Summer landfast sea ice desalination at Point Barrow, Alaska: Modeling and observations
Vancoppenolle, M.; Bitz, C.M.; Fichefet, T. (2007). Summer landfast sea ice desalination at Point Barrow, Alaska: Modeling and observations. J. Geophys. Res. 112(C4). dx.doi.org/10.1029/2006JC003493
In: Journal of Geophysical Research. American Geophysical Union: Richmond. ISSN 0148-0227; e-ISSN 2156-2202, meer
|  |
| Auteurs | | Top |
- Vancoppenolle, M., meer
- Bitz, C.M.
- Fichefet, T., meer
|
|
|
| Abstract |
[1] Landfast sea ice cores from two sites in Point Barrow, Alaska, extracted between 1999 and 2001, show a progressive desalination and a corresponding shape transition in the salinity profile starting at snowmelt onset, around 1 June. The vertical percolation of fresh surface meltwater through the permeable ice matrix (flushing) has long been supposed to control this transition. A parameterization of flushing in bare ice is formulated. It is incorporated into the semiempirical winter sea ice desalination model of Cox and Weeks (1988), coupled to the one-dimensional thermodynamic sea ice model of Bitz and Lipscomb (1999) and forced under Point Barrow conditions. Adjustment to the original Cox and Weeks parameterization of gravity drainage was necessary to give reasonable agreement with observations in winter. With this change the model has the potential to simulate the full seasonal cycle of the Arctic salinity and mass balance of typical Arctic ice. In summer, the model salinity profile closely follows the observations. The model upper ice temperatures are slightly too cold. At the thin snow (Chukchi Sea) site, the model simulates the observed snow depth and ice thickness well. At the thick snow (Elson Lagoon) site, the snow disappears 7 days earlier than observed, which results in underestimating ice thickness. Sensitivity experiments suggest that a more realistic representation of snow and meltwater physics would significantly improve the simulation. Because of the effects of brine drainage on the sea ice mass balance and oceanic circulation, including the time dependence of the ice salinity profile would significantly improve the next generation of sea ice models. |
|
