[MITgcm-support] Understanding the shortwave radiation

[Martin Losch]

Marcello,

1. the shortwave heat can penetrate into the ocean while the remaining  
part of the heatflux cannot and is only applied to the surface  
(layer). That's why you have extra addtion to gT in external_forcing_t  
(and not only to the surface layer).
2. qnet is really qnet. If shortwave_heating is undefined all of this  
qnet is added to the surface layer (surface flux), if qsw is defined  
only (qnet-qsw) is added to the surface layer (see surfaceForcingT in  
external_forcing_surf.F, I think that that is what slipped through  
your careful scrutiny.)

Martin


On Jan 8, 2010, at 10:03 PM, Marcello Gatimu Magaldi wrote:

> Hi folks,
>
> sorry for the clutter, I am a little bit confused by how the  
> shortwave flux is handled in the model.
>
> Ok, first my relevant configuration. I am providing downward short  
> and longwave
> rads, air temperature, rel humidity and winds. Since KPP is used, I  
> am defining SHORTWAVE_HEATING too.
> I tracked the code and I ended up with:
> 1) downward short and longwave rads are converted respectively in  
> upward net short and longwave
> rad in exf_radiation.F.
> 2) sensible and latent heat fluxes are calculated using model fields  
> in exf_bulkformulae.F
> 3) the total upward net heat flux is calculated in exf_getforcing.F  
> summing these contributions. Here the first
> confusing thing: if SHORTWAVE_HEATING is defined the shortwave flux  
> is not added with the other fluxes
> but few lines later. Anyway, exiting exf_getforcing.F the total  
> upward net heat flux (which it seems to me that ALWAYS includes the  
> shortwaves)
> is assigned to the variable qnet via exf_mapfields.F. Similarly the  
> upward net shortwave flux is assigned to qsw.
> 4) the heat flux is used in external_forcing_surf.F (which is called  
> by do_oceanic_phys.F) to change surfaceForcingT.
> surfaceForcingT is ultimately the variable by which the fluxes enter  
> the theta-equation since it affects gT (via subroutine
> external_forcing_T.F called by calc_gt.F which is in turn called by  
> thermodynamics.F... uff... complicated! A beer will help!).
> However, here the second confusing thing: if SHORTWAVE_HEATING is  
> defined the model adds Qnet (ok!) but again
> Qsw! But Qnet already contains shortwaves, why is Qsw added a second  
> time? I am sure the code is right. I am just not
> understanding.
> 5) Third confusing thing: there is another place in  
> external_forcing.F where Qsw is added this time directly to gT  
> always if
> SHORTWAVE_HEATING is defined.
>
> I am sure I am missing something. Can somebody explain the rationale  
> behind this multiple addition?
> Thanks in advance,
> m.
>
>
> -- 
> Marcello Gatimu Magaldi,
> ---------------------------------------------------------------
> The Johns Hopkins University
> Department of Earth and Planetary Sciences
> 328 Olin Hall
> 34th and North Charles Streets
> Baltimore, Maryland 21218, USA
> e-mail: < Marcello.Magaldi at jhu.edu >
> ---------------------------------------------------------------
>
>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support