[MITgcm-support] MITgcm-support] what is obcs balance?

[David Hebert]

It appears the way the code is written the sum of the flux across each 
face is zero. I don't see anywhere in the code where the sum from all 
faces is used to make any sort of correction to OBC values.  I agree 
completely that it is the total flux in/out of a domain that is of 
interest, not each  face independently.

David

Nico Wienders wrote:
>
>
> I still don't understand. LOW_OBCS_BALANCE=.TRUE.
> means:
>
> that the the flow across each face with be equal to zero?
>
> or
>
> That the sum of the flow across each (4) face will be zero?
> (which would have more physical justification).
>
>
>
>
>
>
> On Jul 14, 2009, at 10:17 AM, Martin Losch wrote:
>
>> David,
>>
>> you interpreted the code correctly, and you also observed correctly 
>> that this is not documented in the manual.
>>
>> Martin
>>
>> On Jul 14, 2009, at 3:36 PM, David Hebert wrote:
>>
>>> Hi everyone,
>>>
>>> Thanks Jean-Michel for clarifying the #undef NONLIN_FRSURF flag.
>>>
>>> I am now looking at obcs_calc.F and the portions related to 
>>> ALLOW_OBCS_BALANCE. If I am not mistaken, if I use 
>>> ALLOW_OBCS_BALANCE, the code computes the flux through one 
>>> particular face as the sum of  area * OB[EWNS][uv].  If the sum over 
>>> the whole face is not zero, the code then subtracts the sum/area off 
>>> that particular face, forcing the boundary flux to be zero. Is this 
>>> true? If not, can someone perhaps provide a description of what is 
>>> happening? I am not seeing this description in the manual.
>>>
>>> Thank you,
>>>
>>> David