Hi Aleksander,
I'm understanding that your data have a time and spectral frequency
axes. Maybe also lat/lon position of the satellite as a function of
time? (i.e. defining a trajectory)
With only this to go on it sounds like the simple CDL below might be be
a sufficient outline. (You would want to look up the standard_name for
your frequency axis and other variables. You could optionally insert a
"coordinates" attribute on slope and offset to emphasize the trajectory
structure.)
However, I'd still encourage running this by your satellite colleagues.
It is in some sense a degenerate swath feature type. Tweaking it to
make that apparent might pay off in long term compatibility issues.
- Steve
====================
netcdf test {
dimensions:
XFREQ = 100 ;
TAX = UNLIMITED ;
variables:
double XFREQ(XFREQ) ;
XFREQ:axis = "X" ;
XFREQ:units = "spectral frequency" ;
double TAX(TAX) ;
TAX:units = "DAYS since 1901-01-15 00:00:00" ;
TAX:axis = "T" ;
TAX:time_origin = "15-JAN-1901" ;
float SLOPE(TAX, XFREQ) ;
float OFFSET(TAX, XFREQ) ;
float LAT(TAX) ;
LAT:units="latitude" ;
float LON(TAX) ;
LON:units="longitude" ;
// global attributes:
:Conventions = "CF-1.0" ;
}
- Steve
On 10/8/2010 11:39 AM, Aleksandar Jelenak wrote:
> Hi Steven, John:
>
> Thanks for mentioning the CF-satellite list. I was told about it just
> two days ago and have already joined.
>
> My data is not of the satellite swath type. It is a time series of
> calibration correction coefficients for all channels of a satellite
> instrument. One (time, chan) variable for the slope and one (time,
> chan) variable for the offset of the correction formula.
>
> I came to the Discrete Sampling Geometries convention because I wanted
> to serve my data via the NetCDF Subset web service of the THREDDS
> server. That service supports only the Grid and Discrete Sampling
> Geometry data types. I thought promoting the satellite into a station
> with fixed latitude, longitude, and altitude would do the trick but
> obviously not.
>
> Let's not get hung up on the term "satellite" here. What if it was a
> spectrometer on a pole, looking downward at a target? What CDM type is
> most appropriate for this kind of data?
>
> -Aleksandar
> _______________________________________________
> CF-metadata mailing list
> CF-metadata at cgd.ucar.edu
> http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <
http://mailman.cgd.ucar.edu/pipermail/cf-metadata/attachments/20101008/594a79e1/attachment-0002.html>
Received on Fri Oct 08 2010 - 13:26:09 BST
