# NavList:

## A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding

**Re: It Works.**

**From:**Arthur Pearson

**Date:**2002 Apr 7, 12:29 -0400

Gentlemen, I have been puzzling on this for the week since Bruce made his initial posting. Like George, I am still struggling. I am convinced of the utility of a method for deriving GMT using calculated altitudes. My struggle is to understand the method of minimizing the errors resulting from uncertain time AND position that. Here is where I stand. I am intrigued by Bruce's description of using "SHA meridian", but I don't quite understand it. The key phrase from his posting is: "I'd taken a time sight of the sun the afternoon before the lunar observation, so knew the error of the watch on local apparent time. By subtracting local apparent time SINCE NOON (converted to arc) from SHA sun I got SHA meridian. That nailed my east-west position in the celestial sphere at the moment of observation." I just can't master what is going on here, I probably need a diagram. Here is what I think I understand about the general procedure. Like George, I welcome corrections and challenges. 1. During the day, one establishes latitude by a noon sight. 2. The same sight that is used to determine latitude is used to establish the Watch Time of Local Apparent Noon. 3. Using DR Longitude, WT of LAN, and the equation of time we can come up with an Estimated GMT. 4. KEY FACTORS IN REDUCING ERROR OF SUBSEQUENT LUNAR: I now have eliminated error in Latitude. In addition, my error in Est. GMT and DR Longitude are now CONSISTENT WITH EACH OTHER WITH RESPECT TO THE CELESTIAL SPHERE. Somehow this seems important though I can't put my finger on exactly why. 5. We now need to keep a good DR and trust the rate of our watch up to the moment that we take our lunar distances and need to use position and est. GMT to come up with calculated altitudes. 6. Because the error in position is minimized (Latitude is correct) and the error in est. GMT and DR Longitude are consistent, the need for iterative solutions of GMT by observation is minimized. In Bruce's examples (under extreme conditions) two iterations were adequate. Again, I suspect if I can diagram how Bruce "nailed my east-west position on the celestial sphere at the moment of observation", perhaps the scales will fall from my eyes. Perhaps a more detailed description of the sequence of his almanac look-ups and calculations would also make things more clear. This is a particularly interesting discussion and I hope others will weigh in as we find our way to a common understanding. Regards, Arthur -----Original Message----- From: Navigation Mailing List [mailto:NAVIGATION-L@LISTSERV.WEBKAHUNA.COM] On Behalf Of George Huxtable Sent: Sunday, April 07, 2002 5:22 PM To: NAVIGATION-L@LISTSERV.WEBKAHUNA.COM Subject: Re: It Works. From George Huxtable- I've been "off-watch" for a few days, as far as Nav-L is concerned, getting my boat "Christina" into a seaworthy state after her winter lay-up. Even poked her head out into Poole Bay, but the weather in the English Channel being pretty thick, turned tail again straight back into harbour. Now back home to find a helpful contribution from Bill Noyce, about what a navigator has to do if he needs to work a lunar using altitudes that have been calculated rather than measured. Here's what he said- ==================== >George Huxtable says > >> 1. For a lunar that uses altitudes that are calculated rather than >> measured, it is necessary to make a measurement for local apparent time. >> This should be done within a day or so of the moment of taking the lunar, >> depending on how well the timepiece can be trusted over such a period. >> : >> 4. The local apparent time, so derived, and corrected by the equation of >> time to give local mean time, can be checked against the ship's timepiece, >> and can now be used with the Almanac to establish precise altitudes for >> both the bodies used in a lunar observation, even though the GMT and the >> longitude are both still uncertain. > >I don't think you need to make any special "local apparent time" >observations or calculations. Assuming the navigator has been >using celestial observations all along, but has an incorrect clock, >he will have determined a celestial "fix" whose longitude is off by >almost exactly 15' for every minute of time error. These two errors >will cancel out to reduce errors in computed altitudes, the same way >as Bruce Stark's procedure using local time. The remaining errors >are come from the change in declination (pretty fast for the moon), >and the difference in rate of change of GHA between the sun, planets, >and stars. > >I think, George, that if you go back to your example where you assumed >the watch was 30 minutes fast, and simply change the assumed longitude >to be 7.5 degrees east of the true position, then you'll come up with >results that match Bruce's. This assumed longitude is what a navigator >would have concluded with traditional celestial observations, if he >were misled by a watch that is 30 minutes fast. ------------------ Later, Bill Noyce corrected this as follows- >I wrote: > >> I think, George, that if you go back to your example where you assumed >> the watch was 30 minutes fast, and simply change the assumed longitude >> to be 7.5 degrees east of the true position, ... > >Of course I got this backwards -- the assumed longitude would >be west of the true position in this case (unless I've confused >myself even more). > >Then simply use the (erroneous) watch time as if it were GMT >for looking things up in the almanac. You'll end up with nearly >the same LHA for most bodies, except of course the moon (which >is the point of the exercise). ========================== Well, those comments were most useful, and I am grateful to Bill for clearing matters up. However, I am still having rather a struggle to get the concepts clear in my mind. I hope others on the list will bear with me while I try to thrash the matter out a bit further, perhaps with the help of Bill and Bruce, and maybe others. I am anxious to get to the bottom of it, being aware that the final part 5 of "About Lunars" awaits completion, and this will become part of it. Below is my summary of the matter as I now see it, assembled rather tentatively, with an invitation to knock it down. ==================== If a lunar distance has been measured, together with measured altitudes of the Moon and the other-body at the same time, then that provides a straightforward measure of GMT that stands on its own and needs no other observations. If the azimuths of those bodies are such as to give a reasonable "cut", then position lines of the bodies, taken by comparing observed altitudes with those predicted at that GMT from the Almanac, will give the ship's latitude and longitude directly. --------------------- However, if only the lunar distance has been measured, and the altitudes of the two bodies must instead be calculated from the predictions of the Almanac, then matters are somewhat different. Neither the GMT (needed for looking up the positions of the bodies in the Almanac) nor the observer's position (needed for calculating the altitudes of the bodies as seen by him) is known. In that case, it is necessary to combine the lunar with additional measurements of altitudes, which could be made before or after the moment of the lunar, and very likely, made of different bodies. An important requirement is that these additional measurements are made at a time that is not widely separated from that of the lunar (within 12 hours, preferably). The aim is that any contribution arising from errors in the rate of the clock, or in the dead-reckoning of the vessel's travel during the elapsed interval, will be small. There are occasions when this might be a more satisfactory procedure than direct measurement of altitudes with the lunar. Particularly, if the lunar distance had been measured in the night, in the absence of a visible horizon, then it could be combined with altitude measurements made in the previous, or the following, daytime, when the horizon is clear: perhaps measurements of the Sun. It may not be necessary to make special observations for this purpose. They will be made regularly by the navigator during the course of the voyage, even if he was then uncertain about the accuracy of his clock. Although one observation may be of the Sun's meridian altitude (to provide latitude directly), one should be in a direction well toward the East or West of the meridian, to be sensitive to the longitude. The ship's position, deduced from that (uncertain) clock, but with allowance then made by dead-reckoning to the moment of the lunar, together with the GMT indicated on that same (uncertain) clock, are used to calculate the altitudes of the Moon and other-body at the moment of lunar observation. Although the longitude of the ship, and the GMT itself, may not be known individually, because of the clock error, their combination will be known well enough to provide accurate altitudes. Once the altitudes of the Moon and other-body have been calculated, a precise measurement of the lunar distance should provide GMT to within a minute or two. This allows the ship's position, which until that moment had been based on an uncertain clock, to be reassessed on the basis of the newly known time from the lunar. ================== I would welcome any comments on this matter. I have not only found it difficult to put across, but also difficult to grasp myself. Have I got it right? Could it be expressed more clearly, or more simply? Over to the lunatics... George Huxtable. ------------------------------ george@huxtable.u-net.com George Huxtable, 1 Sandy Lane, Southmoor, Abingdon, Oxon OX13 5HX, UK. Tel. 01865 820222 or (int.) +44 1865 820222. ------------------------------