Re: [thredds] NcML Aggregation

• To: Roy Mendelssohn - NOAA Federal <roy.mendelssohn@xxxxxxxx>
• Subject: Re: [thredds] NcML Aggregation
• From: "Signell, Richard" <rsignell@xxxxxxxx>
• Date: Tue, 19 Mar 2013 16:46:24 -0400

Roy,

> I should add that this arises only because the dateFormatMark  when used in 
> aggregation creates a string for time, rather than a number, and therefore 
> breaks tools looking for CF compliance.  This issue has been brought up 
> several times before.

Totally agree here.  It would be great if dateFormatMark produced a
CF-compliant time variable.   With the new grib aggregation, a
CF-compliant time is created with units of hours since the time in the
1st grib file.   In the more general case, it might be a bit more
tricky to specify appropriate units.  Maybe those could be specified
by the user (since the units parser is already built), and then these
units could be used to transform dates into the CF-compliant time
variable.

-Rich

 --
Dr. Richard P. Signell   (508) 457-2229
USGS, 384 Woods Hole Rd.
Woods Hole, MA 02543-1598

• References:
  • [thredds] NcML Aggregation
    • From: Bob Simons - NOAA Federal
  • Re: [thredds] NcML Aggregation
    • From: Roy Mendelssohn - NOAA Federal