Re: [cf-satellite] [CF-metadata] New CoordinateType: Spectral?

• To: "Tom Whittaker" <whittaker@xxxxxxxx>
• Subject: Re: [cf-satellite] [CF-metadata] New CoordinateType: Spectral?
• From: "rhorne@xxxxxxxxxxxxxxxxx" <rhorne@xxxxxxxxxxxxxxxxx>
• Date: Wed, 1 May 2013 15:49:38 -0400

Hi Tom: 

  

Previous posts on this thread have made it clear to me that using 
wavelength as a coordinate variable is fine. 

  

Also wanted to point out that there is currently discussion on wavelength 
and wavenumber standard names being proposed by Aleksander Jelenek whose 
associated variables could serve as coordinate variables. 

  

The project team I am working with and I are evaluating these standard 
names and plan on posting our thoughts very soon.  Note there is an 
existing standard_name "radiation_wavelength" that may very well be good 
enough.  We do plan on using wavelength as a coordinate variable.  Our data 
sets will only have one image, so this coordinate variable will be a 
scalar. 

  

The subject line for this email chain is: 

  [CF-metadata] New Standard Names for Satellite Data 

  

Here is a good place to start ... 

  

http://mailman.cgd.ucar.edu/pipermail/cf-metadata/2013/056417.html 

  

  

very respectfully, 

  

randy

----------------------------------------
 From: "Tom Whittaker" <whittaker@xxxxxxxx>
Sent: Wednesday, May 01, 2013 2:55 PM
To: "Randy Horne" <rhorne@xxxxxxxxxxxxxxxxx>
Cc: "Jonathan Gregory" <j.m.gregory@xxxxxxxxxxxxx>, 
cf-metadata@xxxxxxxxxxxx, "cf-satellite@xxxxxxxxxxxxxxxx" 
<cf-satellite@xxxxxxxxxxxxxxxx>
Subject: Re: [cf-satellite] [CF-metadata] New CoordinateType: Spectral?

I wanted to point out that last year, we discussed whether the
"wavelength" or "wavenumber" should be a "coordinate variable".

I believe that the consensus at that time was that just a "dimension
variable" should be used, and then to define variables with declared
standard_name attributes (like "wavelength" and/or "wavenumber" --
although I know these are not the agreed upon names...), which were
shaped by that "dimension variable", as a signal applications that the
"dimension variable" really was an index and related to either
wavelength or wavenumber....and more, important, that these data were
"satellite data" (as opposed to a "grid").

In this construct, it was not necessary to treat the "dimension
variable" as a "coordinate variable".

tom

--
Tom Whittaker
University of Wisconsin-Madison
Space Science & Engineering Center (SSEC)
Cooperative Institute for Meteorological Satellite Studies (CIMSS)
1225 W. Dayton Street
Madison, WI 53706 USA
ph: +1 608 262 2759