If you are using
Navigator 4.x
or
Internet Explorer 4.x
or
Omni Web 4.x
, this site will not render
correctly!
gfdl's home page > people > John Dunne >
Notes from Oct 29
The primary goal of this meeting was to decide on the course of action for development of a coarse resolution ocean model in MOM4 for Biogeochemical and Earth System Model development.
After presentations by Gnandesikan dna Dunne and a subsequent
discussion, it was decided that:
1) a
"best configuration" of OM1 should be frozen based on the MOM3
results, and run out with the OCMIP configurations, but no active
development of this model should take place at GFDL.
2) A revised OM1p5 should be
developed and tuned to serve as the work-horse for Biogeochemical and
Earth System Model developmentat GFDL.
- Sensitivity of ventilation to GM/viscosity should be established by tuning to CFCs, 14C and primary productivity.
- The sensitivity of circulation in the interior to Z-level resolution should be assessed by comparing 28 level (base OM1p5) and 50 level (OM3) versions
- Model results should be incorporated into the database for general GFDL use
- Postprocessing tools should be implemented to ease analysis
Dunne's full powerpoint presentation can be downloaded here . A synopsis follows...
Gnanadesikan demonstrated the timescales of biogeochemical equilibration in OM1 to be ~20 years for physical stabilization, ~100 years for nutrient readjustment and upwards of 1000 years for deep CO2 equilibration. The major point was that these timescales will take too much real time and computational power to be done in the fine reslution models already developed within MOM4 (OM2/OM3).
Dunne described the current status of ocean biogeochemistry code development in MOM4:
- Codes now available:
- OCMIP2 Biotic
- Restoring diagnostic biology
- Prognostic biology
- OCMIP2 CFCs
- Hadley center biology
- OM1 PRINCE2 forcing (Simeon)
- OCMIP2 Abiotic and 14C (Slater)
- Field manager (Slater/Cooke)
- Moving gas exchange to coupler (Fan?)
- Biological tuning (Dunne)
Dunne described the deficiencies of the current OM1, which include:
- Makeshift deep water formation in southern ocean
- NADW not at right depth
- NPIW formed from subtropical waters
- Which forcing/configuration is best?
- Does not resolve euphotic zone (plants need light - this reslults in a severe decrease in phytoplankton activity in OM1)
- Limited mixed layer dynamics
- Coarse resolution in the equatorial zone (N-trapping, weak undercurrent, large T errors)
- No ice dynamics
- Poor dynamic coupling with an Atmospheric Model
- Physics has not been tuned for MOM4
Dunne described legacies of MOM3 in the current OM1 due to the grid being inherited from R15 coupled model:
- 96 zonal bins to equal atmospheric model
- Depth scaling doubled from 12 level model
- Deepest depth to 5000m (OM2/OM3 are 5500m)
- 2 empty rows in Antarctica (40 total)
- Free surface rather than rigid lid means time-varying solutions differ (significantly?)
- Smaller time step necessary for free surface. Can we make use of this time to do better physics?
- Tripolar enhancement is there. Does the current grid configuration make best use of it to resolve arctic?
Dunne went on to propose a re-evaluation of the coarse model configuration that would be coarse enough to do many/long runs while:
- Resolving the Euphotic zone
- Explicit Mixed layer dynamics
- Equatorial undercurrent/recirculation cells
- Ice dynamics
- Comparable to NCAR/CCSM coupled paleo-model (100x116x25; POP; shifted pole; high res. tropics)
- have a realistic potential for coupling to atmosphere for Earth System Model development
A comparison of the the MOM3 veriosn, MOM4 OM1 version, and the new model (dubbed - OM1p5) was presented:
