of the Irminger Sea Water. J. of Phys. Ocean., 32, 627-647. Maslowski, W., Marble, D., Walczowski, W., Schauer, U., Clement, J. L., Semtner, A. J., 2004. On climatological mass, heat, and salt transports through the Barents Sea and Fram Strait from a pan-Arctic coupled ice-ocean model simulation. J. of Geophys. Res. 109, C03032, doi:10.1029/2001JC001039. Stroeve, J., W. Maslowski, 2007. Arctic Sea Ice Variability during the last half century, in Special Volume Proceedings of the Conference on Climate Variability and Extremes During the Past 100 Years, Springer, accepted. 1980 1985 1990 1995 2000 -0.4 0 0.4 0.8 1.2 1.6 Bering Strait Volume Transport (Sv) Arctic Ocean Volume Balance Positive volume transport into Arctic Ocean 1980 1985 1990 1995 2000 -4 -2 0 2 Davis Strait Volume Transport (Sv) 1980 1985 1990 1995 2000 -8 -4 0 4 8 Fram Strait Volume Transport (Sv) 1980 1985 1990 1995 2000 Year -4 0 4 8 Barents Sea Opening Volume Transport (Sv) Volume Flux Figure 2. Monthly mean volume fluxes through four major sections shown in Figure 1. Volume flux into the Arctic Ocean is blue, out is red, and net is black. A thick black line represents a 13-month running mean of net volume flux. SUMMARY The 24-year model output was used for analyses of freshwater export from the Arctic Ocean through the Canadian Arctic Archipelago, the Fram, Denmark, Hudson straits, and Cape Farewell to understand the circulation regime in these areas and to quantify the freshwater entering the Labrador and Irminger seas. The importance of this study lies in the quantification of the freshwater export through the CAA into the Labrador Sea, in comparison with the freshwater export through Fram and Denmark straits into the North Atlantic. This is especially important when considering the recent increase of freshwater export via these routes, which could affect deep-water formation. The combined volume fluxes through several sections compare favorably with known observational fluxes. Several important observations can be made from this study. The first is that the Canadian Arctic Archipelago is the largest freshwater contributor to the Labrador Sea and the Hudson Bay is the second largest. Comparison of the absolute freshwater fluxes suggests that the Hudson Bay's freshwater input into the Labrador Sea is more significant than the freshwater contribution from the Fram-Denmark straits and Cape Farewell pathway. Also, the variability of freshwater export into the Labrador Sea is directly affected by large-scale circulation patterns upstream in the Arctic Ocean, which are subject to change due to large-scale atmospheric forcing. This would support a hypothesis that atmospheric regime shifts in the Arctic possibly associated with global warming could hinder convection in the Labrador Sea. Other Aspects of Concern in Ocean Models Table 2. 26-year mean freshwater fluxes for various sections *Red is freshwater due to outflow through Fram and Denmark straits; blue represents freshwater flow through the Canadian Arctic Archipelago. Net FW for S > Sref is only listed when it contributes to the freshwater budget within the Nordic Seas. Greenland Ice sheet Melt Extent 2005 Greenland Ice sheet Melt Extent 2005 – another record melt year another record melt year Causes of Ice Sheets Growth of Reduction - Warmer Ocean Temperature, T : Melting of bottoms of floating ice shelves and tongues of outlet glaciers faster discharge from grounded ice ice loss - Warmer Air Temperature, T : More precipitation (snowfall) ice growth More melting at lower elevations ice loss Jakobshavn Glacier retreat: ~11km between 2000 and 2004 Courtesy Jay Zwally, NASA/GSFC heat flux heat flux (smoothed) melt/growth melt/growth (smoothed) zero reference line for: melt/growth heat flux -40 -20 0 20 40 60 Heat Flux (W/m 2 ) 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 -20 -10 0 10 20 30 40 Corr. coef. (monthly mean) = -0.36; Corr. coef (smoothed) = -0.81 Heat Flux (TW) Greenland (sum) heat flux and area-averaged ice melt/growth (AC removed) -8 -6 -4 -2 0 2 4 x 10 -10 Ice growth/melt (m/day) Figure 4. (a) The area for ice melt/growth calculation (yellow rectangle) and two oceanic sections for the total on-shelf heat flux (black dotted lines). (b) Total monthly mean heat flux onto the Greenland shelf (blue) and area-averaged monthly mean anomalies of ice melt/growth (red) after removing 24-year mean annual cycle. The smoothed lines represent a 13-month running mean. (from Stroeve & Maslowski, 2007). a) b) Melting of the Greenland Ice sheet and subsequent runoff into the ocean is an additional source of freshwater (FW) in the northern North Atlantic, which is not accounted for in ocean models. This FW source is distinct from the Arctic Ocean sources described earlier. Oceanic heat flux and ice melt/growth are negatively correlated in the Greenland Sea. Heat flux and ice melt have recently accelerated. Net (S < Sref) FW (mSv) Net (S > Sref) FW (mSv) Net (S < Sref) FW (mSv) Net (S > Sref) FW (mSv) Net (S < Sref) FW (mSv) Net (S > Sref) FW (mSv) Sref=34.8 Sref=34.8 Sref=35 Sref=35 Sref=35.2 Sref=35.2 Fram Strait 11.893 N/A 24.975 N/A 40.102 N/A S Fram Strait (solid) 50.501 N/A 50.538 N/A 50.575 N/A S Denmark Strait 11.689 -7.19 25.218 N/A 48.904 N/A S Denmark Strait (solid) 10.091 N/A 10.099 N/A 10.106 N/A S Iceland-Scotland Ridge -13.771 -474.647 S Cape Farewell 2.302 -173.352 7.948 -34.864 115.577 N/A W Barents Sea Opening 9.509 -20.859 14.608 -7.244 25.872 -0.007 E Davis Strait 60.935 70.225 78.709 S Davis Strait (solid) 13.079 13.089 13.098 S Hudson Bay Mouth 4.01 4.111 4.211 E Hudson Bay Mouth (solid) 2.035 2.037 2.038 E Smith Sound 15.703 19.989 24.226 S Lancaster Sound 48.171 52.202 56.187 E Bering Strait 47.444 50.913 54.342 N Positive Direction Section CONCLUSIONS Based on calculations of 24-year mean volume transports through four major pathways, flux into and out of the Arctic Ocean is approximately 11.6 Sv. The small imbalance (-0.004 Sv) is due to the Hudson Strait flux, which is not accounted for in the analysis in Table 1. Both Fram Strait and the Barents Sea Opening have high volume fluxes in both directions and are the primary transport pathways. However, Bering Strait and Canadian Archipelago/Davis Strait allow significant freshwater fluxes in and out of the Arctic Ocean. This study shows that the Canadian Arctic Archipelago (CAA) is an important pathway for freshwater export from the Arctic Ocean, both in absolute terms and when compared to the Fram/Denmark Strait route. The CAA provides a direct path for freshwater leaving the Arctic into the Labrador Sea, where it can affect the formation of Labrador Sea Water. The effects of freshwater flux through the CAA on the convection in the northern North Atlantic and on the global thermohaline circulation are yet to be determined in global ocean and climate models (GCMs). The main limitation is due to the horizontal resolution being too coarse in GCMs to realistically account for both volume and freshwater export through the narrow and/or shallow channels of the CAA. The importance of the CAA freshwater contribution to the Labrador Sea and ultimately to the North Atlantic has not been sufficiently studied in part due to the focus of most research on the Fram Strait pathway for water and sea ice export into the sub-arctic seas and the northern North Atlantic. However, our analyses suggest that the total westward freshwater flux off of Cape Farewell is significantly reduced due to the 'salinification' of the signal along its path (Table 2). Strong sources of relatively ‘salty’ water enter the Nordic Seas via Denmark Strait and especially through the Iceland-Scotland Ridge as the Norwegian Atlantic Current. Some of these waters mix with relatively fresh water exported from the Arctic and weaken the freshwater signal as it moves southward along the Greenland coast. This strengthens the relative importance of the freshwater contribution from the CAA, but does not necessarily downplay the role of freshwater from the Denmark Strait. The freshwater flowing southward through Denmark Strait lowers the salinity of large amounts of salty Atlantic water through mixing within the Irminger Sea before it enters into the Labrador Sea. This is different from the freshwater outflow from the CAA, which remains relatively isolated on the northern Labrador shelves (Cuny et al. 2002). Therefore, the water from the CAA has a very low salinity when it is mixed into the Labrador Sea, which consequently could have a major impact upon convection in the region. This further emphasizes the importance of proper representation of freshwater export from the Arctic Ocean via the Canadian Archipelago / Davis Strait and the Hudson Strait in global ocean and climate models. Furthermore, such model improvements would facilitate a more complete answer to the question about the effects of freshwater exported from the Arctic Ocean on the Meridional Overturning Circulation in the North Atlantic.
Transcript of On the importance of freshwater fluxes from the Arctic ...
On the importance of freshwater fluxes from the Arctic Ocean into the North Atlantic: the Nordic Seas versus Canadian Arctic Archipelago
Wieslaw Maslowski1, Jaclyn Clement Kinney
1, Jaromir Jakacki
2
1Naval Postgraduate School, Dept. of Oceanography, Monterey, CA, USA2Institute of Oceanology, Polish Academy of Sciences, Sopot, PolandABSTRACT
We use a high resolution coupled ice-ocean model of the Pan-Arctic region forced with realistic atmospheric data to investigate the variability of sea ice and liquid freshwater fluxes from the Arctic Ocean into the North Atlantic during 1979-2004. Model results are analyzed to compare the relative contribution of the total combined liquid and solid freshwater flux through the two main pathways: Fram-Denmark Strait (FDS) and the Canadian Arctic Archipelago-Davis-Hudson Strait (CAADHS). Our results suggest the relative importance of the freshwater flux through CAADHS into the Labrador Sea. This implies the need for ocean models to adequately represent mass and property fluxes through the narrow and shallow passages of the Canadian Archipelago and Davis and Hudson Straits. We argue that this requirement must be satisfied to advance studies of the Atlantic Meridional Overturning Circulation (MOC) and especially its variability. Given the recent record sea ice melt in the Arctic Ocean, it is critical that global ocean and climate models improve their skill in simulating and predicting effects of changing exports from the Arctic Ocean into the North Atlantic. This talk will outline a possible approach to satisfy such requirements.
Table 1. 24-year mean volume fluxes for various sections
Mean Vol. Trans. (Sv) Net Into AO Out of AOBering Strait 0.649 0.725 -0.076Davis Strait -1.571 0.934 -2.505Fram Strait -2.346 4.741 -7.088Barents Sea Opening 3.264 5.199 -1.934Sum -0.004 11.599 -11.603
MODEL DESCRIPTION In an effort to obtain past estimates of oceanic fluxes through various passages controlling oceanic flow between the Arctic and the lower latitude oceans, we analyze results from a high-resolution coupled ice-ocean model of the Pan-Arctic region. The model domain extends from ~30oN in the North Pacific through the Bering Sea, Arctic Ocean, Nordic Seas, into the North Atlantic to ~45oN. At 9-km and 45-level resolution, the model allows realistic exchanges through important narrow passages. The model has been forced with the atmospheric fields from European Centre for Medium-range Weather Forecast (ECMWF), including 1979-1993 reanalysis and 1994-2004 operational products. Additional details on the model are provided elsewhere (Maslowski et al. 2004; Clement et al. 2005). Volume, heat, salt, and freshwater fluxes are calculated from this integration to allow comparison with historical and future estimates.
Artificial North American Channel
Canada
Alaska
Greenland
Russia
ChinaEurope
Bering Strait
Fram Strait
Smith Strait
Lancaster Sound
Davis StraitCape Farewell
Denmark Strait
Figure 1. Model domain showing the locations of various sections used for frehwater flux analysis.
Barents Sea Opening
Hudson Bay Mouth
Pre-LabradorLabrador Sea
Central Labrador
ACKNOWLEDGEMENTSWe would like to thank the National Science Foundation support of this research. Additonal support has been provided by the Department of Energy (DOE), National Aeronautic and Space Administration (NASA). Computer resources were provided by the Arctic Region Supercomputing Center (ARSC), Army Engineering and Research Development Center (ERDC), and Army High Performance Computing Research Center (AHPCRC) through the Department of Defense High Performance Computer Modernization Program (HPCMP).
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 20020
2
4
6
8
10
12
Time
Fres
hwat
er T
rans
port
(mSv
)
Cape Farewell Freshwater Fluxes
2.342
Reference salinity = 34.8
FW netFW EastFW WestFW mean
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002
5
10
15
20
25
30
35
Time
Fres
hwat
er T
rans
port
(mSv
)
Denmark Strait Freshwater Fluxes
11.429
Reference salinity = 34.8FW netFW NorthFW SouthFW mean
Figure 3. Monthly mean freshwater flux through various sections shown in Figure 1. A thick black line represents a 13-month running mean of net flux.
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 20020
5
10
15
20
25
30
35
Time
Fres
hwat
er T
rans
port
(mSv
)
Fram Strait Freshwater Fluxes
10.638
Reference salinity = 34.8
FW netFW NorthFW SouthFW mean
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002
10
20
30
40
50
60
70
80
90
100
Time
Fres
hwat
er T
rans
port
(mSv
)
Pre-Labrador Freshwater Fluxes
61.642
Reference salinity = 34.8FW netFW NorthFW SouthFW mean
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002
0
5
10
15
20
25
30
Time
Fres
hwat
er T
rans
port
(mSv
)
Hudson Bay Mouth Freshwater Fluxes
4.069
Reference salinity = 34.8 FW netFW EastFW WestFW mean
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 20020
20
40
60
80
100
120
Time
Fres
hwat
er T
rans
port
(mSv
)
Central Labrador Freshwater Fluxes
63.896
Reference salinity = 34.8
FW netFW NorthFW SouthFW mean
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002
10
20
30
40
50
60
70
80
90
100
Time
Fres
hwat
er T
rans
port
(mSv
)
Davis Strait Freshwater Fluxes
60.772
Reference salinity = 34.8FW netFW NorthFW SouthFW mean
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002
10
20
30
40
50
60
70
80
90
Time
Fres
hwat
er T
rans
port
(mSv
)
Lancaster Sound Freshwater Fluxes
48.299
Reference salinity = 34.8
FW netFW EastFW WestFW mean
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002
5
10
15
20
Time
Fres
hwat
er T
rans
port
(mSv
)
Smith Sound Freshwater Fluxes
15.65
Reference salinity = 34.8FW netFW NorthFW SouthFW mean
Freshwater Flux
3.971
67.867
REFERENCESClement, J. L., W. Maslowski, L. Cooper, J. Grebmeier, W. Walczowski, 2005. Ocean circulation and exchanges through the northern Bering Sea - 1979-2001 model results. Deep Sea Research II, 52, 3509-3540, doi: 10.1016/j.dsr2.2005.09.010.Cuny, J., P.B. Rhines, P.P. Niiler, and S. Bacon, 2002. Labrador Sea boundary currents and the fate of the Irminger Sea Water. J. of Phys. Ocean., 32, 627-647.Maslowski, W., Marble, D., Walczowski, W., Schauer, U., Clement, J. L., Semtner, A. J., 2004. On climatological mass, heat, and salt transports through the Barents Sea and Fram Strait from a pan-Arctic coupled ice-ocean model simulation. J. of Geophys. Res. 109, C03032, doi:10.1029/2001JC001039.Stroeve, J., W. Maslowski, 2007. Arctic Sea Ice Variability during the last half century, in Special Volume Proceedings of the Conference on Climate Variability and Extremes During the Past 100 Years, Springer, accepted.
1980 1985 1990 1995 2000
-0.4
0
0.4
0.8
1.2
1.6
Beri
ng
Str
ait
Vo
lum
e T
ransp
ort
(S
v)
Arctic Ocean Volume BalancePositive volume transport into Arctic Ocean
1980 1985 1990 1995 2000
-4
-2
0
2
Da
vis
Str
ait
Vo
lum
e T
ransp
ort
(S
v)
1980 1985 1990 1995 2000
-8
-4
0
4
8
Fra
m S
tra
itV
olu
me T
ransp
ort
(S
v)
1980 1985 1990 1995 2000Year
-4
0
4
8
Ba
ren
ts S
ea O
pe
nin
gV
olu
me T
ransp
ort
(S
v)
Volume Flux
Figure 2. Monthly mean volume fluxes through four major sections shown in Figure 1. Volume flux into the Arctic Ocean is blue, out is red, and net is black. A thick black line represents a 13-month running mean of net volume flux.
SUMMARY The 24-year model output was used for analyses of freshwater export from the Arctic Ocean through the Canadian Arctic Archipelago, the Fram, Denmark, Hudson straits, and Cape Farewell to understand the circulation regime in these areas and to quantify the freshwater entering the Labrador and Irminger seas. The importance of this study lies in the quantification of the freshwater export through the CAA into the Labrador Sea, in comparison with the freshwater export through Fram and Denmark straits into the North Atlantic. This is especially important when considering the recent increase of freshwater export via these routes, which could affect deep-water formation. The combined volume fluxes through several sections compare favorably with known observational fluxes. Several important observations can be made from this study. The first is that the Canadian Arctic Archipelago is the largest freshwater contributor to the Labrador Sea and the Hudson Bay is the second largest. Comparison of the absolute freshwater fluxes suggests that the Hudson Bay's freshwater input into the Labrador Sea is more significant than the freshwater contribution from the Fram-Denmark straits and Cape Farewell pathway. Also, the variability of freshwater export into the Labrador Sea is directly affected by large-scale circulation patterns upstream in the Arctic Ocean, which are subject to change due to large-scale atmospheric forcing. This would support a hypothesis that atmospheric regime shifts in the Arctic possibly associated with global warming could hinder convection in the Labrador Sea.
Other Aspects of Concern in Ocean Models
Table 2. 26-year mean freshwater fluxes for various sections
*Red is freshwater due to outflow through Fram and Denmark straits; blue represents freshwater flow through the Canadian Arctic Archipelago.Net FW for S > Sref is only listed when it contributes to the freshwater budget within the Nordic Seas.
Greenland Ice sheet Melt Extent 2005 Greenland Ice sheet Melt Extent 2005 –– another record melt yearanother record melt year
Causes of Ice Sheets Growth of Reduction- Warmer Ocean Temperature, T :
Melting of bottoms of floating ice shelves andtongues of outlet glaciers fasterdischarge from grounded ice ice loss
- Warmer Air Temperature, T :More precipitation (snowfall) ice growthMore melting at lower elevations ice loss
Jakobshavn Glacier retreat:~11km between 2000 and 2004
Courtesy Jay Zwally, NASA/GSFC
heat flux
heat flux (smoothed)
melt/growth
melt/growth (smoothed)zero reference line for: melt/growth heat flux
-40
-20
0
20
40
60
Hea
t Flu
x (W
/m2 )
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004
-20
-10
0
10
20
30
40
Corr. coef. (monthly mean) = -0.36; Corr. coef (smoothed) = -0.81
Hea
t Flu
x (T
W)
Greenland (sum) heat flux and area-averaged ice melt/growth (AC removed)
-8
-6
-4
-2
0
2
4
x 10-10
Ice
grow
th/m
elt (
m/d
ay)
Figure 4. (a) The area for ice melt/growth calculation (yellow rectangle) and two oceanic sections for the total on-shelf heat flux (black dotted lines). (b) Total monthly mean heat flux onto the Greenland shelf (blue) and area-averaged monthly mean anomalies of ice melt/growth (red) after removing 24-year mean annual cycle. The smoothed lines represent a 13-month running mean. (from Stroeve & Maslowski, 2007).
a) b)
Melting of the Greenland Ice sheet and subsequent runoff into the ocean is an additional source of freshwater (FW) in the northern North Atlantic, which is not accounted for in ocean models. This FW source is distinct from the Arctic Ocean sources described earlier. Oceanic heat flux and ice melt/growth are negatively correlated in the Greenland Sea. Heat flux and ice melt have recently accelerated.
Net (S < Sref) FW (mSv)
Net (S > Sref) FW (mSv)
Net (S < Sref) FW (mSv)
Net (S > Sref) FW (mSv)
Net (S < Sref) FW (mSv)
Net (S > Sref) FW (mSv)
Sref=34.8 Sref=34.8 Sref=35 Sref=35 Sref=35.2 Sref=35.2Fram Strait 11.893 N/A 24.975 N/A 40.102 N/A SFram Strait (solid) 50.501 N/A 50.538 N/A 50.575 N/A SDenmark Strait 11.689 -7.19 25.218 N/A 48.904 N/A SDenmark Strait (solid) 10.091 N/A 10.099 N/A 10.106 N/A SIceland-Scotland Ridge -13.771 -474.647 SCape Farewell 2.302 -173.352 7.948 -34.864 115.577 N/A WBarents Sea Opening 9.509 -20.859 14.608 -7.244 25.872 -0.007 EDavis Strait 60.935 70.225 78.709 SDavis Strait (solid) 13.079 13.089 13.098 SHudson Bay Mouth 4.01 4.111 4.211 EHudson Bay Mouth (solid) 2.035 2.037 2.038 ESmith Sound 15.703 19.989 24.226 SLancaster Sound 48.171 52.202 56.187 EBering Strait 47.444 50.913 54.342 N
Positive DirectionSection
CONCLUSIONS Based on calculations of 24-year mean volume transports through four major pathways, flux into and out of the Arctic Ocean is approximately 11.6 Sv. The small imbalance (-0.004 Sv) is due to the Hudson Strait flux, which is not accounted for in the analysis in Table 1. Both Fram Strait and the Barents Sea Opening have high volume fluxes in both directions and are the primary transport pathways. However, Bering Strait and Canadian Archipelago/Davis Strait allow significant freshwater fluxes in and out of the Arctic Ocean. This study shows that the Canadian Arctic Archipelago (CAA) is an important pathway for freshwater export from the Arctic Ocean, both in absolute terms and when compared to the Fram/Denmark Strait route. The CAA provides a direct path for freshwater leaving the Arctic into the Labrador Sea, where it can affect the formation of Labrador Sea Water. The effects of freshwater flux through the CAA on the convection in the northern North Atlantic and on the global thermohaline circulation are yet to be determined in global ocean and climate models (GCMs). The main limitation is due to the horizontal resolution being too coarse in GCMs to realistically account for both volume and freshwater export through the narrow and/or shallow channels of the CAA. The importance of the CAA freshwater contribution to the Labrador Sea and ultimately to the North Atlantic has not been sufficiently studied in part due to the focus of most research on the Fram Strait pathway for water and sea ice export into the sub-arctic seas and the northern North Atlantic. However, our analyses suggest that the total westward freshwater flux off of Cape Farewell is significantly reduced due to the 'salinification' of the signal along its path (Table 2). Strong sources of relatively ‘salty’ water enter the Nordic Seas via Denmark Strait and especially through the Iceland-Scotland Ridge as the Norwegian Atlantic Current. Some of these waters mix with relatively fresh water exported from the Arctic and weaken the freshwater signal as it moves southward along the Greenland coast. This strengthens the relative importance of the freshwater contribution from the CAA, but does not necessarily
downplay the role of freshwater from the Denmark Strait. The freshwater flowing southward through Denmark Strait lowers the salinity of large amounts of salty Atlantic water through mixing within the Irminger Sea before it enters into the Labrador Sea. This is different from the freshwater outflow from the CAA, which remains relatively isolated on the northern Labrador shelves (Cuny et al. 2002). Therefore, the water from the CAA has a very low salinity when it is mixed into the Labrador Sea, which consequently could have a major impact upon convection in the region. This further emphasizes the importance of proper representation of freshwater export from the Arctic Ocean via the Canadian Archipelago / Davis Strait and the Hudson Strait in global ocean and climate models. Furthermore, such model improvements would facilitate a more complete answer to the question about the effects of freshwater exported from the Arctic Ocean on the Meridional Overturning Circulation in the North Atlantic.
