ARCUS | Arctic Research Consortium of the United States
ARCSS Overview

Message From the ARCSS Committee

Community Planning

Community Surveys

ARCSS Listserve

New ARCSS AO

Current Arctic Announcement

ARCSS Synthesis Process

ARCSS Committee

ARCSS Meetings

ARCSS Publications

ARCSS Research Components

ARCSS Data Coordination

Contact Information

    2002 ARCSS All-Hands Workshop

    February 20, 2002
    Bell Harbor International Conference Center, Seattle WA

    Effects of Horizontal Resolution on GCM Simulations of Mid-and High-Latitude Circulation in the Northern Hemisphere

    Cecilia M. Bitz1, Richard E. Moritz2, Jeffrey Yin3, Philip B. Duffy4
    1Polar Science Center, University of Washington, 1013 NE 40th St, Seattle, WA, 98105, USA, Phone 206-543-1339, Fax 206-616-3142, bitz@apl.washington.edu
    2Polar Science Center, USA
    3JISAO, USA
    4Lawrence Livermore National Laboratory , USA

    Key features of the mid- and high-latitude atmospheric circulation are poorly simulated by current general circulation models and are crucial to the coupled global atmosphere-land-ice-ocean system. These features include the mean annual cycles and variability of Arctic atmospheric circulation. When such GCM's are coupled to dynamical sea ice models, the resulting fields of sea ice velocity and thickness are qualitatively unrealistic, with significant implications for climate feedbacks.

    We have found that horizontal resolution has a significant effect on the wintertime Arctic atmospheric circulation in simulations with the NCAR CCM3.6 using a spectral dynamical core with triangular truncation. Our analysis of simulations at T42, T85, and T170 indicates that the position and amplitude of the wintertime Beaufort high are improved with higher resolution. We find storms enter the northern North Atlantic and Barents Sea more frequently in the higher resolution models, in better agreement with observations. Kinetic energy and meridional heat transport are higher in the storm tracks in the higher resolution runs. In addition, the relative magnitude of the simulated Pacific and Atlantic storm tracks is more realistic, and there is evidence that more storms track across North America, where GCMs have typically failed to capture the observed path of storms. The influence of synoptic-scale eddies on the mean flow is both stronger and at smaller length scales at high resolution. These characteristics match well with observations, although the maxima in this influence are not necessarily well located.

    Some areas show no improvement with increased resolution. For example, eddy kinetic energy and meridional heat transport at mid-tropospheric levels over Alaska increase with resolution although they are higher than observed even at T42 resolution. Most mid-latitude changes in the meancirculation are not clear improvements, such as the changes in position and amplitude of the Aleutian and Icelandic lows.


    Previous Abstract | Next Abstract

    Return to Abstracts page.