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

Re: COARDS time convention

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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: COARDS time convention
  • Date: Fri, 02 Nov 2001 08:26:02 -0700
>To: address@hidden
>From: David Legler <address@hidden>
>Subject: Re: 20011101: COARDS time convention
>Organization: US CLIVAR Office
>Keywords: COARDS, time conventions

Hi David,

> I am currently co-chairing the WOCE Data Products Committee (DPC). 
> The WOCE data system (<http://oceanic.cms.udel.edu/woce/ >)is 
> coordinating a final issue of WOCE data in netCDF format. I have a 
> small group meeting today and tomorrow (FRI) in Delaware to hash out 
> the details. We have agreed that we wish our files to be COARDS 
> compliant to the extent possible (but recognizing the convention set 
> we develop will eventually be a superset of COARDS as well as a bit 
> of EPIC). We have been confused about COARDS-compliant time 
> conventions. Currently we have time tags in the form of yyyymmdd and 
> hhmmss.dd. We believe that this is not strictly COARDS compliant. 
> Determining the explicit COARDS time convention is not easy to figure 
> out (even though we spent the better part of a few hours trying to).
>
> Our question, must we change our current convention to something like 
> "seconds since 1992-10-8 15:15:42.5" or "minutes since ...." to be 
> COARDS compliant?

Yes, COARDS requires times in forms such as that, though it's more
common to use something like

 "days since 1990-1-1 0:0:0"

But you could redundantly store times in a more human-readable form as
well in a separate variable.

By the way, there may be a clearer description of time conventions in
the new "NetCDF Climate and Forecast (CF) Metadata Conventions"
draft document 

  http://www.cgd.ucar.edu/cms/eaton/netcdf/CF-current.htm

referenced from the netCDF conventions page

  http://www.unidata.ucar.edu/packages/netcdf/conventions.html

It explicitly warns about problems with using the "years" or "months"
units in a time convention and discusses some of the issues.  That
document is also an extension of the COARDS conventions that is
supported by the original COARDS authors and other organizations.

I've CC:ed Steve Emmerson and John Caron on this reply, in case they
have anything to add ...

--Russ