IGOR POLYAKOV
Torgny Vinje, Terje Loyning, Igor Polyakov
[for more details read Vinje et al., 2002 available on web http://www.frontier.iarc.uaf.edu/~igor/convec/vinje.pdf]
Convection in the Greenland Sea is an impotant factor which affects the formation of deep waters of the World Ocean. Cyclonic oceanic circulation, weak stratification, and surface forcing (cooling or brine rejection due to ice formation) are some of important preconditions for the open-ocean convection suggested by Killworth [1983]. These preconditions are frequently observed in the Greenland Sea area between 75-77N, east of the East-Greenland ice drift stream (EGIS) proper, known as Odden-Nordbukta (see Figure 1).
Our coupled ice-ocean model demonstrates that in Odden-Nordbukta the ice and ocean surface gyre is correlated with the topographically controlled deep-water cyclonic gyre explaining Comiso et al. [2001] findings that the shape of Odden bares some similarities with the bathymetry. This attests to the important role of ocean circulation in shaping the Nordbukta/Odden ice. Occasionally, a local atmospheric low enhances the cyclonic oceanic circulation, sending additional ice northeast from Jan Mayen, and Nordbukta becomes ice-free. The cooling in Nordbukta destabilizes weak density stratification and starts thermal convection. It also favours ice formation (Figure 2) accompanied by brine rejection and haline convection.
However, observations and modeling results show some indications that ice formation is not that important for the open-ocean convection and that the advection of salty water to the convection cell plays a crucial role. Recent observations show development of convective chimneys down to 2500 m in the Greenland Sea when the Odden and Nordbukta features were absent [Wadhams et al., 2002]. Comiso et al. [2001] found a poor correlation of Odden with convection events. Our model also shows low (R=0.13) correlation between stratification and ice production in Nordbukta. This suggests that sea ice formation may not be that important for deep-water convection as was believed.
The question arises as to what drives the Greenland Sea deep-water convection. Observations [Comiso et al., 2001] and modeling show a strong variability of the Odden/Nordbukta system at various time-scales. According to the model, in 1952-76 the stratification was generally close to neutral whereas prior to 1952 and from the mid-1970s to the present the increased ice and fresh water transport from the Arctic has been spread over the area east of EGIS capping the Nordbukta region and preventing deep-water convection (Figure 3). The above periods closely resemble phases of multi-decadal arctic variability [Polyakov and Johnson, 2000]. This result is corroborated by oceanographic observations and chemical measurements showing that the formation of Greenland Sea Deep Water slowed considerably in the recent decades [Dickson et al., 1996; Schlosser et al., 1991]. Thus, stratification/convection in the Greenland Sea Gyre and NAO index show coherent variations when positive NAO anomalies are associated with depressed deep-water convection and ventilation (as it has been observed in the 1980-90s), whereas negative NAO anomalies are accompanied by vertical convection and deep-water formation (as was observed in the 1960s and 2001).
Comiso, J. C., P. Wadhams, L. T. Pedersen, and R. A. Gersten, Seasonal and interannual variability of the Odden ice tongue and a study of environmental effects, J. Geophys. Res., 106, 9093-9116, 2001.
Dickson, R., J. Lazier, J. Meincke, P. Rhines, and J. Swift, Long-term coordinated changes in the convective activity of the North Atlantic, Prog. Oceanogr., 38, 241-295, 1996.
Killworth, P. D., Deep convection in the world ocean, Rev. Geophys. Space Phys., 21, 1-26, 1983.
Polyakov, I., and M. Johnson, Arctic decadal and interdecadal variability, Geophys. Res. Lett., 27, 4097-4100, 2000.[download pdf]
Schlosser, P., G. Bonisch, M. Rhein, and R. Bayer, Reduction of deepwater formation in the Greenland Sea during the 1980s: Evidence from tracer data, Science, 251, 1054-1056, 1991.
Vinje, T., T. B. Loyning, and I. Polyakov, Effects of melting and freezing in the Greenland Sea, Geophys. Res. Letters, 2002, accepted.
Wadhams, P., J. Holfort, E. Hansen, and J. P. Wilkinson, A deep convective chimney in the winter Greenland Sea, Geophys. Res. Lett., 2002 (accepted)
Simulated 1946-97 ice monthly melt/freezing rates (cm/month, color). Solid line separates melting and freezing regions. Bathymetry is shown by dotted lines. Note Nordbukta region where in winter ice formation occurs whereas ice melts northward and southward from Nordbukta.
Brunt-Vaisala frequency showing generally weaker stratification in Nordbukta in 1952-76 and stronger stratification prior to 1952 and from the mid-1970s to the present.
Last modified: May 06, 2004. 14:57:23 pm