* Assuming that the answer to my question is "yes", is there a single
description for a string to use in the axis attribute that would be
sufficient to cover most of the different types of radiant energy
binning? Would it work to have "band" as the value to put in the axis
attribute, and then allow the standard_name and units attributes to
specialize the type of band, such as wavelength vs polarization? (I
assume that trying to put wavelength and polarization into the same
dimension is probably not going to be a good idea.)
I've only worked with netCDF a little bit, so I still struggle with the
terminology, but this bullet from Jim got my attention. Perhaps it's best
for me to describe how a user of passive microwave data think of their
data.
That person might be working with SSM/I, which measures microwave energy
in 2 frequencies at 2 polarizations, and 1 additional frequency at 1
polarization. For each physical location (say lat/lon points at
approximately 25-km sampling spacing), swath data from SSM/I has 5
brightness temperature values: one each for 19H, 19H, 22V, 37H, and 37V.
At the same time, SSM/I measures one more frequency, both polarizations,
at double the spatial resolution (12.5-km spacing between points), so this
would be 85H and 85V values at each location.
I think that scatterometers also have cross-polarized channels, so they
also care about a frequency and "HV", "VV" and "HH".
I think this means that users of this kinds of data would be putting both
frequency and polarization "into the same dimension" as Jim says. I'm not
sure of CF implications, I just want to provide this as an example that
has caused us metadata concerns because it's not just sensing frequency
that determines a given "band" or "channel". As long as there's enough
flexibility to represent that these are different measurements, it should
be OK.
MJ
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Mary Jo Brodzik, Special Projects Lead, 303-492-8263
NSIDC/CIRES, Univ. of Colo. at Boulder, 449 UCB, Boulder, CO 80309-0449
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
_______________________________________________
cf-satellite mailing list
cf-satellite@xxxxxxxxxxxxxxxx
For list information or to unsubscribe, visit:
http://www.unidata.ucar.edu/mailing_lists/
