回复： Re: [MITgcm-support] parameter set
Maxiaohui
maxiaohui823 at yahoo.cn
Tue Sep 2 10:21:19 EDT 2008
Hi,Martin
Yes, you are right. I put eedata in the other directory.
Now I meet a error which I haven't met before.
forrtl: severe (17): syntax error in NAMELIST input, unit 11, file /tmp/pbs.5256.a330/fortc6LmOR, line 40, position 2
Image PC Routine Line Source
mitgcmuv 40000000004D74D0 Unknown Unknown Unknown
mitgcmuv 40000000004D4FE0 Unknown Unknown Unknown
mitgcmuv 400000000044AA10 Unknown Unknown Unknown
mitgcmuv 40000000003B4710 Unknown Unknown Unknown
mitgcmuv 40000000003B36A0 Unknown Unknown Unknown
mitgcmuv 40000000003F2810 Unknown Unknown Unknown
mitgcmuv 4000000000342780 Unknown Unknown Unknown
mitgcmuv 4000000000363300 Unknown Unknown Unknown
mitgcmuv 4000000000397010 Unknown Unknown Unknown
mitgcmuv 4000000000296970 Unknown Unknown Unknown
mitgcmuv 4000000000004150 Unknown Unknown Unknown
libc.so.6.1 2000000003E85C50 Unknown Unknown Unknown
mitgcmuv 4000000000003F40 Unknown Unknown Unknown
MPI: MPI_COMM_WORLD rank 0 has terminated without calling MPI_Finalize()
MPI: aborting job
So what does this mean？ What is wrong?
Xiaohui
Martin Losch <Martin.Losch at awi.de> 写道：
you should set
OB_Jsouth=121*1,
OB_Ieast=61*-1,
in the vertical all points are assumed to be a boundary point if the
surface points is a boundary poiint. OB_Jnorth and Iwest should be
zeros (the default) if there are no open boundaries there.
[ you can also, if only part of your, e.g., southern boundary is
open boundary, set OB_Jsouth = 10*1,50*0,20*2,40*3,1*0. In this
example the first ten points in x-direction at J=1 would be open
boundary points, the next 50 points would be solid or periodic
boundaries depending on your land mask, the 20 points would be
boundary points at J=2 and then 40 points at J=0 and then another
solid/periodic boundary point. the -1 is equivalent to Nx/Ny, so in
your case OB_least=61*121 is the same as 61*-1. ]
eedata: the error message is obvious, somehow the file eedata could
not be openend. Read permissions? Is the file in the correct
directory (where you run the executable)? That's usually the problem.
Martin
On 2 Sep 2008, at 15:32, Maxiaohui wrote:
> Hi Martin,
>
> Thank you for your suggestion.
> I noticed that in the data.obcs of exp4,
> OB_Jnorth=80*-1,
> OB_Jsouth=80*1,
> OB_Ieast=42*-1,
> OB_Iwest=42*1,
> that means all the points on the boundary must have u v
> values,right? Since my experiment has land points at the boudary,
> moreover, different layers have different land points(there are 23
> layers in my experiment), then how should I set
> OB_Jnorth ,OB_Jsouth,OB_Ieast,OB_Iwest ?
>
> One more qustion ,when I run this experiment for a try , I got the
> followeing error:
> S/R EESET_PARMS
> Unable to open execution environment
> parameter file "eedata"
> ABNORMAL END: S/R EESET_PARMS
> MPI: MPI_COMM_WORLD rank 0 has terminated without calling
> MPI_Finalize()
> MPI: aborting job
> Is this related to my size.h file? My size.h file is:
> & sNx = 121,
> & sNy = 61,
> & OLx = 3,
> & OLy = 3,
> & nSx = 1,
> & nSy = 1,
> & nPx = 1,
> & nPy = 1,
> & Nx = sNx*nSx*nPx,
> & Ny = sNy*nSy*nPy,
> & Nr = 23)
>
> and the eedata file is:
> # Example "eedata" file
> # Lines beginning "#" are comments
> # nTx - No. threads per process in X
> # nTy - No. threads per process in Y
> &EEPARMS
> nTx=1,
> nTy=1,
> &
> # Note: Some systems use & as the
> # namelist terminator. Other systems
> # use a / character (as shown here).
> Could you tell me what is going wrong? Thank you very much!
>
>
> sincerely,
>
> Xiaohui
>
>
>
>
>
>
>
>
> Martin Losch 写道：
> Hi Xiaohui,
>
> in "data":
> to set the monthly interval between forcing field records (months) set
> externForcingPeriod=2592000.,
> to have your twelve month forcing repeat year after year set
> periodicExternalForcing=.TRUE.,
> externForcingCycle=31104000.,
> These values are then also valid for the obcs-boundary conditions. In
> other words, in "data.obcs" you need to specify the boundary value
> files, just as in exp4 and that's it. This is convenient, but
> inflexible, because it requires that all your forcing and boundary
> values have to be available at the same forcing "frequency/period",
> but you have that.
>
> Alternatively you could use the exf/cal package to be able to set the
> period/cycle indiviually for each external field, but for a simple
> configuration I don't recommend that.
>
> Martin
>
> On 1 Sep 2008, at 03:24, Maxiaohui wrote:
>
> > Hi all,
> >
> > I am now trying an experiment in an (121x61x23） area(resolution:
> > 1/6x1/6°) with open boundary. I have prepared u, v t s (121x23x12
> > or 61x23x12) at the south and east boundary with twelve
> > months,since the north and west boundary are on the land .I also
> > use heatflux ,freshwater flux and wind filed(121 x61 x12) as the
> > external forcing and t s field( 121 x61x23 12) as the initial
> > field, all these files are twelve months.
> > Now I am confused at how to set the parameters.Could anyone tell
> > me how to set the parameters in data.obcs and data file? Could you
> > give me an example?
> > I will appreciate your help very much!
> >
> > Xiaohui
> >
> > 雅虎邮箱，您的终生邮箱！
> > _______________________________________________
> > 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
>
>
> 雅虎邮箱，您的终生邮箱！
> _______________________________________________
> 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
---------------------------------
雅虎邮箱，您的终生邮箱！
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mitgcm.org/pipermail/mitgcm-support/attachments/20080902/2224e79a/attachment.htm>
More information about the MITgcm-support
mailing list