[CF-metadata] [cjw_at_ucar.edu: CF standard_name --> reference]

From: Brian Eaton <eaton>
Date: Mon, 21 Mar 2005 19:31:31 -0700

Here is a posting to the netcdfgroup mail list that I thought would be of
interest to this group.

Brian

----- Forwarded message from Chris Webster <cjw at ucar.edu> -----

Delivered-To: eaton at ucar.edu
Organization: UCAR/Unidata
Keywords: 200503220043.j2M0hkv2013571
Date: Mon, 21 Mar 2005 17:43:39 -0700
From: Chris Webster <cjw at ucar.edu>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.3) Gecko/20040913
X-Accept-Language: en-us, en
To: netcdfgroup at unidata.ucar.edu
Subject: CF standard_name --> reference
Precedence: bulk
Reply-To: Chris Webster <cjw at ucar.edu>
X-Virus-Scanned: amavisd-new at ucar.edu

We have chosen to adopt the standard_name attribute to our netCDF files.
for it's ability to allow automatic detection of certain parameters.

However it falls short for redundant instruments. For example
we typically have 3 instruments providing position, 2 giro IRU's and a
GPS, plus an IRS/GPS blended set (i.e. 4 variables with standard_name of
latitude, longitude, altitude, etc).

Which one is the best?

So we are also looking to extend this so that we can mark the best one
as the 'reference', which should be selected.

We currently provide a variable attribute called 'Categories' which
allows for tagging variables with category names (we use this to help
locate variables from our typical list ~400 variables:

        float GLAT(Time) ;
                GLAT:units = "deg_N" ;
                GLAT:long_name = "GPS Latitude" ;
                GLAT:standard_name = "latitude" ;
                GLAT:Category = "Position,Raw" ;
                GLAT:DataQuality = "Preliminary" ;
                ....
        float LAT(Time) ;
                LAT:units = "deg_N" ;
                LAT:long_name = "Inertial Latitude" ;
                LAT:standard_name = "latitude" ;
                LAT:Categories = "Position,Raw" ;
                LAT:DataQuality = "Preliminary" ;
                ....
        float LATC(Time) ;
                LATC:units = "deg_N" ;
                LATC:long_name = "GPS-Corrected Inertial Latitude" ;
                LATC:standard_name = "latitude" ;
                LATC:Categories = "Position,Derived,Reference" ;
                LATC:DataQuality = "Preliminary" ;
                ....

We're considering the method shown above, a category called "Reference".
 So before we implement, is there anyone else considering anything
similar? Comments?

--Chris

----- End forwarded message -----
Received on Mon Mar 21 2005 - 19:31:31 GMT