[MITgcm-support] open boundary condition cost

[Patrick Heimbach]

Hi Suneet,

what's happening is that either during the read of the
OB control variable (xx_obc...) or the read of the
OB cost function, you are trying to read a non-existing record
(not clear from the output provided where the model stops).
This could have three reasons,
1. not enough records in your cost file
2. not enough records in your control file (less likely)
3. mismatch in file precision for cost file (e.g.
    you provide 32bit file, but try to read 64bit file)
Have you checked and properly set open boundary values in
data.ctrl (for OB control) and data.ecco (for OB cost)?
You could try to first only do
#define ALLOW_OBCSN_CONTROL
#undef ALLOW_OBCSN_COST_CONTRIBUTION
i.e. don't use the cost function penalty term.
I think tough the code wants both (at least it used to),
and complains if you don't have the cost term, so you need to
temporarily comment the "complaint".
I (embarrassingly) forgot which routine,
but should be easy to find once you get to that point.

-Patrick



On Mar 13, 2008, at 2:10 PM, Suneet Dwivedi wrote:

> Hi Everyone,
> When I am trying to calculate the model cost corresponding to open
> boundary conditions by defining them in ECCO_CPPOPTIONS.h (define
> ALLOW_OBCSN_COST_CONTRIBUTION; define  ALLOW_OBCSN_CONTROL), I am
> getting the following errors:
> ---------------------------------------------------------------------- 
> ------------------------------------------------
> forrtl: severe (24): end-of-file during read, unit 11, file fortqtPIW1
> Image              PC                Routine            Line         
> Source
> libintlc.so.5      0000002A95D7A05E  Unknown               Unknown   
> Unknown
> forrtl: severe (24): end-of-file during read, unit 11, file fortC2Du11
> Image              PC                Routine            Line         
> Source
> libintlc.so.5      0000002A95D7A05E  Unknown               Unknown   
> Unknown
> ---------------------------------------------------------------------- 
> ------------------------------------------------
>
> If I don't make these changes in ECCO_CPPOPTIONS.h, then the model
> runs fine and open  boundary conditions  show an effect on cost of
> others variables but as soon as I define them this problem occurs. Can
> anyone of you please tell me what is the meaning of this error and how
> to tackle this problem.
> Regards,
> Suneet
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support

---
Patrick Heimbach | heimbach at mit.edu | http://www.mit.edu/~heimbach
MIT | EAPS 54-1518 | 77 Massachusetts Ave | Cambridge MA 02139 USA
FON +1-617-253-5259 | FAX +1-617-253-4464 | SKYPE patrick.heimbach