ok major thanx again grant
so i am still not sure i have understood everything yet
but it appears there may actually be 4 distinct levels of accuracy available
progressing possibly as follows from crudest to finest
1 spherical
2 spheroidal
3 ellipsoidal
4 geoidal
of which you indicate the latter 2 may vary by several meters
& this against the full 12756 km planetary diameter
amounts to something on the order of 1 percent of 1 percent of 1 percent
but the former 3 levels probably vary by very much more than this
perhaps even approaching 1 full percent in all
considering that even at sea level the earth is 41km or about 1 third of 1
percent wider at the equator than it is erect along its axis thru the poles
if i am not multi mistaken
anyway that is just what i can suppose at this point without any confidence
while basically just wanting to check out the accuracy of my first guess in
message 2305
but fascinated much more generally as well
m
>
>Michael:
>> can you also say which of the 3 choices gps survey technology adopts
>> & whether this points to a most consensual usage
>>
>> or is it equally at home geocoordinating in all 3 versions
>
>The basis for GPS is the WGS84 ellipsoid. But my (rather elderly) GPS
>receiver provides conversion to many local mapping systems, based on
>many locally-defined best-fit ellipsoids.
>Traditionally, horizontal mapping has been done using an ellipsoid
>model (because the sums are easier to do than if you treat the Earth
>in all its lumpy reality), but vertical positioning must always be
>referred to the geoid, since we're interested in height above mean
>sea level. In the old days, this was all done with tide gauges and
>sight lines.
>So my old GPS gives horizontal data relative to the ellipsoid, which
>match the mapping nicely, but also *heights* relative to the
>ellipsoid, which can be adrift by several metres from MSL. There's no
>reason in principle why the geoid model couldn't be built into a GPS
>receiver (it's basically just a contour map of MSL relative to the
>ellipsoid) although it's intensive in memory and calculation, I would
>guess that such things do exist, but I haven't seen them.
>
>So: ellipsoid maths is sufficiently simple that there's no reason to
>use the spherical approximation. It also transfers GPS data
>accurately to the horizontal detail of a map, which is what most
>people want for navigation. But for accurate height surveying
>information you need a geoid model built into your GPS receiver
>software.
>
>Grant
>
> Yahoo! Groups Sponsor
><http://rd.yahoo.com/M=190462.1393721.2979173.2/D=egroupmail/S=1700126166:N/A=55
>1014/?http://www.debticated.com target="_top"> Your use of Yahoo!
>Groups is subject to the <http://docs.yahoo.com/info/terms/> Yahoo! Terms
>of Service.