NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Surveyors and celestial
From: Lu Abel
Date: 2012 Jul 6, 10:16 -0700
From: Lu Abel
Date: 2012 Jul 6, 10:16 -0700
When our discussion of the possible effect of the leap-second on surveying, I emailed a friend who is a graduate of one of the best civil engineering universities in the USA, Rensselaer Polytechnic Institute. Here is his reply to me. I find his comments on the use of the movements of Polaris for establishing an accurate North interesting.
----- Forwarded Message -----
From: Gordon Davids <g.davids@verizon.net>
To: Lu Abel <luabel@ymail.com>
Sent: Monday, July 2, 2012 1:04 PM
Subject: Re: A question for a civil engineer
Lu - I'll try to keep my comments with the questions.
On 7/2/2012 11:30 AM, Lu Abel wrote:
Yes, but the accuracy and precision were in keeping with the purpose of the survey. More accuracy is needed to establish the actual area of a surveyed plot, or the relative location of an improvement, than in knowing its precise location on the earth. In building a bridge, it's more important to get the anchor bolts at the tops of the piers in the correct relative positions rather than knowing their geodetic coordinates to the nearest 0.0001 foot.Did surveyors/civil engineers ever do celestial observations (as one would do with a sextant for establishing the position of a ship) to establish the absolute position of a point on the earth?
That, too. You run a traverse of measured distances and angles, and close it to determine the accumulated error. Even triangulation used in geodetic surveying needs at least one highly accurate base line for a start, and several more as we go to check on error and then close the grid.Or is surveying based exclusively on distances and angles from a known starting point (forgot what you call that)?
The leap second should have been accounted for in any ephemeris. However, it is only an issue for land surveyors in determining longitude. Of course, I understand that when you are at sea, you have to use the stars you can see, but unless you are launching missles, you really don't need one centimeter precision.
Reason for question is that I'm on a mailing list for people interested in celestial navigation. Don't know if you were even aware, but we had a leap-second inserted into our time-keeping on Saturday at midnight GMT. Now there's a debate raging about how that affected the Nautical Alamanc, which gives positional data for 63 celestial bodies (57 stars, 4 planets, sun, and moon) accurate to one second of time.
We used the ephemeris to give us star and sun positions, but our really precise azimuthal measurements were based on Polaris. Polaris, of course, is not fixed in relation to the earth (or vice versa :-) but describes a circle about the north pole pro tem. We get the approximate time and angle of east or west elongation, and crank the angle into our transit or theodolite so when the cross hairs track Polaris to the farthest extent of its sideways motion and then see it begin to return, that, with the angle correction, give us true north. If the theodolite is sufficiently accurate, that can be obtained to about 0.2 seconds of angle with a few repeated shots (one per night, not enough to get us really "high.") We can also obtain latitude by observing Polaris at upper or lower culmination, the same way. Neither method requires accurate time. You just have to know when to be awake and watching.
Some sources of error are weird atmospheric refraction, and variations in the gravitational field at different locations. The plum bob and the level vial are not foolproof. I was surprised to see the variations in the earth's gravitational field at the surface when I was working with the railroads and the Air Force on the Peacekeeper Rail Garrison program (1990-92). The AF was surveying pre-determined launch sites with differential GPS and highly sensitive gravimeters.
Later on (2001), in connection with a derailment and fire in the Howard Street Tunnel in Baltimore, the Corps of Engineers assisted us with a gravimeter survey of the tunnel. We were amazed at how well it recorded voids in the surrounding ground - some known and some not known at the time.
I can't see any situation where land surveyors would need one-second accuracy in a time measurement. It isn't done that way.
With respect to celestial ship navigation, the leap-second has no practical effect, since sextant observations can be off by one or two minutes of arc due to a whole host of factors ranging from the heaving of a ship's deck to indistinct horizons. But someone on the list suggests that "surveyors using a theodolite to establish position might need the accuracy"