John, et al.,
I agree with Jonathan that the metadata in files should not include
detailed model documentation, but should instead point to that
documentation. That is what can be done with the CF global "references"
attribute. Examples of the kind of information that would be usefully
included in the the actual model documentation can be found at:
http://www-pcmdi.llnl.gov/ipcc/model_documentation/ipcc_model_documentation.php
With respect to the shape of the Earth in global models, it is
invariably taken to be spherical. This approximation works well (and
allows g, the acceleration due to gravity, to be taken as a constant)
because to good accuracy "effective gravity" (i.e., Newtonian gravity +
centrifugal "force" due to the Earth's rotation) is very nearly
perpendicular to the Earth's surface. (The Earth has effectively
deformed from sphericity so that this is true.)
Larry Gates has recently examined again this approximation in J. Atmos.
Science, 61, 2478-2487, 2004. Here is part of his abstract:
With the exception of two terms in the meridional and vertical equations
of motion that are unique to the spheroidal system, all of the metric
and rotational terms in the spheroidal system correspond to those found
in the familiar spherical formulation, but now have coefficients that
are functions of both the spheroidal latitude and elevation. The unique
spheroidal terms arise from the resolution of the difference between the
directions of apparent gravity and Newtonian gravitation, which is
neglected in the spherical formulation.
Karl
Rich Signell wrote:
>
> John,
>
> There are two "sphere" ellipsoids recognized by the EPSG:
>
> Clarke 1866 Authalic Sphere (EPSG_code=7052); radius=6370997
> GRS 1980 Authalic Sphere (EPSG_code=7048); radius=6371007
>
> Both of these spheres were designed to have the volume of the sphere
> match the true ellipsoid, and the slight difference of 10 meters
> represents the progress in understanding the ellipsoid over the 114 years!
>
> It sounds like you would like to use the older one, and since it hasn't
> been specified, we assume it doesn't matter, but...
>
> I would choose the GRS 1980 value, just to "sound" more modern (so that
> people wouldn't ask me why my netcdf data was based on something that
> came from 1866)
>
> -Rich
>
> John Caron <caron at unidata.ucar.edu>
>
>> So would everyone agree, that in the absence of explicit information,
>> I could use, say, a spherical earth with radius 6370997 m as a
>> reasonable default ?
>>
>>
>> _______________________________________________
>> CF-metadata mailing list
>> CF-metadata at cgd.ucar.edu
>> http://www.cgd.ucar.edu/mailman/listinfo/cf-metadata
>
>
Received on Tue Apr 05 2005 - 10:51:10 BST
