NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
1) you have 27K stations, and 11K obs, so its hard to find a station with any
obs. try "030750-99999". Use NCDump data from the Viewer tab to see data
contents.
2) you need to add the altitude coordinate into
:coordinates = "time lon lat";
3) the time selection widget in ToolsUI sucks
4) the CF dictators are demanding I hand over the Sudetenland, i mean make some
changes to the CF point spec, so its likely some things will have to change.
> Guys,
> I made your changes to my ragged_parent_index point data file, and the
> stations now appear in ToolsUI, but when I select a station I get
> "there are no observations for this selection". I tried to change
> the time selection, but it would not let me select a time range older
> than 1 day from the present date. Is this expected behavior?
>
> My modified file is at:
> http://coast-enviro.er.usgs.gov/models/share/gsod_point_ragged.nc
>
> Thanks,
> Rich
>
>
>
> On Fri, Feb 5, 2010 at 4:55 PM, Unidata netCDF Java Support
> <address@hidden> wrote:
> > Hi Rich,
> >
> > Rich wrote:
> >> I can't get this CDM Point stationTimeSeries with ragged_parent_index to
> >> work
> >>
> >> http://coast-enviro.er.usgs.gov/models/share/gsod_point.nc
> >>
> >> The coordinate systems seem okay, I can read the coordinate values and
> >> data values using other tools, but trying to load as Point FeatureType
> >> in ToolsUI produces this error: "Table Structure(record):
> >> lat/lon/time coord not found".
> >>
> >> How can I go about diagnosing the problem?
> >
> > It took me awhile and some stepping through the code to see this one. Two
> > problems, both very minor:
> >
> > 1) This may be a hold over from an old version of the PointObs spec. Your
> > dataset has the global attribute
> >
> > CF:pointFeature = "stationTimeSeries";
> >
> > but the netCDF-java code is looking for
> >
> > CF:featureType = "stationTimeSeries";
> >
> > 2) Your "stn_name" variables needs the following attribute:
> >
> > standard_name="station_id";
> >
> > The first one is the one that took some time. I don't know the PointObs
> > code very well but also the error message didn't give much info to go on.
> > I've added to that error message some info on what feature type it thought
> > it was looking for ("POINT", the default for this part of ToolsUI). That
> > might help next time.
> >
> > Anyway, here's the NcML I used to get it working (it includes a change from
> > "deg_c" to "degC" on a few variables for udunits compatibility):
> >
> > <?xml version="1.0" encoding="UTF-8"?>
> > <netcdf xmlns="http://www.unidata.ucar.edu/namespaces/netcdf/ncml-2.2";
> > location="gsod_point.nc">
> >
> > <attribute name="CF:featureType" value="stationTimeSeries" />
> > <variable name="stn_name">
> > <attribute name="standard_name" value="station_id" />
> > </variable>
> >
> > <variable name="temp">
> > <attribute name="units" value="degC" />
> > </variable>
> > <variable name="max_temp">
> > <attribute name="units" value="degC" />
> > </variable>
> > <variable name="min_temp">
> > <attribute name="units" value="degC" />
> > </variable>
> > <variable name="dewp">
> > <attribute name="units" value="degC" />
> > </variable>
> > </netcdf>
> >
> >
> > Cheers,
> >
> > Ethan
> >
> >
> >
> > Ticket Details
> > ===================
> > Ticket ID: HOJ-653870
> > Department: Support netCDF Java
> > Priority: High
> > Status: Closed
> >
> >
>
>
>
> --
> Dr. Richard P. Signell (508) 457-2229
> USGS, 384 Woods Hole Rd.
> Woods Hole, MA 02543-1598
>
>
Ticket Details
===================
Ticket ID: HOJ-653870
Department: Support netCDF Java
Priority: High
Status: Open