[MITgcm-support] Smagorinsky 3D tendency diagnostics

[Ruth Musgrave]

Hi Jean-Michel,

I agree that adding the Smag-3D contribution to Um_Diss and Vm_Diss
would be a consistent thing to do. As an update to my earlier email, it
seems like my attempt at adding smag-3D diagnostics was ok in the end -
my budgets weren't closing for other reasons having to do with how I was
computing the time derivative. I am happy to send you the changes I made
to the code, but I am sure that someone who knows what they're doing
would do it better than I did.

I searched online for the problems with no-slip and Smag-3D but didn't
find anything, but I would really like to hear what these problems are.
Could anyone point me in the right direction to find the discussion
online? Or maybe remembers what the specific issues are?

Thanks
Ruth


On 12/8/14, 7:56 PM, Jean-Michel Campin wrote:
> Hi Ruth,
> 
> Thanks for reporting the problem.
> I don't think there is any diagnostics for Smag-3D and we might
> want to add some.
> But regarding Um_Diss,Vm_Diss on one side and Wm_Diss on the other,
> I agree that this asymmetry should be fixed.
> 
> I propose to add the Smag-3D contribution to Um_Diss,Vm_Diss
> the same way the vertical momentum contribution is already 
> part of Wm_Diss.
> 
> Also one thing to keep in mind (I forgot if it has already be
> mentionned on this list): there are known problems in Smag-3D
> with no-slip BC (either no_slip_sides=T or no_slip_bottom=T)
> but I don't remember exactly which line of code is wrong !
> 
> Cheers,
> Jean-Michel
> 
> On Sun, Dec 07, 2014 at 03:02:12PM -0800, Ruth Musgrave wrote:
>> Hi,
>>
>> I am running internal wave type simulations using the Smagorinsky 3D
>> turbulence closure, however I think that there might be some
>> inconsistencies in the momentum dissipation diagnostics when Smag 3D is
>> turned on.
>>
>> Looking in dynamics.F, it seems like Um_Diss and Vm_Diss are filled as
>> diagnostics in mom_fluxform before mom_uv_smag_3d is called, which means
>> that they don't include the smag3D contribution to the tendency.
>> However, Wm_Diss is filled in gw_calc after a call to mom_w_smag_3d,
>> which means it does include the smag 3d contribution. I can't find any
>> specific diagnostics relating to 3D Smagorinsky to disentangle all this.
>>
>> I have tried "fixing" the above by adding more diagnostics specifically
>> for the Smag 3D terms, but my energy budgets still don't close, so I'm
>> probably totally off track, or doing it wrong. Any thoughts/guidance
>> would be greatly appreciated.
>>
>> Thanks,
>> Ruth
>>
>> _______________________________________________
>> MITgcm-support mailing list
>> MITgcm-support at mitgcm.org
>> http://mitgcm.org/mailman/listinfo/mitgcm-support
> 
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support
>