[thredds] Clarification on Unidata Dataset Discovery/THREDDS geospatial

• To: Ethan Davis <edavis@xxxxxxxxxxxxxxxx>
• Subject: [thredds] Clarification on Unidata Dataset Discovery/THREDDS geospatial
• From: Benno Blumenthal <benno@xxxxxxxxxxxxxxxx>
• Date: Thu, 4 Mar 2010 11:57:28 -0500

HI Ethan,

I am working with  NetCDF Attribute Convention for Dataset Discovery
defined at
http://www.unidata.ucar.edu/software/netcdf-java/formats/DataDiscoveryAttConvention.html

In particular, for

geospatial_lat_min, geospatial_lat_max, geospatial_lon_min,
geospatial_lon_max, geospatial_vertical_min geospatial_vertical_max

it says

Describes a simple latitude, longitude, and vertical bounding box.

I thought this was pretty clear, and in particular refers to the edges of
that simple latitude, longitude box.   Edges of the box corresponds to the
edges of the pixels within the image/dataset for lon/lat data.   But I have
seen usage which seems to give the centers of the edge pixels instead.


Meanwhile, THREDDS gives start/span for northsouth and eastwest.  Again I
would think the start gives the edge and the span the difference between the
edges, e.g geospatial_lon_max minus geospatial_lon_min, and that pixel edges
correspond to the starts.

I am also confused about what it would mean if geospatial_lon_units were not
degrees_east.  Are you just allowing degrees_west or is there something more
significantly different in mind?

Benno




-- 
Dr. M. Benno Blumenthal          benno@xxxxxxxxxxxxxxxx
International Research Institute for climate and society
The Earth Institute at Columbia University
Lamont Campus, Palisades NY 10964-8000   (845) 680-4450