HR: 11:20h
AN: C52A-05    [Abstracts]
TI: Comparison of Model- and Satellite-Derived Arctic Sea Ice Thickness
AU: * Lipscomb, W H
EM: lipscomb@lanl.gov
AF: Los Alamos National Laboratory, Group T-3, MS B216, Los Alamos, NM 87545 United States
AU: Hunke, E C
EM: eclare@lanl.gov
AF: Los Alamos National Laboratory, Group T-3, MS B216, Los Alamos, NM 87545 United States
AU: Mills, S C
EM: m054692@usna.edu
AF: Department of Oceanography, U. S. Naval Academy, Nimitz Library, 589 McNair Rd., Annapolis, MD 21402 United States
AU: Laxon, S
EM: swl@cpom.ucl.ac.uk
AF: Centre for Polar Observation and Modelling, University College London, Gower St., London, WC1E 6BT United Kingdom
AB: Sonar measurements suggest that Arctic sea ice has thinned in recent decades. The amount of thinning is uncertain because of the sparseness of observations and the difficulty of distinguishing the long-term signal from natural variability. Climate models predict a thinning ice cover, but until recently it has been impossible to validate models with synoptic-scale data. A new eight-year time series from satellite altimetry provides, for the first time, a basin-scale ice thickness data set for model validation. The satellite data, which cover about half the area of permanent Arctic sea ice, have been compared to ice thicknesses generated by the Los Alamos sea ice model, CICE, coupled to the POP ocean model. The model was run for 50 years on a 0.4-degree global grid using protocols and forcing data from the Arctic Ocean Model Intercomparison Project (AOMIP). Model- and satellite-derived wintertime ice thickness fields were compared for 1993-2001. The modeled mean thickness of 2.70 m over the region of satellite coverage is very close to the mean from the data. The spatial thickness patterns are generally similar, with the thickest ice adjacent to the Canadian Archipelago and Greenland. The model thickness is biased high in the Canadian Archipelago, possibly because of insufficient grid resolution, and low in the Barents Sea, probably because of excessively warm ocean temperatures. The standard deviation of wintertime ice thickness in the model is about 4% of the mean, compared to 9% for the satellite data. Much of the model variability results from dynamic processes. Thermodynamic variability is underestimated, at least in part because the AOMIP radiative forcing is based on monthly climatologies. Future work will aim to correct these biases.
UR: http://climate.lanl.gov/
DE: 4215 Climate and interannual variability (3309)
DE: 4255 Numerical modeling
DE: 4540 Ice mechanics and air/sea/ice exchange processes
DE: 1863 Snow and ice (1827)
DE: 1635 Oceans (4203)
SC: Cryosphere [C]
MN: 2004 AGU Fall Meeting