[MITgcm-support] Problem regarding how to add SSS relaxation

[Jean-Michel Campin]

Hi Ruizhi,

I would suggest to check the surface-salinity field you are relatxing to, i.e., the 
one you provide from a file. It should not contain unrealistic values or NaN where
you have ocean.

And you can also check the magnitude of the relaxation tendency after the first time-step
by turning on diagnostics "SRELAX  " (with a frequency equal to your deltaT).

Cheers,
Jean-Michel

On Fri, Mar 13, 2026 at 01:52:26PM +0800, ?????? wrote:
>  Dear MITgcm community,
> 
>  I am currently running a regional model for the Southern Ocean with a horizontal resolution of ~10 km, using the r* coordinate system (without the sea ice package at this stage). The model runs stably under normal conditions, but it crashes within the first few time steps right after I introduce SSS relaxation (climsssTauRelax is set to 60 days).  
> 
>  Currently, I have exact volume conservation enabled in my setup. The model crashes regardless of whether the real freshwater flux is enabled or not. I am wondering what the potential causes of this crash might be. Could there be a conflict between the SSS relaxation and some of my other settings?
> 
>  I have attached my data and data.obcs files for your reference. Thank you very much for your time and support. Any advice on how to properly set up these variables together would be greatly appreciated!  
> 
>  Best regards,  
>  Ruizhi Xu  



> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support