i agree that satellite data needs to be a seperate convention.
you can join the conversation by subscribing to cf-satellite at
http://www.unidata.ucar.edu/support/help/
On 10/7/2010 3:08 PM, Steve Hankin wrote:
Hi Aleksandar,
Satellite datasets have a range of needs that the
PointObservationConventions (a.k.a. Discrete Sampling Geometries")
don't try to address. It might be worth your while consulting with
some of the individuals who are working on applications of CF to
satellite obs. I've cc'ed Ken Casey of the GHRSST project and Ken
Roberts who is leading CF satellite discussions at NCDC.
- Steve
=============================================
On 10/6/2010 7:24 PM, Aleksandar Jelenak wrote:
Dear Jonathan,
I am seeking a clarification about the proposed addition to the CF
convention for discrete sampling geometries data:
https://cf-pcmdi.llnl.gov/trac/wiki/PointObservationConventions
My data would fit the Single timeSeries type except the variables have
more than one dimension: var(obs, dim2, dim3, ..., dimN). Does such
variables still qualify as discrete sampling geometry data?
Not in the proposals current being considered for this convention.
What are the
additional dimensions?
In my case there is an additional dimension: channels of a satellite
instrument. I wanted to generalize this case, hence all the way to dimN.
I think supporting more dimensions would be possible if the order of
outer dimensions is maintained as described by the current proposal.
-Aleksandar
_______________________________________________
CF-metadata mailing list
[email protected]
http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
_______________________________________________
CF-metadata mailing list
[email protected]
http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
_______________________________________________
CF-metadata mailing list
[email protected]
http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
