The OASIS Coupler Forum

  HOME

NEMO3.6 + OASIS3-MCT problem

Up to Specific issues in real coupled models

Posted by Anonymous at June 27 2014

Salut Laure,

I have updated OASIS3-MCT_2.0 to r1025 and NEMO to r4690.

What happens now is that I get the error: oasis_coupler_setup ERROR send recv pair both Out A_TauX_oce O_OTaux1 from both models.

oasis_coupler_setup model variable info:
model,nvars =      1    21
model,idx,var,ops =      1     1  A_TauX_oce    14  prism_out
model,idx,var,ops =      1     2  A_TauY_oce    14  prism_out
[...]
model,idx,var,ops =      1    16  A_SST    15  prism_in
model,idx,var,ops =      1    17  A_Ice_temp    15  prism_in
[...]
model,nvars =      2    18
model,idx,var,ops =      2     1  O_SSTSST    20  unknown
model,idx,var,ops =      2     2  O_TepIce    20  unknown
[...]

And like this for NEMO:
oasis_coupler_setup model variable info:
model,nvars =      1    21
model,idx,var,ops =      1     1  A_TauX_oce    14  unknown
model,idx,var,ops =      1     2  A_TauY_oce    14  unknown
[...]
model,nvars =      2    18
model,idx,var,ops =      2     1  O_SSTSST    20  prism_out
model,idx,var,ops =      2     2  O_TepIce    20  prism_out
[...]

The def_var calls seem to turn without problems in both models. Any idea? Could this have to do with the update of the code?

As a sidetrack, I have a little question regarding restart files: The names of the files seem to be limited to 8 characters. Is there a special reason for this limit? It would be nice to give the files more descriptive names.

Regards, Uwe.

Posted by Anonymous at June 28 2014

Dear Uwe,

I changed the value of prism_out and prism_in in the last version 1025 of OASIS3-MCT_2.0 (prism_out=20 (instead of 14), prism_in=21 (instead of 15)). Are you sure that in your models you are not using 14 and 15 instead of prism_out and prism_in when calling the routine the routine oasis_def_var ?

Best regards, Laure

Posted by Anonymous at June 29 2014

Hi Laure,

Okay, recompiling (not only re-linking) solved the problem with the new constants.

Cheers, Uwe
Reply to this