[MITgcm-support] Surface waves relaxation without viscosity terms

[Oliver Jahn]

Hello Quentin,

I think you solved your problem, but just to clarify (and at the risk of 
stating the obvious) I'd like to point out that for simple monthly 
averages you do not need averagingFreq nor repeatCycle, but just 
frequency(1) = 2592000.  I am getting the impression that this may be 
what you want, but maybe I am wrong.

averagingFreq and repeatCycle are basically for climatologies: you want 
to average the January of each year and then only write the average of 
all the Januaries, all the Februaries, etc, i.e. only 12 averages for 
several years.  If frequency is also set to a number of years smaller 
than the model run, you will get multiple climatologies.  Say frequency 
is 10 years.  Then you get a separate climatology for every decade.  And 
yes, in the mnc file there will be funny time values, clustered at the 
end of each decade.  But for a climatology, there isn't really a single 
"time of averaging" for each record, but several.

Cheers,
Oliver



On 2013-01-31 05:55, quentin jamet wrote:
> Hello Jean-Michel,
>
> I made a mistake when copying the data.diagnostics file. I did remove
> the frequency=50yearfor my previous tests.
>
> I made better tests, and I think I fanally find the way out : when using
> mnc file format combined to repeatCycle, averaging diagnostics are saved
> at the end of the cycle. This results in a model_time 'T' into the mnc
> file which corresponds to the date of saving instead of the date of
> averaging.
>
> I hope I made myself clear enought.
>
> Thanks again
>
> Quentin
>
> 2013/1/30 Jean-Michel Campin <jmc at ocean.mit.edu <mailto:jmc at ocean.mit.edu>>
>
>     Hi Quentin,
>
>     I sometimes uses this repeatCycle and averagingFreq feature
>     but not with NetCDF files (although I thought I checked that
>     it was also working with mnc).
>
>     One thing that I found curious in your data.diagnostics file is
>     the frequency (50 yrs) which is much longer than the simulated
>     time (4.yrs). Don't know if this has something to do with your problem.
>
>     Also, for the 2nd output file, should be:
>      >    fields(1:4,2) = 'THETA   ','SALT    ','PHIHYD  ','CONVADJ ',
>     instead of:
>      >    fields(1,2) = 'THETA   ','SALT    ','PHIHYD  ','CONVADJ ',
>
>     Cheers,
>     Jean-Michel
>
>     On Tue, Jan 29, 2013 at 11:33:18AM +0100, quentin jamet wrote:
>      > Dear MITgcm developpers and users,
>      >
>      > I have a problem using the diagnistic options repeatCycle and
>     averagingFreq.
>      > When runing with this configuration for 4 years, variables within
>     the output
>      > files DiagOcnEta.*.nc contains the good dimensions, i.e. 48 = 12*4. I
>      > expect to get one average output each month during my 4 year
>     simulation, but
>      > by looking at the *.nc file, the number of iteration at which
>     diagnostics
>      > are made correpond to 1 diagnostics each year, and 1 diagnostic
>     at the 12
>      > previous iterations, but not 1 diagnostic each month.
>      > I did some test but I did not find out the answer. Did anyone
>     have asolution to
>      > fix my problem?
>      >
>      > Many thanks
>      >
>      > Quentin
>      >
>      > PS : I copy a part of the file data.diagnostics below :
>      >
>      >   frequency(1) = 31104000.,
>      >   frequency(1) = 1555200000.,
>      > # frequency(1) = 2592000.,
>      >    filename(1) = 'DiagOcnEta',
>      >    fields(1,1) = 'ETAN    ',
>      >  averagingFreq(1) = 2592000.,
>      >    repeatCycle(1) = 12,
>      >
>      >   frequency(2) = 31104000.,
>      >   frequency(2) = 1555200000.,
>      > # frequency(2) = 2592000.,
>      >    filename(2) = 'DiagOcnTave',
>      >    fields(1,2) = 'THETA   ','SALT    ','PHIHYD  ','CONVADJ ',
>      > #                'UVELSQ  ','VVELSQ  ','UV_VEL_C','THETASQ ',
>      > #                'WVELTH  ','WVELSLT ',
>      >  averagingFreq(2) = 2592000.,
>      >    repeatCycle(2) = 12,
>      >
>      >
>      > --
>      > Quentin Jamet
>
>      > _______________________________________________
>      > MITgcm-support mailing list
>      > MITgcm-support at mitgcm.org <mailto:MITgcm-support at mitgcm.org>
>      > http://mitgcm.org/mailman/listinfo/mitgcm-support
>
>
>     _______________________________________________
>     MITgcm-support mailing list
>     MITgcm-support at mitgcm.org <mailto:MITgcm-support at mitgcm.org>
>     http://mitgcm.org/mailman/listinfo/mitgcm-support
>
>
>
>
> --
> Quentin Jamet
>
>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support
>