Model blowup 2009
-
- Posts: 30
- Joined: Fri Mar 10, 2017 11:22 pm
- Location: Universidade Federal do Maranhão, Brazil
Model blowup 2009
I was performing a long integration (1985 to 2015), when in December 2009 the model stopped, the .out shows me the following error:
NETCDF_GET_DIM - too small dimension parameter, Mdims = 50*****
change file mod_netcdf.F and recompile
Found Error: ** Line: 330 Source: ROMS/Modules/mod_netcdf.F, netcdf_get_dim
NETCDF_GET_DIM - unable to inquire about contents of input NetCDF file: /scratch/leonardo.lima/Data/Forcing/Atmosphere/CFSR/CFSR-2010-blk-TSA-roms-Uwind_2.nc
call from: ROMS/Utility/inquiry.F
Found Error: 02 Line: 462 Source: ROMS/Modules/mod_netcdf.F, netcdf_check_dim
Found Error: 02 Line: 176 Source: ROMS/Utility/inquiry.F
Found Error: 02 Line: 119 Source: ROMS/Utility/get_2dfld.F
Found Error: 02 Line: 134 Source: ROMS/Nonlinear/get_data.F
Found Error: 02 Line: 205 Source: ROMS/Nonlinear/main3d.F
Found Error: 02 Line: 274 Source: ROMS/Drivers/nl_ocean.h
Is it a problem with my atmospheric forcing?
Can anyone help?
Thank you very much!
NETCDF_GET_DIM - too small dimension parameter, Mdims = 50*****
change file mod_netcdf.F and recompile
Found Error: ** Line: 330 Source: ROMS/Modules/mod_netcdf.F, netcdf_get_dim
NETCDF_GET_DIM - unable to inquire about contents of input NetCDF file: /scratch/leonardo.lima/Data/Forcing/Atmosphere/CFSR/CFSR-2010-blk-TSA-roms-Uwind_2.nc
call from: ROMS/Utility/inquiry.F
Found Error: 02 Line: 462 Source: ROMS/Modules/mod_netcdf.F, netcdf_check_dim
Found Error: 02 Line: 176 Source: ROMS/Utility/inquiry.F
Found Error: 02 Line: 119 Source: ROMS/Utility/get_2dfld.F
Found Error: 02 Line: 134 Source: ROMS/Nonlinear/get_data.F
Found Error: 02 Line: 205 Source: ROMS/Nonlinear/main3d.F
Found Error: 02 Line: 274 Source: ROMS/Drivers/nl_ocean.h
Is it a problem with my atmospheric forcing?
Can anyone help?
Thank you very much!
Re: Model blowup 2009
I believe the problem is described here. Can you do as they suggest?NETCDF_GET_DIM - too small dimension parameter, Mdims = 50*****
change file mod_netcdf.F and recompile
-
- Posts: 30
- Joined: Fri Mar 10, 2017 11:22 pm
- Location: Universidade Federal do Maranhão, Brazil
Re: Model blowup 2009
Thank you for reply, Kate
But the model run until 2009 before changing the folder where the atmospheric forcing from 2010 to 2015. Do you think I should recompile?
But the model run until 2009 before changing the folder where the atmospheric forcing from 2010 to 2015. Do you think I should recompile?
Re: Model blowup 2009
I don't know what you changed to make it run out of dimensions. I've not yet run out of dimensions or I would have changed it myself. You have at least two choices - recompile it and run with it if it works or dig into what's really going on. If the first doesn't work, then you're down to the second choice anyway.
-
- Posts: 30
- Joined: Fri Mar 10, 2017 11:22 pm
- Location: Universidade Federal do Maranhão, Brazil
Re: Model blowup 2009
Where can I have changed the dimensions? All right, Kate. I will do this. Thank you very much for your attention.
Best regards,
Best regards,
Re: Model blowup 2009
From the top of the ROMS source directory, I do "vi */*/mod_netcdf.F", then search for the string Mdims.change file mod_netcdf.F and recompile
My "vi" is aliased to "vim". I assume there's some text editor you know how to use.
-
- Posts: 30
- Joined: Fri Mar 10, 2017 11:22 pm
- Location: Universidade Federal do Maranhão, Brazil
Re: Model blowup 2009
Ok Kate, Thank you,
Do you totally rule out some mistake in the forcing? In the folders set in the .in file?
Do you totally rule out some mistake in the forcing? In the folders set in the .in file?
Re: Model blowup 2009
Not at all. As I said, I've never run out of dimensions and I use a lot of forcing files. Tools for trouble-shooting these things include (a) modifying the code as suggested and seeing what happens - shouldn't take long, after all and (b) running in a debugger and watching it run out of dimensions, seeing if something looks odd and fixing it. Option (b) requires a debugger and some idea of how to use it. I can't tell from here what's going on, honestly.adilsonborges wrote:Ok Kate, Thank you,
Do you totally rule out some mistake in the forcing? In the folders set in the .in file?
-
- Posts: 30
- Joined: Fri Mar 10, 2017 11:22 pm
- Location: Universidade Federal do Maranhão, Brazil
Re: Model blowup 2009
Ok, Kate,
Thank you very much!
Thank you very much!
-
- Posts: 30
- Joined: Fri Mar 10, 2017 11:22 pm
- Location: Universidade Federal do Maranhão, Brazil
Re: Model blowup 2009
Hi Kate and all,
I recompiled ROMS by changing the Mdims to 100 and to 1000, I did tests starting on 01 of 2010 and this runs normally. The model is no more than month 12 of 2009 for month 01 of 2010. I found that forcants do not influence. I do not know what this error is about, maybe a problem when compiling, I do not know. My grid contemplates the Equatorial Atlantic (10N 10S and 60W 20W), with all edges of the grid open. In my .h file is set:
#undef EASTERN_WALL
#undef NORTHERN_WALL
#define WESTERN_WALL
#undef SOUTHERN_WALL
#define RADIATION_2D
Is there any influence of these settings on this error?
Best regrads
I recompiled ROMS by changing the Mdims to 100 and to 1000, I did tests starting on 01 of 2010 and this runs normally. The model is no more than month 12 of 2009 for month 01 of 2010. I found that forcants do not influence. I do not know what this error is about, maybe a problem when compiling, I do not know. My grid contemplates the Equatorial Atlantic (10N 10S and 60W 20W), with all edges of the grid open. In my .h file is set:
#undef EASTERN_WALL
#undef NORTHERN_WALL
#define WESTERN_WALL
#undef SOUTHERN_WALL
#define RADIATION_2D
Is there any influence of these settings on this error?
Best regrads
Re: Model blowup 2009
I have no idea what's going on, but this:
implies that your ROMS version is really very dated. You might want to download something more modern to see if that makes a difference.#undef EASTERN_WALL
#undef NORTHERN_WALL
#define WESTERN_WALL
#undef SOUTHERN_WALL
-
- Posts: 30
- Joined: Fri Mar 10, 2017 11:22 pm
- Location: Universidade Federal do Maranhão, Brazil
Re: Model blowup 2009
Kate, thank you very much
I will continue trying to solve, who knows how to download again the roms code does not work?
Thanks for the help always!
Best regards
I will continue trying to solve, who knows how to download again the roms code does not work?
Thanks for the help always!
Best regards
Re: Model blowup 2009
What sort of problems are you having with downloading the code?
-
- Posts: 30
- Joined: Fri Mar 10, 2017 11:22 pm
- Location: Universidade Federal do Maranhão, Brazil
Re: Model blowup 2009
Sorry, Kate, no problem.
I compiled using an older code (April 2017) and it worked.
Thanks a lot for the help!
I compiled using an older code (April 2017) and it worked.
Thanks a lot for the help!