[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: 961015: netcdf/udunits

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: Re: 961015: netcdf/udunits
  • Date: Wed, 16 Oct 1996 10:57:44 -0600
>To: address@hidden
>From: address@hidden (Jim Hines (awdnsun) 472-6708)
>Subject: netcdf/udunits
>Organization: .
>Keywords: 199610151305.AA28322

Jim,

> >> When I do 
> >> call ncagtc(ncid,NCGLOBAL,'_basetime',basetime,80,rcode)
> >> utdec(basetime,t_unit)
> >> uticaltime(1948,8,1,0,0,0,t_unit,ref_val)
> >> 
> >> ref_val=0.0
> >> 
> >> and when I do a
> >> uticaltime(1948,8,1,24,0,0,t_unit,ref_val)
> >> ref_val=1.0
> >
> >Those both seem correct, since the first specifies 0.0 days since
> >1948-08-01 00:00, and the second specifies 1.0 days (24 hours) since
> >1948-08-01 00:00.
> >
> >> ref_val of 0.0 does not work in ncvgt1 and 
> >> 1.0 does
> >
> >I don't understand this.  How are you calling ncvgt1 and why is ref_val
> >used as an argument?  I wouldn't think ncvgt1 would take a
> >double-precision argument such as ref_val.  It requires integer indices
> >to specify which value of a variable to get.
> >
> >> the data does start on Aug 1 1948
> >> 
> >> My question is ...  is the second way the
> >> correct way of working with time or does the
> >> CDL need to be changed (with respect to bastime)?
> >
> >The CDL looks fine, and both your examples of calling uticaltime seem to
> >return the correct value for the arguments you have given.  It sounds
> >like the problem is in the way you are calling ncvgt1.
>
> I am setting the ref_val to a integer indices that 
> is used as an agruement in ncvgt1.  This I thought
> would index me to the correct day in the file.
> Is my thinking correct??  Will I get into trouble
> doing it this way?  It looks like I would have to
> specify the end time of the day
> 
> uticaltime(1948,8,1,24,0,0,t_unit,ref_val)
> 
> to get the correct index into the file.

Without seeing your NCVGT1 call, I can only guess what is going on here.

Since you are using Fortran with array indices starting at 1 rather than
0, if your first day is 1948-08-01, then you should just add 1 to
whatever is returned as ref_val by uticaltime and use that as a day
index.  That would give you an index of 1 for any valid time during the
day 1948-08-01, 2 for 1948-08-02, etc.

If you were using C, array indices start at 0, and you could use the
integer-truncated ref_val value directly.  

Specifying the end time of the day using a time of 24:00 is another more
roundabout way of just adding 1 to the number of days returned by your
uticaltime call.

_____________________________________________________________________

Russ Rew                                         UCAR Unidata Program
address@hidden                     http://www.unidata.ucar.edu