[MITgcm-devel] ICE workshop - recap and homework!

[Patrick Heimbach]

Hi Jean-Michel,

no objections from my part.
The sooner the better, both on model changes and lab_sea config.
And the fix suggested by Martin/Dimitris re. limiting SEAICE_mcPheePiston.

I'll do the solve_…diagonal this WE.

-p.

On Mar 10, 2012, at 11:42 AM, Jean-Michel Campin wrote:

> Hi,
> 
> I took a look at "seaice-verif-notes.txt" file (things we discuss
> on Thursday afternoon of our seaice-workshop), and I propose to check-in 
> the 2 remaining changes (secion: model development:) that are not
> yet beeing implemented, i.e., to rename parameters:
>> SIsalFrac to SEAICE_saltFrac
>> SIsal0    to SEAICE_salt0
> Will proceed soon.
> 
> I would like also not to wait anymore before changing "lab_sea" 
> verification experiment (starting with forward test),
> as we discuss during our meeting.
> 
> Any objection ?
> 
> Cheers,
> Jean-Michel
> 
> On Thu, Mar 08, 2012 at 05:50:22PM +0100, Martin Losch wrote:
>> Dear seaice-folks,
>> 
>> based on the online-discussion of this week, I have come up with a slightly updated version of SEAICE_OPTIONS.h (actually not changed from the previous one, but now we do not need to edit it any longer to make it work, thanks to Gael's recent fixes) and data.seaice. The "only" change is in the "turbulent flux" parameters. The new data.seaice contains
>>> SEAICE_mcPheeTaper=0.92,
>> which is meant to reproduce the default behavior of the McPhee parameterization that used to be invoked with defining MCPHEE_OCEAN_ICE_HEAT_FLUX and GRADIENT_MIXED_LAYER_TURBULENCE_FACTOR. Note, however, that last week (it seems so far away) we agreed to allow supercooled water to turn into ice. This is reflected in the default behavior of the new flag SEAICE_frazilFrac=1. (if for some reason you don't want this, then set this flag to 0., but I suggest that for our "baseline" tests, that you leave this parameter unchanged).
>> 
>> Now, based on today's code, I have used the SEAICE_OPTIONS.h and data.seaice, and even the data.diagnostics, that I constructed from Gael's message, to run my configuration for 7 days, and so far things look good, except that the diagnostics 'oceSPflx' is not available, so I commented out the corresponding line in data.diagnostics. Also note, that I copied Gael's list of diagnostics to the DIAG_STATIS_PARMS namelist, so that building the budget might be a little easier. Whatever you prefer.
>> 
>> So, having in mind the deadline of Mar30, I will not keep you any longer from starting your homework: start your integrations with the parameters included in the attachment, I already have. Good luck! 
>> 
>> Martin
>> 
>> PS. I just hope that I didn't screw up (o:
> 
> 
>> _______________________________________________
>> 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

---
Patrick Heimbach | heimbach at mit.edu | http://www.mit.edu/~heimbach
MIT | EAPS 54-1420 | 77 Massachusetts Ave | Cambridge MA 02139 USA
FON +1-617-253-5259 | FAX +1-617-253-4464 | SKYPE patrick.heimbach