On 12/30/11 9:45 AM, Al Kozakiewicz wrote:
> That's really interesting.
>
> Jim Lux already mentioned the pitfalls of GPS and barometric measurements of
> altitude. The technology of ILS doesn't require a functioning altimeter to
> land and I don't believe IFR rules allow landing without some minimum
> visibility - enough to render a functioning altimeter "optional" on approach.
> The point being that no one expects a barometric altimeter to be accurate
> enough to be the only reference for height above ground as you land.
>
> Many, if not most, USGS elevation benchmarks were set using a rod and level
> starting in the 19th century. They are surprisingly accurate considering
> that those in the middle of the country are more than a thousand miles from
> the nearest ocean.
>
USGS references to NAVD29. A note I have says:
"Regarding the actual leveling undertaken by USC&GS/NGS, the 1929
adjustment (NGVD29) entailed about 75,000 km in the U.S. and about
32,000 km in Canada. By contrast, the 1988 readjustment (NAVD88)
involved re-running much of the first-order vertical control network --
about 81,500 km -- to First Order Class II specifications."
First Order survey is typically good to 1 ppm.. so, for 1000 miles,
that's good to 5 feet.
Imagine doing this carrying all your gear on horse/mule, etc. Those
guys were tough.
http://www.ngs.noaa.gov/web/about_ngs/history/ is an interesting source.
(A great book: "The Great Arc" describes the Great Trigonometric Survey
of India.. 5 foot diameter theodolites, precision chaining, crew eaten
by tigers, etc.)
> As an aside, I thought I had read that Google Earth had a problem with
> elevations. I don't remember the exact circumstances, but the context was
> some sort of modeling as GE assumes a spherical earth. Basically, GE
> underestimates altitudes and sea levels at low latitudes due to the "bulging"
> of the earth at the equator. That sound familiar to anyone?
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|