NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: Sextant Positions versus Map Datums?
From: Michael Wescott
Date: 2002 Jan 15, 12:41 PM
From: Michael Wescott
Date: 2002 Jan 15, 12:41 PM
jared.sherman@VERIZON.NET said: > Okay, a sextant position is based on the almanac which is based > on...ergh...How exactly does that match up to a map datum? Should a > sextant position line up with a WGS84 chart position? An NAD27? A > what-ever-is-most recent datum? Or can it be expected to be off > by...What distance actually? The almanac is based on astronomical observations that are independent of any specific model of the earth as well as independent of the locale in which the measurements are made. WGS84 et al. are mathematical constructs, i.e. " sea-level surfaces" that are designed to closely fit local observations. The details are in what is meant by "local" and by "closely fit". For example a reference ellipsoid might ignore the fit of altitude and try to make latitude and longitude fit more closely to actual observations. North American ellipsoids will model the sea-level surface of North America at the expense of big errors in say Austrailia. For a more detailed and more lucid explanation see Chapter 2 of Bowditch: http://pollux.nss.nima.mil/NAV_PUBS/APN/Chapt-02.pdf > If NAD27 and WGS84 disagree by a mile, then wouldn't the "most > accurate" sextant position still either be off by a mile, or have been > off by a mile? Or maybe still be off wherever the map datum doesn't > match a "spherical earth" model? > How exactly does one accomodate the "no datum here" from a sextant > position? Aside from the fact that the difference may be meaningless > out at sea, where the maps all agree that "you are here" in the big > empty spaces and a mile or two doesn't always matter. The better fit should be the more "local" model. That said, I would think that the uncertainty of sextant observations would be the same order on magnitude or larger than the differences between any two reasonable models. Looking at chart 25641 (it's on my wall here at work) shows that it's based on NAD83, "which for charting purposes is considered identical to WGS84." The note goes on to discuss the difference between charted positions and those referenced to the Puerto Rico Datum as about 7" north-south and 1.5" east-west. Seven seconds of latitude is about 12 feet. -- Mike Wescott Wescott_Mike@EMC.COM