Re: [netcdfgroup] data output error

• To: netcdfgroup@xxxxxxxxxxxxxxxx
• Subject: Re: [netcdfgroup] data output error
• From: "H.Dang" <danghy@xxxxxxxxx>
• Date: Fri, 31 Jul 2009 17:56:12 -0400

Hi Dave,

Thanks for sharing this with me! I'll check the data types.

Yeah the new subroutines I made to create netCDF file are fortran 90. But
most of the original model codes (CMAQ) are fortran 77. I just found if I
call nc_chk1_putvar in an outside subroutine, I can get the correct
magnitude.. It's a bit tricky for me... Anyway, I'll solve it.

Have a good weekend!
Hongyan
On Fri, Jul 31, 2009 at 2:56 PM, Dave Allured <dave.allured@xxxxxxxx> wrote:

> Hongyan,
>
> Those wrong data values are symptomatic of a data type error in a
> subroutine call.  For example, passing an integer array in a subroutine
> argument that is supposed to be real or double precision.  Please check the
> subroutine and function calls carefully for type mismatches.
>
> I see that you are using the Netcdf fortran 90 interface.  If you convert
> your own programs to fortran 90, and use module interfaces, then the
> compiler will find these mistakes for you.  I strongly recommend that you
> start using fortran 90 instead of fortran 77. Either way, good luck.
>
> Dave Allured
> CU/CIRES Climate Diagnostics Center (CDC)
> http://cires.colorado.edu/science/centers/cdc/
> NOAA/ESRL/PSD, Climate Analysis Branch (CAB)
> http://www.cdc.noaa.gov/psd1/
>
> H.Dang wrote:
>
>> Hello~
>>
>> The data I got in the netCDF file seems wrong. My code is:
>>
>>      nstep_nreps = nstep_nreps + 1
>>      DO S = 1, N_GC_SPC
>>       call nc_chk1_putvar(
>>     &              jdate, jtime
>>     &            , conc_varid(s)
>>     &            , CGRID, S
>>     &            , nstep_nreps )
>>      END DO      write(56,*) 'cgrid(1,1,1,1)',cgrid(1,1,1,1)
>>
>> I checked the first of the data values, in the 56th file, it's corrct:
>> "cgrid(1,1,1,1)=1.6699868E-04".
>> But in the netCDF file, it is:  2.802597e-45. Actually all values in
>> netCDF has magnitude e-19~e-45.
>> The subroutine "nc_chk1_putvar" is like this:
>>
>>  count = (/NCOLS, NROWS, NLAYS, 1, 1/)
>>  start = (/ 1, 1, 1, Nspecies, nstep_nreps/)      call
>> nc_check(nf90_put_var( &
>>                    chk_ncid,  &
>>                    nc_data_varid,  &
>>                    CGRID(:,:,:,Nspecies), &
>>                    start=start,  &
>>                    count=count  &
>>                   ))
>>
>> Could you please find apparent mistake in my code? Thank you!
>>
>> --
>> Cordially,
>> Hongyan
>>
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> netcdfgroup mailing list
>> netcdfgroup@xxxxxxxxxxxxxxxx
>> For list information or to unsubscribe,  visit:
>> http://www.unidata.ucar.edu/mailing_lists/
>>
>
> _______________________________________________
> netcdfgroup mailing list
> netcdfgroup@xxxxxxxxxxxxxxxx
> For list information or to unsubscribe,  visit:
> http://www.unidata.ucar.edu/mailing_lists/
>



-- 
Cordially,
Hongyan

• References:
  • [netcdfgroup] data output error
    • From: H.Dang
  • Re: [netcdfgroup] data output error
    • From: Dave Allured