Need help on the meaning of the error message

[fenghe@wisc_edu]

Dear CCSM Software Development Group

I am running coupled CCSM3 (T31_g3v5) with slight code modification in forcing_coupled.F from pop.

The model could run successfully for 39 years, but stopped at the 1st month of year 40.

I pasted part of poe.stderr.955414 here and the full information from poe.stderr.955414/poe.stdout.955414 is added as the attachments.

It seems that the model blows-up, but we didn't observe any "unrealistic" numbers from the output.

I just want to get a little clue on why the model blows-up. Thanks!

The code directory is

/homebv/jcheng2/ccsm3/scripts/b30.PIDVHSE

And the error message is poe.stderr.955414





24: Traceback:
24:
24: Traceback:
24: Offset 0x0000010c in procedure __abortutils_NMOD_endrun Offset 0x0000010c in procedure __abortutils_NMOD_endrun, near line 38 in file /ptmp/jcheng2/b30.PIDVHSE/atm/obj/abortutils.f90
24:, near line 38 in file /ptmp/jcheng2/b30.PIDVHSE/atm/obj/abortutils.f90
24: Offset 0x00000640 in procedure cubydr Offset 0x00000640 in procedure cubydr, near line 132 in file /ptmp/jcheng2/b30.PIDVHSE/atm/obj/cubydr.f90
24:, near line 132 in file /ptmp/jcheng2/b30.PIDVHSE/atm/obj/cubydr.f90
24: Offset 0x00000084 in procedure hrintp Offset 0x00000084 in procedure hrintp, near line 139 in file /ptmp/jcheng2/b30.PIDVHSE/atm/obj/hrintp.f90
24:, near line 139 in file /ptmp/jcheng2/b30.PIDVHSE/atm/obj/hrintp.f90
24: Offset 0x00000254 in procedure sltb1 Offset 0x00000254 in procedure sltb1, near line 183 in file , near line 183 in file /ptmp/jcheng2/b30.PIDVHSE/atm/obj/sltb1.f90
24:/ptmp/jcheng2/b30.PIDVHSE/atm/obj/sltb1.f90
24: Offset 0x00000490 in procedure scanslt@OL@1 Offset 0x00000490 in procedure scanslt@OL@1
24:
24: Location 0x0900000001031164
24: Location 0x0900000001031164
24: Location 0x090000000102a3a0
24: Location 0x090000000102a3a0
24: Offset 0x000000dc in procedure _pthread_body
24: --- End of call chain ---
24: Offset 0x000000dc in procedure _pthread_body
24: --- End of call chain ---

 

[olson]

This is the relevant error message:


> CUBYDR: Departure point out of bounds: jcen,icount,nlon*plev= 38 2495 2496


It typically means the CFL condition has been violated because the winds became too
strong for the timestep setting of DTIME in the namelist. You may need to branch the run
using a smaller time step to get a day or two beyond the blow up point. Once past the blowup
point, you may be able to branch again, using the original time step without further incident.

In some instances, the winds are becoming unrealistically strong because some pathological
problem has occurred in the simulation, in which case you may need to track down the problem.