[MITgcm-devel] [MITgcm-cvs] MITgcm/model/src CVS Commit

Patrick Heimbach heimbach at MIT.EDU
Thu May 22 09:19:18 EDT 2014


Hi again,

so the last checking (kpp_forcing_surf.F) appears to fix all forward problems (I reran on baudelaire), with and without use of salt_plume. The culprit was another recip_Cp replacement that I had missed. Will test the adjoints soon.

By the way, the testreport page doesn’t indicate seaice_itd to have had problems, no?

Y Y Y Y 16 16 16 16 16 16 16 16 13 13 16 16 16 16 12 16 16 16 16 16 22 14 14>16< 0  .  .  .  .  .  .  .  .  .  .  .  . pass  seaice_itd
Y Y Y Y 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 22 16 16>16<16  .  .  .  .  .  .  .  .  .  .  .  . pass  seaice_itd.growth
Y Y Y Y 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16>16<16  .  .  .  .  .  .  .  .  .  .  .  . pass  seaice_itd.melt

Cheers
p.

On May 22, 2014, at 8:55 AM, Jean-Michel Campin <jmc at ocean.mit.edu> wrote:

> Hi Patrick,
> 
> Your plan sounds good to me. Let's focus first on forward exp (kpp ?)
> and regarding baltic test, Oliver will let us know.
> 
> Cheers,
> Jean-Michel
> 
> On Thu, May 22, 2014 at 08:46:11AM -0400, Patrick Heimbach wrote:
>> Hi Jean-Michel,
>> 
>> argh, looks like I got caught (preaching to test thoroughly before checking in, but being negligent myself). An issue is that there’s some pressure to get this sorted out before the weekend for a number of reasons.
>> 
>> How about this:
>> 1. we keep the changes in model/src/ (there shouldn’t be a problem now after the fixes)
>> 2. we also keep the changes in pkg/salt_plume/ for now
>> 3. I will try to sort out the issue in kpp today (I agree it’s likely to come from there).
>> 
>> Also, I don’t know anything about 
>>> monod_baltic_76x72x24 (x1)
>>> darwin_baltic_76x72x24 (x1)
>> so can’t test these.
>> 
>> Cheers
>> p.
>> 
>> On May 22, 2014, at 7:48 AM, Jean-Michel Campin <jmc at ocean.mit.edu> wrote:
>> 
>>> Hi Patrick,
>>> 
>>> After putting back local variable recip_Cp, the results are more
>>> similar to what they used to be.
>>> However, it's still not very clear that some experiments which do not
>>> even compile pkg/salt_plume are giving different output.
>>> For forward test:
>>> lab_sea (x4, only 1 test uses pkg/salt_plume)
>>> natl_box (x2)
>>> seaice_itd (x1)
>>> seaice_obcs (x1, but all 3 tests use pkg/salt_plume)
>>> monod_baltic_76x72x24 (x1)
>>> darwin_baltic_76x72x24 (x1)
>>> 
>>> I would suggest that we revert back to code that produces exactly
>>> the same outputs when pkg/salt_plume is not compiled.
>>> The reasons being that:
>>> 1) if we need to make changes outside ALLOW_SALT_PLUME, they 
>>> should be tested + checked-in + documented separately from
>>> major changes within pkg/salt_+plume
>>> 2) in term of following why/when results change (not only now, but later
>>> on when some user asks the usual question: "I updated my code
>>> and I don't now why I got different results"), it would be good to isolate
>>> within few lines of code changes what is the cause of different output.
>>> 3) the issue here is very likely related to pkg/kpp changes.
>>> And since this pkg has other issues (e.g., case where non local
>>> term is problematic, cf Dimitris experience with very high res. set-up),
>>> I would recommend being specially careful here.
>>> 
>>> Cheers,
>>> Jean-Michel
>>> 
>>> On Wed, May 21, 2014 at 11:30:11AM -0400, Patrick Heimbach wrote:
>>>> Sorry, I may have misunderstood the purpose of your removing
>>>> recip_Cp as a global variable and making it local-only in few routines.
>>>> In merging it with An’s code (which still dates from the time when it was global),
>>>> I simply removed it altogether.
>>>> If that was not the intention, I’m happy to put it back and make it local.
>>>> 
>>>> Cheers
>>>> p.
>>>> 
>>>> On May 21, 2014, at 11:26 AM, Jean-Michel Campin <jmc at ocean.mit.edu> wrote:
>>>> 
>>>>> Hi Patrick,
>>>>> 
>>>>> I don't know why you decide to remove the local variable
>>>>> "recip_Cp" from external_forcing.F & external_forcing_surf.F
>>>>> but I have the impression that this affects many experiments, 
>>>>> with few output.txt (I counted 17) that would need an update 
>>>>> (no longer 13 digits).
>>>>> 
>>>>> Cheers,
>>>>> Jean-Michel
>>>>> 
>>>>> Wed, May 21, 2014 at 06:44:59AM -0400, Patrick Heimbach wrote:
>>>>>> Update of /u/gcmpack/MITgcm/model/src
>>>>>> In directory forge:/tmp/cvs-serv19798/model/src
>>>>>> 
>>>>>> Modified Files:
>>>>>> 	do_oceanic_phys.F external_forcing.F external_forcing_surf.F 
>>>>>> Log Message:
>>>>>> An Nguyen's extensions to salt_plume package
>>>>>> (carry corresponding heat flux along with salt redistribution)
>>>>>> 
>>>>>> 
>>>>>> _______________________________________________
>>>>>> MITgcm-cvs mailing list
>>>>>> MITgcm-cvs at mitgcm.org
>>>>>> http://mitgcm.org/mailman/listinfo/mitgcm-cvs
>>>>> 
>>>>> _______________________________________________
>>>>> 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
>>>> 
>>> 
>>> 
>>> 
>>>> _______________________________________________
>>>> 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
>> 
> 
> 
> 
>> _______________________________________________
>> 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

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 1588 bytes
Desc: not available
URL: <http://mitgcm.org/pipermail/mitgcm-devel/attachments/20140522/c014f909/attachment-0001.p7s>


More information about the MITgcm-devel mailing list