Re: [cf-satellite] band dimension...

• To: Tom Whittaker <whittaker@xxxxxxxx>
• Subject: Re: [cf-satellite] band dimension...
• From: Russ Rew <russ@xxxxxxxxxxxxxxxx>
• Date: Tue, 22 Jun 2010 15:13:46 -0600

Tom Whittaker wrote:
> On Tue, Jun 22, 2010 at 1:42 PM, Tom Rink <rink@xxxxxxxxxxxxx> wrote:
> >
> > For single band files, will it be necessary to dimension the data array
> > like Data(band, y, x) =A0where band is length=3D1, or is there some other
> > metadata construct which would allow a band=3D1 dimension in the
> > file, but allow Data(y,x), with other band indexed variables like
> > wavelength(band)?
> 
> My option:  since the application will have to recognize the other
> metadata in order to ascertain that this file has "image data", then
> it probably could be smart enough to realize that if the "band"
> dimension is not part of the shape of the data, that it can (and
> should) default to "1".

The CF Conventions provide for this, with the concept of s "scalar
coordinate variable", defined as:

    A scalar variable that contains coordinate data. Functionally
    equivalent to either a size one coordinate variable or a size one
    auxiliary coordinate variable. 

In section 2.4, this is further explained:

    Dimensions may be of any size, including unity.  When a single value
    of some coordinate applies to all the values in a variable, the
    recommended means of attaching this information to the variable is
    by use of a dimension of size unity with a one-element coordinate
    variable. It is also acceptable to use a scalar coordinate variable
    which eliminates the need for an associated size one dimension in
    the data variable.  The advantage of using a coordinate variable is
    that all its attributes can be used to describe the single-valued
    quantity, including boundaries.  For example, a variable containing
    data for temperature at 1.5 m above the ground has a single-valued
    coordinate supplying a height of 1.5 m, and a time-mean quantity has
    a single-valued time coordinate with an associated boundary variable
    to record the start and end of the averaging period.

So you could have Data(y,x), a scalar "band" variable, and connnect it
with the Data variable through the CF "coordinates" attribute, declare
the bounds of the band, etc.

    float Data (y, x);
      Data:coordinates = "band y x";
    double band;
      band:bounds = "band_bounds";
    double band_bounds(2);
      band_bounds:units = ...;

--Russ

• References:
  • Re: [cf-satellite] band dimension...
    • From: Tom Whittaker