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

[Support #RNL-400249]: Problems with the day/time display for netCDF imagery in IDV and V.

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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: [Support #RNL-400249]: Problems with the day/time display for netCDF imagery in IDV and V.
  • Date: Sat, 14 Feb 2009 06:17:05 -0700
Hi Jay-

Please send these questions to support-idv instead of me.  That
way you'll be sure to get a timely response by someone here in
case I'm out of the office.

> Becky and I made some MSG eumetcast images into netCDF files with our
> new CF=YES keyword in IMGCOPY (which makes CF compliant netCDF files).

Well, since there isn't a standard for images yet, I guess they are
CF compliant grids. ;-)

> When we display them in -V and the IDV the time and date are wrong.  The
> day/times seem to be in the netCDF file header correctly.  Can you check
> these files and let us know what you think?
> 
> ftp://ftp.ssec.wisc.edu/pub/mug/eumetcast/
> 
> File:MSG-12UM0003.nc - 600x600 centered at lat/lon 0,0 (4 MB) CF
> compliant (CF=YES)
> File:MSG-12UM0004.nc - 600x600 centered at lat/lon 0.0 (4 MB) old format
> (CF=NO)

There is no value given for the time variable.  ncdump shows it
as missing, so it's probably a NaN in the IDV which would be a large
negative number.  


 time = _ ;

Since your units are seconds since the epoch, that will give you
something back in early 20th century.

A couple more notes.  Your time variable is listed as:

        int time(time) ;
                time:long_name = "seconds since 1970-1-1 0:0:0" ;
                time:units = "seconds since 1970-1-1 0:0:0" ;

The long name should probably be "nominal image time" and
the units should ideally have the leading zeros:

seconds since 1970-01-01 00:00:00

It works as is, but is much more exact as I have it.

Also, if you want them to be CF compliant, you'll need to have
the global attribute:

                :Conventions = "CF-1.0" ;

It would be helpful to have some other global attributes like:

                :nominal_image_time = "1998-06-30 00:15:00Z" ;
                :history = "netCDF encoded on 2008-09-13 22:21:05Z" ;

to help with the readbility.  It might also be a good place to
put the audit trail.


> To view them in -V or the IDV I use the General - Files/Directories
> chooser and use "I'm Feeling Lucky" to make a Color-Shaded Plan View.
> The actual time on the image is 0Z on day 2009042.  -V and the IDV lists
> the date as 1901-12-13 at 20:45:53Z.

That would be the way to load them in, but the time needs to be
fixed first.


Ticket Details
===================
Ticket ID: RNL-400249
Department: Support IDV
Priority: Normal
Status: Open