[MITgcm-support] Vertical temperature profile tRef

[Martin Losch]

Hi Kevin,

I am assuming that you are using the linear equation of state, because (in the MITgcm) the reference profile only makes sense in this context.
hydrogThetaFile should contain the initial temperature field theta (no anomaly or perturbation), can be a 3D field

tRef contains a vertical profile and I think it makes no difference whether you use the namelist parameter tRef or tRefFile to specify it. Density (anomaly) is then computed as
rho - rhoConst = -rhoNil*tAlpha*(T - tRef) + salinity contribution. + rhoNil-rhoConst  (in most cases rhoNil-rhoConst=0)

If you do not specify a file for hydrogThetaFile, then tRef is also used to initialize theta. In that case rho-rhoConst = 0 initially, and you get a marginally stable state.

Martin

PS for a non-linear EOS, tRef is ignored.

> On 12. May 2020, at 13:23, Kevin Ha <kevin.ha.pro at gmail.com> wrote:
> 
> Dear all,
> 
> I am trying to implement a configuration with a confined internal gravity wave in the ocean without topography with boundary conditions w=0 at z=0 and z=-H.
> 
> My question is : if I want a vertical reference profile in temperature (my medium is stratified in temperature, but not in salinity), should I specify it only as a binary file in tRefFile or should I do it both in tRefFile and hydrogThetaFile ? 
> 
> I know the code is written in terms of perturbation field but when I do it in terms of tRefFile only, the results don't match the theoretical prediction whereas when I do it on both, meaning that :
> - tRefFile is my vertical profile T(z)=N²/(alpha*g)*(z+H),
> - hydrogThetaFile is the sum of my vertical profile T(z)=N²/(alpha*g)*(z+H) and the initial condition for the potential temperature of the confined wave T(x,z) = T0.*sin(kz0.*z).*cos(kx0.*x), 
> the results are quite better in this last case. I need to be clarified with the difference between tRefFile and hydrogThetaFile, which I suspect to be respectively the reference profile and the perturbation ONLY.
> 
> Here is some of the key files of my configuration.
> 
> Thank you in advance,
> 
> Best
> 
> Kevin
> <packages.conf><gendata.m><data><ini_fields.F><SIZE.h><CPP_OPTIONS.h><FFIELDS.h><MOM_FLUXFORM_OPTIONS.h>_______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support