[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Question about CDM



Hi Pauline:

I think Andrew's the expert on this, sounds like WCS 1.1 does not yet support 
irregular x,y coordinates. WCS 1.1.0 
(http://www.opengeospatial.org/standards/wcs) is final, but there are ongoing 
efforts to improve the next version (1.2 i suppose).

The GALEON document was written before the spec was final, so it is not 
authoritative.

Pauline Mak wrote:
Hi John & Andrew :)

Thanks for the insight into WCS - I must confess I'm not very
geographically-minded and find the standard document somewhat daunting!

I'm still a little confused though... So is WCS 1.1 still in
development?  And that, once it's publish, it will include irregular
grid?  I've also found an interesting document here:
http://www.unidata.ucar.edu/projects/THREDDS/GALEON/WCSnetCDFencoding.htm and 
it seems to imply that WCS can support irregular grid.  I could have 
misunderstood what it meant about row 6...  Some clarification would be much 
appreciated :)

Cheers,

-Pauline.

On Tue, 2007-04-24 at 18:10 +0100, Woolf, A (Andrew) wrote:
WCS 1.1 does not yet support other than regular grids, though note there
is now a change proposal in (OGC document 06-160) to get irregular grids
(ISO 19123 CV_ReferenceableGrid in ISO-speak) into GML which should
help.

In my view it is crucial that conceptual models for grids remain unified
between GML, WCS, and ISO 19123.



-----Original Message-----
From: address@hidden [mailto:address@hidden On Behalf Of John Caron
Sent: 24 April 2007 16:45
To: Pauline Mak
Cc: address@hidden
Subject: Re: Question about CDM




Pauline Mak wrote:
Hi all,

Thanks for the help - through the second validator, I think
I've found
out what's wrong with the dataset. Some of the axis
doesn't have any
values under the regular column. I've written a little
Java app that
goes through a directory and find all nc files and checks
that a) it
opens up as a GriddedDataset and then 2) the
GridCoordSystem from the
GriddedDataset isRegularSpatial(). Is this what it means
by the WCS
standard that it has to be gridded and regularly spaced?
yes. I think WCS 1.1 relaxes the regular spacing need, which will implement later this year (probably).


Thanks,

-Pauline.

On 4/21/07, *Ethan Davis* <address@hidden <mailto:address@hidden>> wrote:

    Hi Pauline,

Another resource for validating datasets against the CDM is available at

    http://motherlode.ucar.edu:8080/thredds/cdmValidate.html
    <http://motherlode.ucar.edu:8080/thredds/cdmValidate.html>

And I believe it is distributed with all TDS
installations, so you could
try it locally at
http://yourserver:port/thredds/cdmValidate.html
Which datasets can be served by the TDS depends on the
service over
which you want to make them available. The HTTP bulk
file download
service (/thredds/fileServer/) can serve any file. The
OPeNDAP service
(/thredds/dodsC/) can serve any dataset that the
netCDF-java library
can
read (the file types that can be read are listed on the
home page,
http://www.unidata.ucar.edu/software/netcdf-java/). The
WCS service
currently only handles gridded data. The key here is
that they have to
be recognized as gridded datasets by the netCDF-java
library. The
library can recognize gridded data and the coordinates
in the data using
a few different conventions. As Tom and/or Roy
mentioned, CF is the
convention we recommend. The CDM validator above should
give you more
    clues on how well you dataset validates.

    Hope that helps.

    Ethan

Pauline Mak wrote:
> Hi all,
>
> I'm currently setting up a THREDDS server to server our
datasets. The
> setup was fairly easy! However, I'm running into
problems with
NetCDF
> files which doesn't conform to the "Common Data
(Access) Model" - we
> want to serve all of our data through WCS (which
comes for free with
> THREDDS data server!). I've searched high and low
for documentation
> on how to 'correct' incompatible files. but failed
to come up with
> anything. Am I missing something? Should all NetCDF
files "just
work"
> with the NetCDF-Java 2.2 library? I'm aware that
files should be
> gridded and regular. Am I understanding correctly
that gridded means
> the files should have an x and y axis with optional
z and t? Is
> there a nice method to find out if a file is
compatible and if it's
> not compatible, why it is so?
>
> My understanding of NetCDF file is rather shallow...
so please bear
     > with me if I'm asking a simple question...
     >
     > Thanks & cheers,
     >
     > -Pauline.
     >
     > --
     > Pauline Mak
     > Software Developer, Insight4
     > Suite 2 / Level 1, 114 Murray St.
     > Hobart, Australia 7000
     > p:  (03) 6234 5203
     > f:  (03) 6234 5603
     > address@hidden <mailto:address@hidden>
    <mailto:address@hidden <mailto:address@hidden>>

--
Ethan R. Davis
Telephone: (303) 497-8155
Software Engineer Fax:
(303) 497-8690
UCAR Unidata Program Center E-mail: address@hidden <mailto:address@hidden>
P.O. Box 3000
Boulder, CO 80307-3000 http://www.unidata.ucar.edu/


----------------------------------------------------------------------
-----





--
--
Pauline Mak
Software Developer, Insight4
Suite 2 / Level 1, 114 Murray St.
Hobart, Australia 7000
p:  (03) 6234 5203
f:  (03) 6234 5603
address@hidden <mailto:address@hidden>
==============================================================
=================
To unsubscribe thredds, visit:
http://www.unidata.ucar.edu/mailing-list-delete-form.html
==============================================================
=================


===============================================================================
To unsubscribe thredds, visit:
http://www.unidata.ucar.edu/mailing-list-delete-form.html
===============================================================================


=============================================================================== To unsubscribe thredds, visit: http://www.unidata.ucar.edu/mailing-list-delete-form.html ===============================================================================


NOTE: All email exchanges with Unidata User Support are recorded in the Unidata inquiry tracking system and then made publicly available through the web. If you do not want to have your interactions made available in this way, you must let us know in each email you send to us.