| Summary: | The article of record as published may be found at http://dx.doi.org/10.1029/2006JC003642 As a part of the Arctic Ocean Model Intercomparison Project, results from 10 Arctic ocean/ice models are intercompared over the period 1970 through 1999. Models' monthly mean outputs are laterally integrated over two subdomains (Amerasian and Eurasian basins), then examined as functions of depth and time. Differences in such fields as averaged temperature and salinity arise from models' differences in parameterizations and numerical methods and from different domain sizes, with anomalies that develop at lower latitudes carried into the Arctic. A systematic deficiency is seen as AOMIP models tend to produce thermally stratified upper layers rather than the “cold halocline”, suggesting missing physics perhaps related to vertical mixing or to shelf‐basin exchanges. Flow fields pose a challenge for intercomparison. We introduce topostrophy, the vertical component of V×∇D where V is monthly mean velocity and ∇D is the gradient of total depth, characterizing the tendency to follow topographic slopes. Positive topostrophy expresses a tendency for cyclonic “rim currents”. Systematic differences of models' circulations are found to depend strongly upon assumed roles of unresolved eddies. OPP-0002239 OPP-0327664 National Science Foundation Office of Polar Programs International Arctic Research Center, University of Alaska Fairbanks U.S. Department of Energy, Climate Change Prediction Program Center for Computational Sciences at Oak Ridge National Laboratory NSF/ARCSS, by the NASA Global Modeling and Analysis, Radiation Sciences, Cryospheric Sciences Programs Russian Foundation for Basic Research
|
|---|