Hi again,
Within BODC, we address your concerns by having a much richer set of feature types that include concepts like timeSeriesSpectrum to indicate that the z-axis is non-spatial and so give display clients a bit more help. However, I think that with the right parameter attribute semantic information and the CF feature types we should be OK in SeaDataNet.
My thinking so far has been confined to non-directional spectra. I can't see how directional spectra could be handled in CF without specification of an additional feature type, but (not for the first time) I might be misunderstanding something in John's view of feature types.
Cheers, Roy.
-----Original Message-----
From: Ellyn Montgomery [mailto:emontgomery at usgs.gov]
Sent: 29 August 2012 15:47
To: Lowry, Roy K.
Cc: cf-metadata at cgd.ucar.edu
Subject: Re: [CF-metadata] example of sea_surface_wave_directional_variance_spectral_density
Roy-
I'm glad that it's on your plate!
Conceptually, I am a bit uncomfortable with having frequency as Z since
many of us think of Z as depth/altitude. Having frequency as Z fits the
shapes allowed, but I wonder what confusion it would start. Are there
ways for clients to identify Z as a non-depth content? Is it as simple
as calling the Z coordinate "frequency" instead of "depth", and then
clients would be able to know what to do?
Then of course the direction component would need to be wedged into the
coordinate system somehow. Such fun!
Looking forward to the discussion!
Ellyn
On 08/29/2012 09:54 AM, Lowry, Roy K. wrote:
> Hi Ellyn,
>
> Encoding wave spectra in a CF-compliant manner is an issue that I plan to address on behalf of SeaDataNet in the next month or so. The point I'd reached in thinking this through was to use the timeSeriesProfile feature type with frequency as the z co-ordinate. I think this OK, but if not I'm sure somebody on this list will shout.
>
> I prefer single-value vectors rather to scalars because it opens the way to packaging multiple instances into a single NetCDF file.
>
> Cheers, Roy.
>
> -----Original Message-----
> From: CF-metadata [mailto:cf-metadata-bounces at cgd.ucar.edu] On Behalf Of Ellyn Montgomery
> Sent: 28 August 2012 15:04
> To: cf-metadata at cgd.ucar.edu
> Subject: [CF-metadata] example of sea_surface_wave_directional_variance_spectral_density
>
> Hello folks-
>
> We want our wave data observations to be CF compliant, with the
> dimensions and representation of time correct. I believe the data we
> collect fits somewhere under Discrete Samples, since we end up with a
> time series of wave spectra from our stationary, bottom-mounted acoustic
> instruments. I couldn't find how the file might look in the 1.6
> conventions document or appendices, so could someone please direct me to
> an example dataset that includes some wave spectra? Help with which
> featureType should be used would also be appreciated.
>
> The Standard Name table says the dimension order should be
> S(t,x,y,f,theta). Since our data is from a point observation (lat and
> lon are size (1)), can the X,Y dimensions be dropped, or are they needed
> as place-holders?
>
> Thanks very much for the help!
> Ellyn
>
--
Ellyn T. Montgomery, Oceanographer and Data Manager
U.S. Geological Survey
Woods Hole Coastal and Marine Science Center
384 Woods Hole Road, Woods Hole, MA 02543-1598
(508)457-2356
--
This message (and any attachments) is for the recipient only. NERC
is subject to the Freedom of Information Act 2000 and the contents
of this email and any reply you make may be disclosed by NERC unless
it is exempt from release under the Act. Any material supplied to
NERC may be stored in an electronic records management system.
Received on Wed Aug 29 2012 - 09:26:16 BST
