NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: Lunars with SNO-T
From: George Huxtable
Date: 2004 Oct 28, 22:27 +0100
From: George Huxtable
Date: 2004 Oct 28, 22:27 +0100
In our dialogue about rejecting observations, Alex Eremenko concluded- >If I am asked to state a general rule for rejection, the >only rule would be: >USE YOUR COMMON SENSE. And I fully agree. We are at one about that. And now that he has said that, other aspects of our argument (which as he said, seem to be based on misunderstanding) become irrelevant. His following question deserves an answer- >A question to George: >Suppose you are in Sea, have no radio, and taking lunars >to compute your chronometer correction. >And you obtained exactly that series of lunar distances I had. >How would you proceed? Would you average all 6 or reject one? I agree that I, too would reject that rogue observation, in the light of the precision that was being achieved for the others. Just as I conceded on 27 Oct, "In the light of the consistent precision Alex shows he can achieve in the second set, I don't disagree with his decision to reject that point in the first set." What I was really arguing with was his stated reason for making that rejection, but we now agree about that matter. ============= Finally, I enquired into the reason for the small difference that Frank Reed's program gave between the average of longitudes and the one longitude calculated from averaged distances / times. It turned out that the input routine Frank's program will only accept Lunar distances given to the neaest 0.1', and not any further digits. So Alex, too, was forced to quantise his average distance to the nearest tenth of a minute. As a result, the resulting longitudes will be quantised in steps of about 3'. There's nothing wrong with that; it corresponds rather well with the physical precision that might be achievable. But it means that investigating such tiny differences in the result, as I was trying to do, becomes rather meaningless. No complaints about that: the program does the job it sets out to do, and well. Which is why I wrote "no discredit to Frank's program". ============= But the questions remain, related to our previous long discussion on "averaging". Is there a non-linearity effect which will upset the result of a longitude deduced by averaging lunar distances, especially of a series that's protracted in time? Is there in that case an advantage in clearing each one, then averaging the result? If such a difference could be glimpsed in Alex's series, though trivial in amount, are there situations where it can become important? If the answer to these questions is "yes", then I think the culprit will be found in the non-linearity of the Moon parallax correction. George. ================================================================ contact George Huxtable by email at george@huxtable.u-net.com, by phone at 01865 820222 (from outside UK, +44 1865 820222), or by mail at 1 Sandy Lane, Southmoor, Abingdon, Oxon OX13 5HX, UK. ================================================================