[MITgcm-support] about 510*510*6*50 cubed sphere example

[Dimitris Menemenlis]

Liuyu,

> Unfortunately, I can not access to NASA Ames computers, and have no 
> "ncep_rgau" forcing files either, so I wonder where I can find these needed 
> files.

Chris Hill has just made available all the needed configuration and forcing
files on the web:  http://ecco2.org/products/model_setups/cs510.html

In particular all the NCEP forcing files from 1948 to present are here:
http://ecco2.org/products/model_setups/ncep_rgau/
You don't need to download all of them, only the fields and years that you need.

I need to warn you, however, that this is a rather large model configuration and
that I would not attempt to run it unless you have access to a reasonably large
and fast supercomputer.

If you are only interested in model output rather than in running
high-resolution integrations yourself, the ECCO2 project also makes available
model output from several high-resolution integrations.  See http://ecco2.org
and http://ecco2.org/products/output/.

> By the way, do you think that using the cubed sphere coordinate is better
> than using other coordinates for global ocean and sea ice simulation?

There are tradeoffs either way.  Nice thing about cubed-sphere is relatively
even horizontal grid spacing throughout the domain, and a good representation of
Arctic Ocean.  But we are also using other grids with different tradeoffs.

Dimitris

-- 
Dimitris Menemenlis <menemenlis at jpl.nasa.gov>
Jet Propulsion Lab, California Institute of Technology
MS 300-323, 4800 Oak Grove Dr, Pasadena CA 91109-8099
tel: 818-354-1656;  fax: 818-393-6720