Re: [netcdf-java] Questions about handling dimensionless coordinates

• To: Jon Blower <j.d.blower@xxxxxxxxxxxxx>
• Subject: Re: [netcdf-java] Questions about handling dimensionless coordinates
• From: Rich Signell <rsignell@xxxxxxxx>
• Date: Wed, 4 May 2011 12:38:55 -0400

Jon,

> I’d like to add support for dimensionless vertical coordinates in ncWMS and
> I have a few questions:
>
> 1.       How common is it for the transformation between dimensionless
> coords and natural coords to be time-dependent?  Under what circumstances
> would this occur?  Does it happen when the coordinate depends on a pressure
> field that varies with time?

Very common.  Nearly all ocean models with dimensionless coordinates
have a "free surface" variable (e.g. sea surface height) that changes
with time.   So all the z-coordinates of the layers below the free
surface also change with time.  Since the ocean surface only varies by
a few meters or less (places like Bay of Fundy are a bit more!), some
users don't care if the vertical coordinate is changing in time by a
few meters, and prefer to read the z levels at the first time step,
and just keep visualizing the results on that grid.  For others,
perhaps studying processes near the sea surface, getting these z
values exactly right many be essential.

> 2.       If the VerticalTransform exists, can I assume that
> getCoordinateArray() will always return a valid array, i.e. that nj4 has
> automatically found all the auxiliary fields necessary to create the
> transform (such as pressure)?

> 3.       I can see how to get the units of the transformed vertical
> coordinate (i.e. the “natural” coords, via
> VerticalTransform.getUnitString()) but how can I get the “positive”
> direction?  In other words, if the natural coordinates are in metres, how do
> I know whether this is depth or height?

>
> 4.       What would be the most “natural” interpolation method in the
> vertical when transforming a gridded field from dimensionless coordinates
> into levels – nearest-neighbour or linear?  (Or something else?)
>

Aren't issues 2, 3 and 4 handled by this existing "sliceAtLeve"
function from Unidata?
Can't you just use that?

<http://www.unidata.ucar.edu/software/idv/docs/javadoc/ucar/unidata/data/grid/GridUtil.html#sliceAtLevel%28visad.FieldImpl,%20visad.Real,%20int,%20int%29>

Thanks,
Rich
-- 
Dr. Richard P. Signell   (508) 457-2229
USGS, 384 Woods Hole Rd.
Woods Hole, MA 02543-1598

• References:
  • [netcdf-java] Questions about handling dimensionless coordinates
    • From: Jon Blower