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

[IDV #RCR-783387]: Odd grid data generation behavior...

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: [IDV #RCR-783387]: Odd grid data generation behavior...
  • Date: Fri, 30 Jan 2009 09:27:57 -0700
Hi Tom-

> How are linear model grids represented in NetCDF,
> array of x map coordinates, and separate array of y map coordinates?
> Or is there some compact metadata form that accounts for the
> linearity?

Separate arrays of evenly spaced coordinates.

The problem with the projections is that although the grid is
nrows x ncols, the coordinates for the projection are the
km along the projection grid.  In the case you sent me, it looks
like you are getting hit by a rounding error.

Don

> Tom
> Unidata IDV Support wrote:
> > Hi Tom-
> >
> >
> >>>> Sometimes the domain of the adapted FlatField is a Linear2DSet,
> >>>> sometimes it's a Gridded2DSet.  The data has a projection, in this
> >>>> VerticalPerspective, so I'm at a loss to understand why a GriddedSet
> >>>> is needed *sometimes*.  In fact, a GriddedSet is only used when
> >>>> a stride is applied, and even then, not all of the time.  This is a
> >>>> potential performance hit for NetCDF files that have large dimensioned
> >>>> grids.
> >>>>
> >>>>
> >>> That's all handled down in GeoGridAdapter.  Maybe the striding is
> >>> screwing up the determination of linearity?  If you run the
> >>> IDV with -debug, you'll get some printouts from GeoGridAdapter
> >>> on what it has determined is needed.  If you can give some examples,
> >>> I can look further.
> >>>
> >
> >
> >>> A linear set is used if it can determine that
> >>> the coordinate axes are linear and the Z dimension is > 1 (if there
> >>> is a Z coordinate).  Single level grids (3D on 2D manifold) have to
> >>> be GriddedSets or we get weird displays.
> >>>
> >>>
> >> I understand that, but this file only contains (x,y)->Q, where x,y->lon,lat
> >> according to proper NetCDF conventions.  I'm sure this is a memory
> >> hit, but maybe it's more of a 'back-burner' thing for now.
> >>
> >
> > The axes are not linear which is why a Gridded2DSet is created.
> >
> > For example, the first 5 numbers in the west-east coordinate are:
> >
> > -4365.84, -4363.806, -4361.771, -4359.736, -4357.702
> >
> > with intervals 2.034, 2.035, 2.035, 2.034  between them.
> >
> > This fails the checkLinearity check.
> >
> > Don
> >
> >
> > Ticket Details
> > ===================
> > Ticket ID: RCR-783387
> > Department: Support IDV
> > Priority: Normal
> > Status: Open
> >
> >
> 
> 


Ticket Details
===================
Ticket ID: RCR-783387
Department: Support IDV
Priority: Normal
Status: Open