cvmix short wave penetration

[milicak]

Hi,

I have a question in impementation of the CVmix short wave penetration in MOM6.
In the MOM_CVMix_KPP.F90 there is the following line;
surfBuoyFlux2(k) = buoy_scale * (buoyFlux(i,j,1) - buoyFlux(i,j,k+1))

where surfBuoyFlux2 gets the bouyancy flux that includes SW penetration values as well. However this part is in a full k=1:GV%ke do loop.
I thought the loop should be just in the boundary layer defined by KPP, not the whole column.

I don't know maybe the way buoyFlux is defined (maybe cumulative sum, or something different) might be the case,
but I was thinking surfBuoyFlux2 should have a profile where it is high at the surface layers, then it should goto zero in the deep ocean.
However since this is a full column loop deep layers will also have some forcing.

I guess I am missing something.

Best,
Mehmet

 

[gmarques]

Hi Mehmet,

This particular code is used in the calculation of the bulk Ri at all depths and that's why surfBuoyFlux2 is computed from k=1:GV%ke. I think surfBuoyFlux2 has the profile you said: "high at the surface layers, then it should goto zero in the deep ocean". The boundary layer is computed later in the code (see call to CVMix_kpp_compute_OBL_depth).

Best,
Gustavo