[MITgcm-support] Model loosing salt

[Jody Klymak]

In my opinion, the easiest way to deal with issues like this is to make your boundary far from your control region by stretching the grid so that you have a very large reservoir.  If the forcing in the control region is steady state, then the model will eventually settle into a steady state between the control region and the reservoir.

Did you try applying a sponge layer on your southern boundary?  You can sponge T and S, and in your case that might add salinity to the domain faster than just prescribing the boundary condition at the southern most grid point.  

Cheers,   Jody

> On 27 Apr 2016, at  12:43 PM, Jean-Michel Campin <jmc at mit.edu> wrote:
> 
> Hi Abhisek and Martin,
> 
> Just a comment on this: 
>> One can imaging prescribing fluxes, 
> Just prescribing tracer fluxes is probably not the right way to go (and this is why
> it has not been implemented): any tracer anomaly that develops in the interior
> should be able to be advected away when approaching an outflow OB region.
> But prescribing tracer fluxes would prevent this, resulting in un-physical solution
> (+ with potential numerical problems) with tracer anomaly being trapped in the
> vicinity of OB outflow.
> 
> Cheers,
> Jean-Michel
> 
> On Wed, Apr 27, 2016 at 11:41:31AM +0200, Martin Losch wrote:
>> in the OBCS package, velocites, temperature and salinity are prescribed (it does not matter how they are provided, by Orlanski, Stevens or just from a file). There is no constraints on fluxes. One can imaging prescribing fluxes, but that would require a total rewrite of the code. Sorry ???
>> 
>> Martin
>> 
>>> On 26 Apr 2016, at 17:56, Abhisek Chakraborty <abhisek.sac at gmail.com> wrote:
>>> 
>>> Dear Martin,
>>> Thanks for your prompt response. Surely the atmospheric forcing is adding up non-zero freshwater flux. I have found the mean salinity is also decreasing. I am using OBCS_BALANCE with a non-zero sponge layer. I have tried with and without Orlanski OBC at my southern bounday but the results are similar. Is it always required to modify the code for balancing the tracers for a regional configuration of the model despite of using the standard OBCS package?
>>> With best regards,
>>> Abhisek
>>> 
> her> > On 26-Apr-2016 4:54 PM, "Martin Losch" <Martin.Losch at awi.de> wrote:
>>> Abhisek,
>>> 
>>> as far as I can see, you have an open boundary and you use atmospheric surface forcing. Both of them will most likely have non-zero freshwater flux that will change your mean salinity (btw, does the mean salinity also change, or only the surface salinity as shown in your plots?). You could balance the surface freshwater flux with a flag (balanceEmPmR=.TRUE.) at each timestep, but you'd have to code balancing the freshwater flux (implied salinity flux, salt*vVel) through the open boundary yourself.
>>> 
>>> Martin
>>> 
>>> On 04/26/2016 11:53 AM, Abhisek Chakraborty wrote:
>>> Dear All,
>>> 
>>> I am trying to configure Mitgcm (version 65r) for Bay of Bengal with a
>>> constant horizontal resolution of 2km. After 3 years of model runs
>>> (2013-2015) using daily ERA forcing, I found the model is losing
>>> salinity. The domain averaged salinity shows a decreasing trend. The
>>> trend is more prominent for the head Bay. I am using yearly run off
>>> climatology. A figure is attached to show my model domain and salinity
>>> trend.
>>> 
>>> The relevant data files are also attached.  Can anybody please help me
>>> pointing out the possible causes of this decreasing salinity trend?
>>> 
>>> Thanking in advance,
>>> Abhisek
>>> 
>>> 
>>> 
>>> _______________________________________________
>>> MITgcm-support mailing list
>>> MITgcm-support at mitgcm.org
>>> http://mitgcm.org/mailman/listinfo/mitgcm-support
>>> 
>>> 
>>> -- 
>>> Martin Losch
>>> Alfred Wegener Institute for Polar and Marine Research
>>> Postfach 120161, 27515 Bremerhaven, Germany;
>>> Tel./Fax: ++49(0471)4831-1872/1797
>>> 
>>> 
>>> _______________________________________________
>>> 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
>> 
>> 
>> _______________________________________________
>> 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

--
Jody Klymak    
http://web.uvic.ca/~jklymak/