[Esip-documentation] Geospatial Bounds

[Aleksandar Jelenak via Esip-documentation]

Hello Bob, all:

Here’s my version for the geospatial_bounds attribute:

Describes geospatial extent using geometric objects (2D or 3D) defined in
the Well-Known Text (WKT) format. geospatial_bounds points are always
space-separated latitude (decimal degrees_north), longitude (decimal
degrees_east), and, for 3D objects, altitude (meters, up) values of the
WGS-84 coordinate reference system. Specifically WGS-84 (EPSG:4326) for 2D
objects and WGS-84 (EPSG:4979) for 3D objects.  Like the
geospatial_lon/lat_min/max attributes, these are typically approximate
values. Example: "POLYGON ((40.26 -111.29, 41.26 -111.29, 41.26 -110.29,
40.26 -110.29, 40.26 -111.29))".


Below are my responses to your comments:

On 9/23/14, 12:31 PM, “B <mailto:esip-documentation at lists.esipfed.org>ob
Simons" <esip-documentation at lists.esipfed.org> wrote:
>I could be wrong, but it is my understanding that:
>* EPSG:4326 specifies lon in the range -180 to 180, but many of my
>datasets are 0 to 360 (which requires another CRS), so using EPSG:4326 is
>wrong.

That is not my understanding nor I have never seen any standards document
that prescribes a WGS-84 official longitude range.

>* In the proposed ACDD 1.3 there is currently no place to specify a CRS.
>  Notably, not in the definitions of geospatial_lat/lon_min/max and
>geospatial_bounds.

My definition of geospatial_bounds attribute clarifies that the WGS-84 CRS
is used. 

>* Are we supplying datums, too, when they are relevant?

Isn’t a datum the same thing as a CRS?

>Summary: 
>So now we're requiring all data providers, all data users, and all
>software libraries (in particular) to have expert and encyclopedic
>knowledge of CRS's, WKT, and datums (which I don't have and I suspect few
>of us have) in order to understand the geospatial_bounds
> WKT?

No. WGS-84 CRS is as plain vanilla, default, widely in use, etc. CRS as
you can get.

>The points are not tied to a specific CRS, but instead provide users with
>the approximate bounds of the data.

Approximate values and CRS are two completely different things. Even
approximate values without a known CRS are useless.

	-Aleksandar