NavList:

Martin, how is Southampton? A nice evening sailing up the Solent? :)

You wrote:
"Regarding Lunars, I took the sights two days ago because I had to wait until the Moon was visible at night but also had clear skies to take the sights."

As far as that goes, in the era when lunars were widely used to get GMT, lunars were mostly (majority of cases, but not overwhelming majority) taken using the Sun as the other body. Like most celestial navigation, lunars were a daytime activity. But of course sometimes the weather and the calendar don't allow that, and the stars do work.

You continued:
"I am far from being an expert in Lunars, so all suggestions and comments are welcome on the following procedure:
I picked Aldebaran to measure the lunar distance because its declination was close to the one of the Moon (around 2 degrees of north declination difference). It is my understanding that the relative similar path on the sky increases the speed between the two bodies and should contribute to get a better precision."

Uh, no, that's not an important condition. Do you recall reading that somewhere... the idea that the two bodies should be at similar declinations? Do you remember where, by chance? Because that's not so, but Aldebaran is still a good choice usually. The Moon has a motion on the celestial sphere relative to the other celestial bodies. That motion can be described by a little arrow or vector, and that vector is always close to parallel to the ecliptic. So that's what you look for on the sky: find the ecliptic. And the prefered other bodies for lunars will be those that are fairly close to the ecliptic. That's how the lunars stars were selected 250 years ago. The bright planets work, of course. And obviously the Sun works since it's dead-on the ecliptic. Aldebaran, in fact, is almost always a good choice for lunars since it is quite close to the ecliptic. Its declination, close to the Moon's or otherwise, is not relevant. Aldebaran would only rarely be out of line with the Moon's motion vector on the celestial sphere (relative to the stars! ignore the common daily motion), and that would happen only when the angular distance from the Moon to Aldebaran is relatively small (this is the lunar distance we are trying to measure). There are other good reasons for avoiding lunars with short angles (usually avoid lunars less than about 15 to 20 degrees of distance).

You wrote:
"Because of cloudy skies, I took the sights a couple hours after sunset. I used the A-12 bubble sextant to take the Hs' of both the Moon and Aldebaran and then the Mark II for the lunar distance. The Mark II telescope is not very powerful, so it was kind of difficult to match the Aldebaran with the Moon lower limb."

Ah yes, cloudy skies... Even more so than standard celestial navigation, lunars need relatively cloudless skies -- we need to be able to see both objects clearly. Here you mentioned bringing the star to the Moon's "lower limb". Did you mean to write that? If it was just a "typo", no problem, of course. But to clarify for anyone else new to lunars who may be reading along, you need to bring the star to the middle of the Moon's bright limb, generally the leading or trailing limb of the Moon relative to its "vector" of motion against the background stars. Think of that bright limb as your horizon (in a real sense, it is... just very far away), and you sweep the star across that limb so that it just barely touches --this is essentially identical to "swinging the arc" in a common altitude sight. Assuming you were shooting your Aldebaran lunar on the evening of Jan 18 (local date), then the distance would have been around 33°, and you would have been shooting a "Far Limb" lunar.

"After all calculations, the difference between the real time and the Lunar method time was 4 minutes and 40 second."

Assuming you have your sight details, you should try running them up in my web app, which is exceptionally accurate and well-tested ...and it does all the work for you, which is nice. :) Here's a case with some preset values: Lunars Analysis web app. The error you quote here is considerably larger than you should expect so something is off somewhere. But I find when working with lunarian fans that the errors decrease very quickly. :)

You concluded:
"It is my understanding that the best possible precision with the Lunar distance method is 2 minutes"

No, definitely not. I've seen this claim a few times quoted in various books in the past few decades. It's founded on an armchair analysis of lunars (note: the authors of the books are armchair analysts --not you! :)). The logic goes something like this:

"The Moon moves at a rate of about one minute of arc in two minutes of time. We "all know" that we usually consider ourselves doing good work when we shoot Sun and star altitudes in common celestial navigation accurate to one minute of arc. That's about as good as it gets, right? And therefore [here's the mistake] when we turn our sextants to measuring lunar distances, we can only expect to get to the nearest minute of arc so that implies nothing better than about two minutes of GMT."

Sextants can, in fact, measure angles five to ten times more accurate than a minute of arc, depending on the quality of the instrument, the care that has been taken adjusting it (especially eliminating or measuring its index error), and the skill of the observer (this is less significant than most people imagine, in my experience). We don't get that level of accuracy in common celestial altitudes because we are bumping up against the 'system limit' of the process. And the reason for that system limit is the uncertainty of the sea horizon. That is what limits us to +/- a mile or so in common celestial fixes. By contrast, with lunars, the edge of the Moon is quite sharp down to a a few percent of a minute of arc (mountains and valleys on the lunar limb are the final limiting factor. I advise most navigators with basic skill and decent quality metal sextants that they should expect about +/- 0.25' for most lunar sights (count that as a 'one standard deviation' width), and if you average four or five in a row, you can expect accuracy about four times better or +/- 0.1'. And just as +/- 1.0' in lunars would imply GMT errors of about 120 seconds, then +/- 0.1' in lunars implies GMT errors of about 12 seconds. That's equivalent to +/- 3 minutes (of arc) of longitude which in mid-latitudes is a little more than +/- 2 nautical miles in east/west position. That is what you can expect from lunars under very good to excellent conditions.

Can lunars be better even than that? Probably not by much, but some rare observers have exquisite visual acuity and coupled with an excellent sextant and excellent observing conditions, it's possible to increase accuracy by another factor of two or even five. Mordris Fersters, in his NavList posts, has reported some high accuracy lunar statistics which strongly suggest he has that rare level of exquisite visual acuity.

Frank Reed