[MITgcm-devel] new salt_plume param?

Fri May 27 00:20:20 EDT 2011

Gael, if I understand your suggestion correctly, you are trying to add a redistribution
that would leave (1-SaltPlumeIntensity) in the surface level, in addition to whatever
is left there by the particular "PlumeMethod" that has been selected.
I think this is a good suggestion but a cleaner way to implement would be to
add one or more PlumeMethod distributions to salt_plume_frac.F in addition
to the ones that are already there.  For example, the current PlumeMethod=5
is  equivalent to your SaltPlumeIntensity=0 and the other PlumeMethod
distributions all leave behind some percentage of salt in the surface level.

Another argument against adding a constant SaltPlumeIntensity in seaice_growth.F
is that conceptually this parameter should not be everywhere constant.  Instead it
should probably be modulated by the fraction of lateral vs bottom freezing occurring
in each grid cell at a particular moment in the integration.  Again, salt_plume_frac.F
would be a better location to carry out such calculation rather than seaice_growth.F

Dimitris Menemenlis

On May 26, 2011, at 12:55 PM, Gael Forget wrote:

> Hi An and co,
> 
> my understanding is that the current set of parameters
> allows various redistribution profiles shape and reach.
> 
> I was surprised not to find one to tune, not the shape, 
> but the very intensity of the parameterization.  Did I miss it?
> 
> In my tests I added the following, which seems to provide for an 
> easy tuning of the salt plume parameterization strength. 
> I thought it could be of interest to other users. So, do you mind if 
> I introduce a SaltPlumeEfficiency factor in seaice_growth.F to do
>             saltPlumeFlux(I,J,bi,bj)=
>     &            HEFFM(I,J,bi,bj)/SEAICE_deltaTtherm
>     &            *(1-SIsalFRAC)*salt(I,j,kSurface,bi,bj)
>     &            *tmpscal1*SEAICE_rhoIce
>     &            *SaltPlumeIntensity
> It would range from 0 to 1, and have a default of 1. I believe the 
> above snippet would be a correct way to do this. Isn't it?
> 
> Cheers,
> Gael
> 
> 
> 
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-devel