[cam-users] (no subject)

[Norm Wood]

On 30 Sep., eastwind2000 wrote:
> Dear ccm3.x users:
>      When I use CCM,I meet with sevral questions :
> (1)I have set up ccm3.2 on my linux PC(Redhat 8.0+PGI Workstation), but I can not read the result of model, can any one give me suggestions to get the result of model? If the result is on the sigma levels, how can I change them to pressure cordinates?

For doing quick visualization of the results in the netcdf
output files, I like using VCDAT ("Visual Climate Data
Analysis Tools"), which you can find at:

http://esg.llnl.gov/cdat/

> (2)when I set up ccm3.6.6 on my linux PC, I have got the object files in the 'run' directory , but the information in compile_log tell me that when linking with netcdf ,it can not get the definitions of netcdf functions in model source files. I have download the netcdf tar package and set it up on my PC, and 'make test' successfully. I do guess it maybe because the compatible problems between PGI compiler and netcdf, but I do not know how to resolve it .
> 

A couple of suggestions.  First check to see if your
netcdf library was compiled with support for fortran:

$ nm /usr/local/lib/libnetcdf.a | grep " nf.*" | less

See the man page for nm for a description of what it does.
You should see a list of function names beginning with "nf_".
If not, you'll need to recompile the netcdf library - see the
documentation re how to include the Fortran programming 
interface.

Second, if your netcdf library _does_ include the fortran
functions, check to see if their names have double underscores
at the end.  Chances are, if the library was compiled with pgf77/pgf90,
they will have only single underscores.  If that's the case,
try removing the -Msecond_underscore option from the FFLAGS
variable in the makefile.

Norm
-- 
Norm Wood

     Department of Atmospheric Science
     Colorado State University
     norm@atmos.colostate.edu
     phone:  970-491-8587