Preconditioning the Greenland Sea for deep convection: Ice formation and ice drift
The role of sea ice in preconditioning the mixed layers of the central Greenland Sea for deep convection is investigated, with particular emphasis on the formation of the “Nordbukta.” The opening of the ice free bay in late January 1989 indicated that the upper layer was well preconditioned for deep...
| Published in: | Journal of Geophysical Research |
|---|---|
| Main Authors: | , , |
| Format: | Article in Journal/Newspaper |
| Language: | English |
| Published: |
AGU (American Geophysical Union)
1995
|
| Subjects: | |
| Online Access: | https://oceanrep.geomar.de/id/eprint/1519/ https://oceanrep.geomar.de/id/eprint/1519/1/Visbeck%20et%20al.%20%281995%29.pdf https://doi.org/10.1029/95JC01611 |
| Summary: | The role of sea ice in preconditioning the mixed layers of the central Greenland Sea for deep convection is investigated, with particular emphasis on the formation of the “Nordbukta.” The opening of the ice free bay in late January 1989 indicated that the upper layer was well preconditioned for deep convection which reached down to 1500 m depth in March 1989. We propose that the ice free bay occurred due to diminishing new ice formation without extensive ice melt. A key process is wind‐driven ice drift to the southwest, as observed by upward looking acoustic Doppler current profilers, which will alter the upper ocean freshwater budget when an ice volume gradient along the ice‐drift direction exists. We investigated the importance and effects of such an ice‐drift‐induced freshwater loss on upper ocean properties using an ice‐ocean mixed‐layer model. Observed temperature and salt profiles from December 1988 served as initial conditions, and the model was integrated over the winter season. Given the one‐dimensional physics and climatological surface fluxes, the model was not able to produce a reasonable ice and mixed‐layer evolution. However, allowing ice drift to reduce the local ice thickness improved the ice‐ocean model performance dramatically. An average ice export of 5–8 mm d−1 was needed to be consistent with the observed evolution of mixed‐layer properties and ice cover. Using the same fluxes and ice export, but initial conditions from the “Is Odden” region, yielded ice cover throughout the winter over a shallow mixed layer, both of which are consistent with the observations from the Odden region. |
|---|