[MITgcm-devel] free drift ice model

[Matthew R. Mazloff]

Hi Martin,

I do not doubt that the seaice dynamics code works beautifully (and 
congrats on that!).  I am 
concerned 
about computational efficiency, however, and I will get a significant 
savings by turning seaice dynamics off and just setting Uice to some 
linear function of the atmospheric wind and ocean surface velocity.  So if 
this is a decent approximation over much of the Southern Ocean I would 
like to go that route. I was hoping for some advice and references on the 
best way to do this.  From what I've read so far I'm leaning towards just 
setting Uice = Uvel(k=1).  I am going to keep reading though, and I'm also 
going to evaluate how the ice moves wrt to wind & ocean in Patrick's 
model.

-Matt



On Thu, 30 Nov 2006, Martin Losch wrote:

> Hi Matt, Patrick et al.
>
> I am very confident about the C-grid dynamics in seaice. As D. says, we need 
> to write it up! With the C-grid the ice-ocean drag works. You can even use 
> different formulations, the "intuitive" one where ice assumed to float on top 
> of water (with no draft), and the Hibler+Bryan87 formulation, which has other 
> assumptions that may not be optimal, but both work, the former runs in 
> Dimitris' high res cube. Correct my if I am wrong.
> I am only using C-grid in my (coarse) runs (2deg "isotropic" grid).
>
> M.
> On 30 Nov 2006, at 00:59, Patrick Heimbach wrote:
>
>> 
>> Hi there,
>> 
>> just some small comments:
>> 
>> On Nov 29, 2006, at 6:40 PM, Dimitris Menemenlis wrote:
>> 
>>> Matt, I am going to guess that free drift is probably a good estimate for 
>>> most
>>> of Southern Ocean, except close to coastlines.  I don't know what your A 
>>> and B
>>> numbers should be, they would depend on drag coefficients, etc.  Look for 
>>> papers
>>> by Miles McPhee on topic.  Although most of his work is for Arctic Ocean, 
>>> I
>>> think that his estimates over first year ice should be applicable for 
>>> Southern
>>> Ocean.
>> 
>> Should maybe be a bit careful re. applicability of Arctic numbers
>> for Antarctic sea-ice (latter seems more "loose", so more in free drift).
>> But maybe ok as first guess.
>> 
>>>> I believe in the current model set up the wind stress acts on the ocean 
>>>> as if
>>>> the ice wasn't there.
>>> 
>>> Not true anymore.  Martin's magic C-grid formulation makes ice-affected 
>>> ocean stress possible.  We need to write it up.  Apparently it's a big 
>>> deal for sea-ice models.  Trying to iron out residual problems with 
>>> growth, etc., first.
>> 
>> Maybe again be careful in stating what works when.
>> Matt probably still has B-grid code enabled for which
>> his assumption is correct.
>> But good to hear, the C-grid code has it.
>> Might be worth switching then if it's "well" tested. Is it?
>> 
>> BTW,
>> I ran v3.iter0 with latest growth code.
>> Seaice concentration misfits is reduced by 2/3, not bad.
>> Still unclear re. fin.diff. sensitivities.
>> Question is, which part of the code did it.
>> 
>> -p.
>> 
>> 
>> 
>> 
>>> Cheers, D.
>>> 
>>> -- 
>>> 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
>>> _______________________________________________
>>> MITgcm-devel mailing list
>>> MITgcm-devel at mitgcm.org
>>> http://mitgcm.org/mailman/listinfo/mitgcm-devel
>> 
>> Dr Patrick Heimbach | heimbach at mit.edu | http://www.mit.edu/~heimbach
>> MIT | EAPS, 54-1518 | 77 Massachusetts Ave | Cambridge, MA 02139, USA
>> FON: +1-617-253-5259 | FAX: +1-617-253-4464 | SKYPE: patrick.heimbach
>> 
>> 
>> 
>> _______________________________________________
>> MITgcm-devel mailing list
>> MITgcm-devel at mitgcm.org
>> http://mitgcm.org/mailman/listinfo/mitgcm-devel
>
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-devel
>