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

[IDV #KWE-346293]: degrees_east > 180 plots off map!



Hi Rich-

> Wow, talk about a blast from the past!

Better late than never. ;-)
 
> As you say, it seems that Steve Cousins changed the longitude values
> in his test.nc file to match the values we were previously supplying
> in the NcML so that the test.nc file now works fine.
> 
> But we can happily use the NcML file to change the longitude back to
> illustrate the IDV problem.
> 
> <netcdf xmlns="http://www.unidata.ucar.edu/namespaces/netcdf/ncml-2.2";
> location="http://rocky.umeoce.maine.edu/cousins/test.nc";>
> <variable name="Depth">
> <attribute name="positive" value="up"/>
> </variable>
> <variable name="Longitude">
> <attribute name="units" value="degrees_east"/>
> <!--    <values start="-159.8819" increment="0.125"/>-->
> <values start="200.1181" increment="0.125"/>
> </variable>
> <variable name="Latitude">
> <attribute name="units" value="degrees_north"/>
> </variable>
> <attribute name="Conventions" type="String" value="CF-1.0"/>
> </netcdf>
> 
> Using this NcML reproduces the original problem of the data plotting
> off the coastline map.  See attached:

This seems to work now in the nightly build.  If you could test
it on other datasets, that would be good.

Don

> On Sun, Feb 8, 2009 at 3:25 PM, Unidata IDV Support
> <address@hidden> wrote:
> > Hi Rich-
> >
> > I've finally gotten back to this (only taken a year).  If I
> > load in the test.nc file that the peru.ncml points to, it
> > looks like the lat/lons have been changed.  Is that true?
> > If so, do you have a file that exhibits the initial behavior?
> >
> > Thanks.
> >
> > Don
> >> > On Tue, Apr 15, 2008 at 5:30 PM, Unidata IDV Support
> >> > <address@hidden> wrote:
> >> >
> >> > >  It's really that the map lines go from -180 to 180 and the data
> >> > >  goes from 0-360.  Depending on the projection, one will not
> >> > >  line up.
> >> >
> >> > Ah, okay.
> >>
> >> Let me explain that a bit more.  The map projection from the
> >> data is based on the range of the longitude values in the data.
> >> In your case, you have values > 180.  The  map lines are in
> >> -180 to 180.  So, the map projection is going to be to the
> >> right of the map lines:
> >>
> >> -180       180  200+(your area)
> >>
> >> map        data
> >>
> >> However, when you change the data to be in the -180 to 180 range,
> >> then both are within the same bounds.
> >>
> >> (not sure if that clears it up or makes it more  confusing).
> >>
> >> > >
> >> > >  > If would be nice if IDV did this so I didn't have to!
> >> > >
> >> > >  That would be nice. ;-)
> >> > >
> >> > >  > http://stellwagen.er.usgs.gov/models/share/peru.ncml
> >> > >
> >> > >  That dataset only goes to about -160.  Is that the correct
> >> > >  link?  I'm not sure I'm understanding the problem you are seeing
> >> > >  with this example.
> >> >
> >> > If you delete the line in the NcML where I overwrote the Longitude
> >> > values you will see the problem.
> >>
> >> I've put a check in there to see if the bounds of the map
> >> projection fall within 0-360 or -180 to 180.  Based on that
> >> I normalize all longitudes to fit in that range.  I've got
> >> a lot of testing to do, but this might just work.  Thanks
> >> for the suggestion.
> >>
> >> Don
> >>
> >
> >
> > Ticket Details
> > ===================
> > Ticket ID: KWE-346293
> > Department: Support IDV
> > Priority: Normal
> > Status: Open
> >
> >
> 
> 
> 
> --
> Dr. Richard P. Signell   (508) 457-2229
> USGS, 384 Woods Hole Rd.
> Woods Hole, MA 02543-1598
> 
> 


Ticket Details
===================
Ticket ID: KWE-346293
Department: Support IDV
Priority: Normal
Status: Open