NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: Point Venus, May 1774
From: Alexandre Eremenko
Date: 2007 Apr 27, 14:10 -0700
From: Alexandre Eremenko
Date: 2007 Apr 27, 14:10 -0700
This is the continuation of my messages of April 24 and April 25 on the Point Venus observations of Cook's expedition on May 2-6 1774. 1. I fixed the "clock problem". The Clock was not so bad after all. Except an unexplained leap of 1 minute on May 1 (see my message on April 24) it was slowing down by 1m 22.6s per day plus-minus a fraction of one second. The astronomers of the expedition checked its going against the Sun, and they determined the rate perfectly, and this is confirmed (to 1s) by my computations. This was a Syderial Clock (showing GHA Aries, rather than Mean solar time). 2. I can prove that what they call "Quadrant error" (for the instrument used in measuring the lunar distance) in their observation journal is in fact the CORRECTION, not "error". They don't say which instrument exactly was this. To prove this, I reduce their own observation using their own almanac, and this correction (which they call error) and obtain the same longitude AS THEY COMPUTED. 3. The residual error of their lunars (after applying the correction mentioned above in part 2), was -0.5' on May 2, -0.6' on May 3 and -0.45' on May 6. These are the averages of 5-10 observations each time and sigma in each series is from 0.2' to 0.4'. These errors are the errors of OBSERVATION, have nothing to do with the almanac. These errors are SYSTEMATIC, every single shot is an overshot. Exactly the same picture as I had for years with my own observations. Averaging does not help at all in this situation. 4. The resulting error in longitude is a sum of the observation error described in section 3, and the error resulting from the almanac. On these dates these two errors ADDED. On May 2, the almanac contributed 30' and the observation error contributed 14', giving the total error in longitude of 44' in longitude. On May 3, the almanac error contributed 32' and observation error 17' to the total of approx. 50' On May 6, the almanac contributed 20' and the observation error 23' , and there was probably some smaller reduction error. Thus we see that the role of the observation errors was roughly of the same magnitude as the errors in almanac. 5. The only thing about these observations that I still don't understand is the errors in the Moon altitudes. They are very large and also systematic. 6' on May 2, 15' on May 3 (!) and 8' on May 6. For this I have no clue. These numbers do not take into account their "quadrant error" which was less than 3' for this quadrant all the time. Alex. --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to NavList@fer3.com To , send email to NavList-@fer3.com -~----------~----~----~----~------~----~------~--~---