[MITgcm-devel] partial cell under ice-shelf

[Martin Losch]

Hi Dimitris,
On Apr 26, 2012, at 2:15 PM, Menemenlis, Dimitris (3248) wrote:

> d) summarize (in words that non-specialists can understand)
> what it is exactly that this new switch will achieve.
with the new code it's possible to get smooth Eta, for what it's worth. That's all, everything else stays pretty much the same. There are small differences in T/S  near the ice shelf interface, but that's only a consequence of a slightly different hydrostatic pressure compuation (numerical issues, but because pressure is so huge in SI units, it does make a difference)
> 
> In particular, does this partial-cell-under-ice-shelves update have
> implications for the addition of tidal forcing?
No.
> 
> Also an old point of confusion - not sure if relevant in this thread:
> what is the interpretation of ETAN under an ice shelf?
> Does the "volume" of water in an ice shelf cavity "change" with ETAN?
> Is this equivalent to an ice shelf cavity flexing at the grounding line
> and moving up and down with tides?
This should be part of the new FAQ-section that we should install: You can view the (local) variation/change of ETAN as the change of (local) ice-ocean interface position. Near the grounding line this viewpoint may not be so accurate (in theory you can have large excursions of ETAN, but the real ice will probably not move up/down very much).

Martin

> 
> Thanks
> 
> Dimitris Menemenlis
> 
> On Apr 25, 2012, at 12:44 PM, Martin Losch wrote:
> 
>> Hi Jean-Michel,
>> 
>> that sounds good to me. Let's do it that way. I'll do (b) after you've done (a)
>> 
>> Martin
>> 
>> On Apr 25, 2012, at 9:17 PM, Jean-Michel Campin wrote:
>> 
>>> Hi Martin,
>>> 
>>> What I propose is, in this order, to:
>>> a) check-in the switch tomorrow.
>>> b) update ismop test experiment
>>> c) send email to the list of pkg/shelfice users.
>>> (easier to explain how to revert to old version when the new version
>>> is already there)
>>> I will not remove the piece of code in cal_phi_hyd.F, so that it 
>>> will be easy enough to get back to the previous formulation
>>> (just 1 line to comment out in set_parms.F).
>>> 
>>> Cheers,
>>> Jean-Michel
> 
> 
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-devel