NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: FW: A noon sight conundrum
From: Herbert Prinz
Date: 2003 Nov 26, 18:08 -0500
From: Herbert Prinz
Date: 2003 Nov 26, 18:08 -0500
Paul Hirose wrote: > By switching the MICA time scale to TT > and applying the actual offset between TT and UT1 on the date of > Kieran's observation, I got these values for zenith distance and > azimuth: > > 2002 Jul 20 03:16:43.4 42 30 54.3 0 00 02.6 > 2002 Jul 20 03:16:44.4 42 30 54.3 359 59 41.9 Paul and All, Unless you have a new magic version of MICA, things are not as simple as they look on the surface. There is some more processing under the hood, which Paul has been hiding from us. When one switches the time scale in MICA to TT, longitudes are referenced to the ephemeris meridian. That is a line east of Greenwich where the zero meridian would be, if the Earth would rotate at uniform speed. In order to obtain the above results, one has to adjust Kieran's longitude by the angle corresponding to the rotation during the time interval deltaT, i.e. to 132deg 23' 57" or thereabouts. In the following way it's easier to see without much calculation that inaccurate deltaT cannot have much impact on transit times: UT is, by definition, the time scale that leaves the time of sun transit invariant. Ephemeris time (TT) governs the position of the sun w.r.t. the equinox. Each time we introduce a leap second into UTC, TT gets a second ahead. But how much do the sun's celestial coordinate, i.e. the RA or SHA and its declination change in a second? And Paul concluded... > > Those times correspond to the UT1 times that I previously posted: > > 2002 Jul 20 03:15:39.0 42 30 54.3 0 00 02.3 > 2002 Jul 20 03:15:40.0 42 30 54.3 359 59 41.6 > > So it looks like meridian passage occurred at 03:15:39 UT1, which > agrees with what Kieran estimated. But the agreement is a mere fluke. First, Kieran says that he estimates meridian transit from the two "equal" altitudes at 3:07:42 and 3:23:36, given as 47deg 26.6'. In fact, these altitudes differ by 7.5" from each other when computed with higher precision. And they remain within a +/- 0.05' bandwith for a duration of ca. 8 seconds. This is enough to throw the result off by 5 seconds. One simply cannot compute the time of meridian transit from two altitudes taken within such a short time span or so close to noon. Second, your own data obtained from MICA that you posted in an earlier message shows beautifully that meridian passage preceeds culmination by 5.5 seconds, in perfect agreement with the correction formula posted in various forms by Zorbec Legras and George Huxtable. The two errors seem to cancel each other in Kieran's analysis. Herbert Prinz