[MITgcm-support] Eta, ETAN and SIheff 2

Tue Sep 2 08:05:32 EDT 2008

Elja,

Eta as in the output of write_state.F is an instantaneous snap shot  
of the model variable etaN.
'ETAN    ' as the diagnostics variable can be snap shot but is most  
likely a  time average of etaN (if the corresponding frequency was  
positive), that's most likely the cause of the difference. Otherwise  
they should be the same.

a parameter that describes the total deformation of the system: Not  
quite clear what you want, please have a look at the definition of  
phiSurf, e.g., in seaice_dynsolver.F:
- if .not.realfreshwaterflux, then ETAN is the surface elevation  
according to dynamics "as if" there were no ice (put e.g. atmospheric  
pressure loading is taking into account).
- if realfreshwaterflux, then ETAN is the surface underneath the ice,  
depressed by the load of the atmosphere, and the ice mass.
phiSurf is the dynamically relevant variable, and you can construct  
that from the diagnostics variables 'ETAN    ', 'atmPload', and  
'sIceLoad' according to its definition in, e.g. seaice_dynsolver.F  
(and other places).

when your entire ocean is covered by ice with constant thickness,  
then  there is no effect, only horizontal gradients matter, in the  
same way that the atmospherical pressure load has not effect, unless  
it contains horizontal gradients (which is normally the case).

Martin

On 2 Sep 2008, at 12:17, Elja Huibregtse wrote:

> Hi Martin,
>
> In addition to my previous mail, I still have another question:
>
> - You wrote that:
> In the presence of sea ice it is the surface underneath sea ice.  
> For a particular
> parameter configuration (#define ATMOSPHERIC_LOADING and
> useRealFreshWaterFlux=.true.,) the free surface is actually depressed
> by ice load (so that ice is not floating "on top" of the ocean but  
> really in the ocean).
>
> In my case, the global ocean is completely covered with ice. So, if  
> the water is depressed by the ice load, where does the water go to?  
> Or is it assumed that water is compressible?
>
> Thanks, I really appreciate your help!
> Elja
>
> %----------------------------
>
> Hi Martin,
>
> Thank you for your reply.
> 'My' definitions of ETA and ETAN are:
>
> * ETA is the output defined in write_state.f:
>  CALL WRITE_FLD_XY_RL('Eta.',suff,etaN,myIter,myThid)
>   Now, I see that the variable 'etaN' is written to ETA.
>
> * ETAN follows from the diagnostics package (for example, in  
> data.diagnostics)
>   filename(1)  = 'ETAN'
>
> Since, in my case, the output data for ETA and ETAN significantly  
> differ from each other, I thought that one of them described the  
> excitation of the ice layer and that the other one described the  
> excitation of the ocean. This is probably incorrect...
>
> Should ETA be equal to ETAN? If so, do you have some (general)  
> suggestions what could be wrong in case they are not equal?
> And secondly, does a parameter exist that describes the total  
> deformation of the system, that is the deformation of the ocean and  
> ice? Or is this (simply) ETAN+SIheff?
>
> Thank you in advance,
>
> Elja
>
>
> -----Oorspronkelijk bericht-----
> Van: mitgcm-support-bounces at mitgcm.org namens Martin Losch
> Verzonden: di 9/2/2008 9:45
> Aan: mitgcm-support at mitgcm.org
> Onderwerp: Re: [MITgcm-support] Eta, ETAN and SIheff
>
> Elja,
>
> what do you mean by Eta and EtaN? From DYNVARS.h:
> C     etaN  - free-surface r-anomaly (r unit) at current time level
> so this is the name of the model variable for SSH. In the presence of
> sea ice it is the surface underneath sea ice. For a particular
> parameter configuration (#define ATMOSPHERIC_LOADING and
> useRealFreshWaterFlux=.true.,) the free surface is actually depressed
> by ice load (so that ice is not floating "on top" of the ocean but
> really in the ocean).
>
> what is "Eta" in your question?
>
> You can try to initialize SIheff (effective ice thickness = volume/
> grid-cell are) from a file, and with useSEAICE=.true. set
> usePW79thermodynamics = .false.,
>        SEAICEadvHeff      = .false.,
>        SEAICEadvArea      = .false.,
>        SEAICEadvSnow      = .FALSE.,
>        SEAICEadvSalt      = .FALSE.,
> SEAICEuseDYNAMICS = .false., ! only to turn off the computation of
> unused drift velocities.
>
> You'll still have some overhead from the ice model, but ice thickness
> should not change.
>
> Martin
>
> On 1 Sep 2008, at 13:12, Elja Huibregtse wrote:
>
> > Hi Matt,
> >
> > Thanks for your reply!
> >
> > Actually, my results for Eta and Etan differ significantly from
> > each other. I modeled an ice layer on top of an ocean, and now I am
> > wondering whether Eta gives the displacement on top of the ocean
> > and Etan on top of the ice layer. Does this make sense?
> >
> > Regards,
> > Elja
> >
> >
> >
> >
> >
> >
> >
> > -----Oorspronkelijk bericht-----
> > Van: mitgcm-support-bounces at mitgcm.org namens Matthew Mazloff
> > Verzonden: zo 8/31/2008 8:52
> > Aan: mitgcm-support at mitgcm.org
> > Onderwerp: Re: [MITgcm-support] Eta, ETAN and SIheff
> >
> > Hi Elja
> >
> > 1)  I think eta = etan ....not sure where discrepancies (if any)
> > come in
> > SIheff is the effective ice thickness...the relationship would  
> depend
> > on surface boundary condition using...
> >
> > 2) sure you can fix SIheff, but i doubt there's a flag to do this.
> > You'll have to hardcode a resetting of SIheff in the model...
> >
> > probably not to helpful...but thought id throw out a response
> > -matt
> >
> > On Aug 29, 2008, at 11:35 AM, Elja Huibregtse wrote:
> >
> > > Dear all,
> > >
> > > I have two (trivial?) questions concerning modeling of ice:
> > >
> > > 1. What is the exact relation between Eta, ETAN and SIheff?
> > > 2. Is it possible to fix the thickness of the ice, i.e. is it
> > > possible to make the ice thickness independent of the ocean
> > > temperature?
> > >
> > > Thank you very much in advance!
> > >
> > > Elja
> > >
> > >
> > >
> > > _______________________________________________
> > > 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