[MITgcm-support] Backtrace error while implementing NOBCS

[kunal madkaiker]

Thank you Gus. Martin, rightly pointed out, the error was caused in the
NOBC boundary file as I had 1 open point (1x48x2), but it was not in the
required 660x48x2 format. I added zeros for every other point and it
worked! Thanks a lot Martin.

I am trying to keep 19 open points in NOBC boundary through data.obcs:
 OB_Jnorth(550)=-100,
 OB_Jnorth(551)=-100,
 OB_Jnorth(552)=-99,
 OB_Jnorth(553)=-98,
 OB_Jnorth(554)=-98,
 OB_Jnorth(555)=-97,
 OB_Jnorth(556)=-97,
 OB_Jnorth(557)=-97,
 OB_Jnorth(558)=-97,
 OB_Jnorth(559)=-97,
 OB_Jnorth(560)=-97,
 OB_Jnorth(561)=-96,
 OB_Jnorth(562)=-96,
 OB_Jnorth(563)=-95,
 OB_Jnorth(564)=-95,
 OB_Jnorth(565)=-95,
 OB_Jnorth(566)=-95,
 OB_Jnorth(567)=-94,
 OB_Jnorth(568)=-94,
 OB_Jsouth=660*1,
 OB_Iwest=480*1,
 OB_Ieast=480*-1,

Is there a simpler way to feed it? I referred to the OBCS section in the
manual and tried a few things but it didn't work out. Any suggestions?

Regards,
Kunal

On Thu, Aug 19, 2021 at 7:32 AM Martin Losch <Martin.Losch at awi.de> wrote:

> Kunal,
>
> the error message (as usual) means that there are not enough records in
> your input file (‘OBNu.seaice_obcs’)
> you domain dimensions are (according to STDOUT.0000):
> (PID.TID 0000.0001)      nPx =   10 ; /* No. processes in X */
> (PID.TID 0000.0001)      nPy =    6 ; /* No. processes in Y */
> (PID.TID 0000.0001)      nSx =    1 ; /* No. tiles in X per process */
> (PID.TID 0000.0001)      nSy =    1 ; /* No. tiles in Y per process */
> (PID.TID 0000.0001)      sNx =   66 ; /* Tile size in X */
> (PID.TID 0000.0001)      sNy =   80 ; /* Tile size in Y */
> (PID.TID 0000.0001)      OLx =    4 ; /* Tile overlap distance in X */
> (PID.TID 0000.0001)      OLy =    4 ; /* Tile overlap distance in Y */
> (PID.TID 0000.0001)      nTx =    1 ; /* No. threads in X per process */
> (PID.TID 0000.0001)      nTy =    1 ; /* No. threads in Y per process */
> (PID.TID 0000.0001)       Nr =   48 ; /* No. levels in the vertical   */
> (PID.TID 0000.0001)       Nx =  660 ; /* Total domain size in X ( =
> nPx*nSx*sNx ) */
> (PID.TID 0000.0001)       Ny =  480 ; /* Total domain size in Y ( =
> nPy*nSy*sNy ) */
>
> So your input file for the OBCS pkg for the northern or southern boundary
> for a 3D field like temperature needs to have a field with the dimensions
> (t,z,x) = (nrec,48,660) and for the eastern/western boundaries
> (nrec,48,480). nrec needs to be at least 1 (for constant obcs forcing).
> This is independent of the number of actual boundary points. I.e. if you
> have only one open boundary point (OB_Jnorth(550)=-100,) your file still
> need the entire array, it can be zero everywhere except for the point i=550
> (for i starting at 1, for i starting a 0 as in python, it should be at 549).
>
> For a 2D field like sea ice velocity you drop the z-dimension (but I guess
> you only give you files these names, and don’t use sea ice).
>
> does that make sense?
>
> Martin
> > On 18. Aug 2021, at 15:44, kunal madkaiker <kunal.madkaiker02 at gmail.com>
> wrote:
> >
> > Dear All,
> >
> > I am using MITgcm for regional analysis. I have kept a single OB point
> on the northern boundary. But I keep getting backtrace error and model
> blows.
> >
> > Error termination. Backtrace:
> > At line 1946 of file mdsio_read_section.f (unit = 9, file =
> 'OBNu.seaice_obcs')
> > Fortran runtime error: Non-existing record number
> >
> > I have gone through similar asked questions on the support group and
> checked accordingly. The dimensions for my NOBCS pt is 1x48x2 (nx x nz x
> nt). I have prepared the T,S,U and V data and written as real*8. But still
> error persists.
> > Attaching my data, data.exf, data.obcs files. Kindly assist.
> >
> > Regards,
> > Kunal
> >
> <data><data.obcs><data.exf><STDOUT.0000>_______________________________________________
> > MITgcm-support mailing list
> > MITgcm-support at mitgcm.org
> > http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support
>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.mitgcm.org/pipermail/mitgcm-support/attachments/20210819/ab34b61a/attachment.html>