[RegCNET] tutorial question / post-processing

[Moetasim Ashfaq]

Ok- this issue i had seen earlier. This has two possible reasons: 

1) This comes from the latest pgi compiler version. Which version you are using
? Its works with 6.05 and below and gives this error with 6.12 version. I didn't
fix it for latest version. You may try using IFC or older pgi version. I will
check this issue of new pgi version.

2) You are low in disk space. 

BTW, You may communicate personally.

Moet


Quoting Michael Notaro <mnotaro at wisc.edu>:

> Thanks.  But my netcdf data file is empty, except for a header.
> Why didn't it put the data into the nc file?
> 
> Mike
> 
> On Jun 6, 2007, at 3:25 PM, Moetasim Ashfaq wrote:
> 
> > Following error should be ignored while doing post-processing -  
> > they have no
> > effect on output.
> > ncvarid: ncid 6: Variable not found
> > ncdimid: ncid 6: Invalid dimension id or name
> > ncvarid: ncid 6: Variable not found
> > ncvarid: ncid 6: Variable not found
> >
> > Moet
> >
> >
> >
> >
> >
> > Quoting Michael Notaro <mnotaro at wisc.edu>:
> >
> >> Thanks for your email.  I initially tried 2002090100 and the post
> >> processing crashed with this error:
> >>
> >> ........
> >> Reading output:     2002083118
> >> AVERAGING DATA:    2002083118    899946.0000000000                 3
> >> Reading output:     2002090100
> >> AVERAGING DATA:    2002090100    899952.0000000000                 4
> >> WRITING CONTINUAL OUTPUT    899544.0000000000        2002081500
> >> COMPUTING AVERAGE OUT FIELDS:           31           31           31
> >>             31
> >> xhravg=    899544.0000000000
> >> ncvarid: ncid 6: Variable not found
> >> ncdimid: ncid 6: Invalid dimension id or name
> >> ncvarid: ncid 6: Variable not found
> >> ncvarid: ncid 6: Variable not found
> >> *ERROR* in putdef in grbtst
> >>
> >> What am I doing wrong?  I thought I was closely following the
> >> tutorial, so I do not
> >> understand why I get errors.  I know the output is OK since I can
> >> view it in grads.
> >>
> >> Mike
> >>
> >> On Jun 6, 2007, at 3:03 PM, Moetasim Ashfaq wrote:
> >>
> >>> You are doing monthly average but your idate2 is set to 2002083121.
> >>> For monthly
> >>> average it should be set to 2002090100.
> >>>
> >>> Moet
> >>>
> >>>
> >>> Quoting Michael Notaro <mnotaro at wisc.edu>:
> >>>
> >>>> Hello.  I have run the RegCM3 tutorial for July and August 2002  
> >>>> over
> >>>> Europe.
> >>>> I have been unable to postprocess the output.
> >>>>
> >>>> In the output dir, data files were produced:
> >>>>
> >>>> -rw-rw-r--  1 notaro notaro 88592000 Jun  5 15:40 ATM.2002070100
> >>>> -rw-rw-r--  1 notaro notaro 87883264 Jun  5 18:02 ATM.2002080100
> >>>> -rw-rw-r--  1 notaro notaro 63504000 Jun  5 15:40 RAD.2002070100
> >>>> -rw-rw-r--  1 notaro notaro 62995968 Jun  5 18:02 RAD.2002080100
> >>>> -rw-rw-r--  1 notaro notaro 11083528 Jun  5 15:40 SAV.2002080100
> >>>> -rw-rw-r--  1 notaro notaro 11083528 Jun  5 18:02 SAV.2002090100
> >>>> -rw-rw-r--  1 notaro notaro 42166656 Jun  5 15:40 SRF.2002070100
> >>>> -rw-rw-r--  1 notaro notaro 41997312 Jun  5 18:02 SRF.2002080100
> >>>>
> >>>> Now, I try to postprocess.  postproc.in says the following:
> >>>>
> >>>> 2002080100,     ! idate0 = First date in File (yymmddhh)
> >>>> 2002080100,     ! idate1 = Start date for averaging and re-writing
> >>>> 2002083121,     ! idate2 = End date for averaging and re-writing
> >>>> 2,              ! iotype: 1=I*2 NetCDF; 2=r*4 NetCDF; 3=grads;
> >>>> 4=vis5d
> >>>> .false.,            ! plv
> >>>> .false.,        !
> >>>> .false.,        ! Write out header?
> >>>> .false.,        ! Write out all RegCM data b/twn idate1 & idate2?
> >>>> .false.,        ! Average RegCM data b/twn idate1 & idate2?
> >>>> .false.,        ! Diurnali avg of RegCM data b/twn idate1 & idate2?
> >>>> .true.,         ! Continually average ATM data b/twn idate1 &  
> >>>> idate2?
> >>>> -1.,            ! No. Days for continual averaging (-1=monthly;
> >>>> 1=daily;5=5day)
> >>>> '200208',       ! postproc output filename (not including type &  
> >>>> ext)
> >>>> '../Input',     ! ICBC directory
> >>>> 'output',       ! RegCM Output directory
> >>>> 'DOMAIN.INFO',  ! Domain Info Filename (from terrain)
> >>>> 'OUT_HEAD',     ! Header File name (from RegCM)
> >>>> '2002080100',   !  1st RegCM Output File Extension
> >>>>
> >>>> Next, I run postproc.x:
> >>>>
> >>>>
> >>>>   ....
> >>>> ....
> >>>> Reading output:     2002083100
> >>>> AVERAGING DATA:    2002083100     
> >>>> 899928.0000000000                 4
> >>>> Reading output:     2002083106
> >>>> AVERAGING DATA:    2002083106     
> >>>> 899934.0000000000                 1
> >>>> Reading output:     2002083112
> >>>> AVERAGING DATA:    2002083112     
> >>>> 899940.0000000000                 2
> >>>> Reading output:     2002083118
> >>>> AVERAGING DATA:    2002083118     
> >>>> 899946.0000000000                 3
> >>>> DONE WRITING OUT DATA!!!
> >>>> DONE!!!
> >>>>
> >>>>
> >>>> It creates a netcdf file, but it is empty.
> >>>> -rw-rw-r--  1 notaro notaro 176 Jun  6 10:07 ATM200208AVG.nc
> >>>>
> >>>> [notaro at ccrcam PracticeRun]$ ncdump -h *nc
> >>>> netcdf ATM200208AVG {
> >>>> // global attributes:
> >>>>                  :domxmin = -4.888416f ;
> >>>>                  :domxmax = 32.60123f ;
> >>>>                  :domymin = 37.03973f ;
> >>>>                  :domymax = 54.29552f ;
> >>>>                  :domzmin = 1050.f ;
> >>>>                  :domzmax = 0.f ;
> >>>>
> >>>> What did I do wrong?
> >>>>
> >>>> Sometimes, if I change some parameters in postproc.in, I get this
> >>>> error:
> >>>> COMPUTING AVERAGE OUT FIELDS:            1             
> >>>> 1            1
> >>>>              1
> >>>> xhravg=    899208.0000000000
> >>>> ncvarid: ncid 6: Variable not found
> >>>> ncdimid: ncid 6: Invalid dimension id or name
> >>>> ncvarid: ncid 6: Variable not found
> >>>> ncvarid: ncid 6: Variable not found
> >>>> *ERROR* in putdef in grbtst
> >>>>
> >>>>
> >>>> Thanks,
> >>>> Mike
> >>>>
> >>>>
> >>
> >>
> 
>