[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: a very simple question



Ocmip modelers:

Matthew  Hecht has noticed a problem concerning the CFC output routines.
It concerns the way of filling the two-dimensional bound arrays and comes
from
a confusion between C and FORTRAN notation.

In the old output routines, it is indicated that

bounds_lon(i,j,1,1) is the lower left corner;
bounds_lon(i,j,1,2) is the lower right corner;
bounds_lon(i,j,2,1) is the upper left corner; and
bounds_lon(i,j,2,2) is the upper right corner.

AND IT SHOULD BE INDICATED AS

bounds_lon(i,j,1,1) is the lower left corner;
bounds_lon(i,j,2,1) is the lower right corner;
bounds_lon(i,j,1,2) is the upper left corner; and
bounds_lon(i,j,2,2) is the upper right corner.

So please download the new version of these routines (all currently at
revision 1.2)
1)  write_nc_CFC_year.f
2)  write_nc_MaskAreaBathy.f
3)  write_nc_SurfTS_year.f

Sorry for the inconvenience and thank you Matthew for noticing this error.

The web documentation is being changed right now.

Regards, Patrick

********************************************************
Matthew Hecht wrote:

> Jim--
>
> we were looking over the specification of output format for the cfc runs.
> There's one thing I just wanted to double check. We'll be including the
> coordinates of the corners of the tracer grid cells. If i,j index the
> coordinates of the grid cell and ii,jj index the 4 corners, it appears
> that we've been asked to send bounds_lon(i, j, jj, ii). I would have
> probably laid this out as (i, j, ii, jj), and I just wanted to verify that
> the former is what you want.
>
> We're running now, and if we don't make any mistakes we should get done on
> schedule. You can put us down as an on-line model.
>
> --Matthew