[an error occurred while processing this directive] [an error occurred while processing this directive]

Report of the CSM Ocean Model Working Group Meeting
January 18 and 19, 2000, NCAR
Michael Spall and Peter Gent, Co-Chairs


Presentations were given as follows:


Model Development

Frank Bryan - Status of NCOM/POP Comparisons

POP spin-up exhibits oscillatory mode in the southern ocean, period 20-25 years (also present in NCOM, just very weak). Amplitude depends on time step acceleration, viscosity tapering with latitude. Wave activity is correlated with the trend in global temperature and also appears to alter the equilibrium state. That raises the following questions:

- How should we spin-up models?
- Is the mode computational or physical?

The NCOM and POP models are as similar as can be expected given different numerics.

Rick Smith - Anisotropic Viscosity in POP

Derived anisotropic stress tensor for generalized coordinates and tested in a zonal channel and x2' global model. The mean kinetic energy increases by a factor of two; there is better western boundary current separation and stronger subpolar gyres, ACC, and Labrador current. There are similar results to NCOM, but implementation is very different.

Tony Craig - Evaluation of Variable Diffusivity in POP Based on Local Flow Properties

He discussed the Visbeck et al. scheme and testing in 2/3-degree POP model with 50-year runs. The model fields in WBC regions are better with smaller values of kappa. Testing and implementation continues.

General discussion - what next with POP?

NCOM and POP comparisons are useful and nearly completed. New sensitivities appear as models become less diffusive. We need to add basin runoff and biogeochemistry to POP. We will do further testing and development in the BBL scheme of Beckmann-Doescher and partial cells. There are resolution and spin-up demands on computer time.

Tony Song - Testing Various Grids and BBL Models

A test case was proposed with wind-driven flow over a coastal canyon, combined z-sigma model z-level in interior (no pressure gradient errors), and sigma near bottom with BBL parameterization.

Analysis

Gokhan Danabasoglu - Upper Ocean Coupled Models in Equatorial Pacific

Many models are being tested (CSM1, CSM Upper Ocean Model, Gent-Cane), and the mean SST is too cold, the semi-annual signal is too large, and the undercurrent appears much better with finer meridional grids. The reasons for poor SST is unknown but is sensitive to details of vertical mixing, regional vs. global model, and vertical coordinate (G-C vs. level/upper ocean model).

Craig Tierney- Using CSM to Estimate Sea Level Rise

He used CSM to provide bounds on errors that may be introduced to estimates of sea level rise due to sampling bias. The power spectra were reasonable at low frequency (except near equator). His conclusion was that errors due to sampling might be significant.

Yi Chao- Atlantic Water Masses in x3' NCOM

He presented his analysis of water masses in the eastern Atlantic in the x3' model that showed the water masses were quite poorly represented.

Attendees from Ocean and Polar Working Groups

Julie McClean, Polar
Michael Spall
Craig Tierney
Amanda Lynch, Polar
Bill Lipscomb, Polar
Esther Brady
Phil Jones
Tony Craig
Tom Bettge
Yi Chao
Elizabeth Hunke, Polar
William Holland
Frank Bryan
Jim McWilliams
Matthew Hecht
Gokhan Danabasoglu
Yuxia Zhang, Polar
Rick Smith
JoAnn Lysne
Peter Gent
Steven Jayne
Bette Otto-Bliesner
Casper Ammann
Steve Yeager
Wieslaw Maszowski
Robin Tolcmalcian
Richard Moritz, Polar
Bob Malone
Keith Lindsay
Marika Holland, Polar
Todd Ringler
Bruce Briegleb, Polar
J. Keith Moore
Y. Tony Song
Warren Washington
Bert Semtner
Todd Arbetter, Polar
Scott Doney