get_ngfld.F

[nrecord]

Hi.

I'm trying to run a 1-d ROMS model coupled with a biology model (CoSiNE). I get this error:

NETCDF_OPEN - unable to open existing NetCDF file:
/dev/null
call from: get_ngfld.F

I'm not sure what netcdf file it's looking for, or what non-gridded fields do in general. Any suggestions? Is there anyone out there who runs ROMS with CoSiNE successfully?

Thanks.
-Nick

[kate]

That's with my branch? I haven't actually run with CoSiNE yet, not having the support files. My guess is you're missing the CoSiNE variables in the varinfo.dat file and it's failing on reading boundary conditions (non-gridded). What boundary conditions are you using? Do you have the fields in a file?

[nrecord]

Thanks Kate. I think you're right. I'll double check my variables.

[nrecord]

Hi. This is my follow-up on a post I sent awhile back.

I'm trying to get the UMaine (CoSiNE) ecosystem model running, by modifying the 1D BioToy example. I'm pretty new to ROMS, and I don't understand the way everything works.

When I try running the UMaine model, it seems like it's looking for the BRYNAME file. But in the BioToy example, the BRYNAME file is set to /dev/null in the .in file. What do I need to change so that my UMaine run isn't looking for all these boundary fields?

Thanks.
-Nick

[kate]

It looks like bio_toy.h has:

Code: Select all

#define EW_PERIODIC
#define NS_PERIODIC

With this, there's no reason for it to be looking for any boundary condition information. If it is, there's something wrong. globaldefs.h has checks which turn on the search for a boundary condition file.