[MITgcm-devel] pkg/seaice budgets

[Martin Losch]

Hi Gael,

I agree that there is something wrong with the sw heat flux term:
in seaice_solve4temp F_ia (which become a_QbyAtm_cover in seaice_growth) does not contain the sw heat flux under the ice: sw heat is absorbed (F_swi), but the sw heat that penetrates cannot melt the ice. So far so good. When you compute QNET, should be all heat flux contributions: SW+rest over water and penetrating SW+rest under ice. Ian's line
#ifdef FENTY_DELTA_HEFF_OPEN_WATER_FLUXES 
     &         +   a_QSWbyATM_cover(I,J)
#endif /* FENTY_DELTA_HEFF_OPEN_WATER_FLUXES */          
adds the part under ice that is missing from F_ia (a/r_QbyAtm_cover). This part should also be added to the corresponding diagnostic SIatmQnt (didn't check that).
So I completely agree with Ian (and you): The CPP-flag should vanish and the contribution needs to be added to the heat flux diagnostics to close the budget. Will you do it?

2nd question: I used oceOnet, because Jean-Michel told me to do so (he used this in his tests for thsice). Why do you want to use TFLUX? Because my tests did not include restoring, they are the same anyway, right?

Martin

On Apr 1, 2011, at 11:26 PM, Gael Forget wrote:

> Martin,
> 
> two questions. 
> 
> - Ian argued that seaice_growth.F has been erroneously
> dropping a sw heat flux term (see line 1716 in r1.117, and 
> our Feb17th email exchange). It first appeared to break heat 
> your conservation checks because Ian did not add this term to 
> the SIatmQnt diagnostic that is used in those checks. However
> I think Ian is right that this term is missing. What do you think?
> 
> - is there a reason why you chose oceQnet rather than TFLUX 
> to test the ocean heat conservation? From my series of tests, and 
> MITgcm/doc/diags_changes.txt, it seems that TFLUX is 
> the one we want to use. Do you concur?
> 
> Have a good week end,
> Gael
> 
> 
> 
> 
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-devel