NavList:

Bob B continued:
"On the other hand, I read a statement Slocum made about his Lunars (source "The Lunar Distance Method in the 19th Century. A simulation of Joshua Slocum's observation on June 16, 1896", author Siebren Y. van der Werf). ...."her (lunars) longitude agreed within five miles of that by dead-reckoning". "

The "source" you list here, the article by van der Werf from 1997, is a trivial thing. He makes a guess at the date, likely wrong by a day, picks an approximate location, then runs through some armchair math for a lunar observation that Slocum described in general terms in his book. You're learning nothing about Slocum's lunar observation in that article that you couldn't find already in the pages of Slocum's popular account of his voyage. The math in van der Werf's article is fanciful. Nothing about it is related to Slocum's work except the estimated date and approximate location. A lunar "could be" worked in the mathematical fashion that van der Werf provides, but there's no expectation that Slocum did it that way. And remember: van der Werf was writing in 1997... Back then armchair navigation enthusiasts had almost no internet, no access to thousands of online books, no communities with expertise to consult. Like John Karl in 2005, van der Werf in 1997 was working in a vacuum.

You added:
"Plus Joshua only took 3 sights (Sun Altitude, Moon Altitude, LD Sun-Moon). No run of sights, no averaging the sights."

Step back for a moment... How do you know what Slocum did? I'm not challenging you with this question... I'm simply asking you to consider how you know what Slocum did on that single day (just one lunar in the whole circumnavigation ...at least two-thirds of it, with certainty). His logbook is gone. We have only his words from his articles / book (the book was serialized in magazine articles first, then compiled into the final book) and a tiny handful of other references that he wrote. Did he average a few lunars in a row? He could have. He had experience with lunars from the early days of his career at sea. He certainly knew that he could average multiple lunars, as an option. Did he on that day? We don't know. My own guess? Probably no averaging... Not because he didn't consider it, but because he didn't care. He was shooting a lunar out of anxiety... any rough confirmation would help.

You asked:
"What was the accuracy experience during the Age of Sail when they used Lunars to derive UTC?"

Occasionally... excellent (within five miles). Mostly... impossible to judge (because no better longitude to compare against). But the best argument for the accuracy of lunars in that era is simply that many navigators used them regularly for the better part of a century. See the examples from a primary source that I posted yesterday. They worked. And it is worth saying: if Slocum approaching the Marquesas in pleasant weather in 1896 had done everything right, he could have gotten his longitude within a few miles. Did it matter? He was suffering from anxiety (not using that word, but he described it in his book) after 43 days at sea and was perhaps worried that somehow, something was wrong with his dead reckoning... Where were the peaks of Fatu Hiva (southern island of the Marquesas)? All he wanted to know was, 'am I in the right part of the ocean?!' When the peaks of Fatu Hiva appeared on the distant horizon, he decided his lunar must be accurate to five miles! But we don't know that. It's not a simple conclusion, and the details of his original observations do not exist.

A small note: mentioning "UTC" here is anachronistic since that specifically implies UT "coordinated" with atomic time. UTC has only been around for a bit more than half a century. In early eras, the safe term is "Greenwich Time" (not GMT necessarily because during most of the period when lunars were actively used at sea, the tables for lunar distances yielded "Greenwich Apparent Time" which saved the navigator the minor trouble of correcting a time sight for the equation of time). You can also just say UT or if you want to be more precise, you can say UT1, since these are just modernized terms for the original GMT.

You concluded:
"I'd assume the Navigator had a Good Sextant (for the time) and was skilled. And he/she was taking sights onboard a ship (and not on land). I'm curious what accuracy a navigator on a 35 foot offshore sailboat can achieve? "

Under good conditions, you can assume +/-0.2' for the lunar itself. Assuming all steps are included and handled properly in the clearing arithmetic, that accuracy translates to +/-24 seconds in absolute time. And then to longitude, that's +/-6 minutes of longitude [6 nautical miles in equatorial latitudes, 6·cos(Lat) miles in higher latitudes]. But the only proof is in the logbooks, and these can be difficult to interpret for the modern navigation enthusiast's requirement of a specific accuracy. There's lots of variety, and in most cases lunars were employed far from land. Without some other more accurate determination of longitude, the accuracy of lunars was normally unknown. Yet the evidence is clear enough: many navigators voyaging great distances, crossing oceans, shot lunars. They worked, by the evidence of the lunarians themselves.

And what about today? Should we compel navigators to learn lunars today "just in case" they lose Greenwich Time?? Of course not! And not because there's anything wrong with lunars but because the second half of that sentence is nonsense. In the modern world, no one is ever going to lose absolute time in a way that can't be easily "fixed". Sure, people can invent fantasies in their heads... and I have heard plenty of them. What if your boat is demasted in a hurricane and you lose all you electronics, huh? But you still have a sextant?? And lunar distance tables?? That's absurd. Or what if nuclear war breaks out and all radio time signals are terminated? And you're worried about celestial navigation?! Also absurd.

Although we don't need lunars in the modern world for time, they do serve a valuable purpose: they are one of the best tests of a sextant's quality and a navigator's skill. You can calibrate your sextant's arc correction by shooting lunars. They're not the only option... maybe not even the best option. But when you add in the "meta" (not practical navigation factor) of experiencing one of the most famous tools in the history of celestial navigation, then there's a solid case for experimenting with lunars. At sea? Maybe for fun, for Slocum-esque "bragging rights"?

Frank Reed
Clockwork Mapping / ReedNavigation.com
Conanicut Island USA

PS: More than twenty years ago, I attended a lecture by John Rousmaniere, prolific author, famous for "Fastnet Force 10". He spoke briefly about Joshua Slocum and his status as an icon in the world of ocean sailing. And Rousmaniere noted that eventually the world of yachting would have to come to terms with Slocum's criminal record involving some sort of assault on a young girl. That still hasn't happened. Any fan of Slocum should be concerned here. Again, we face the same problem with his navigation... we don't know what really happened. There is little primary source evidence. But we do know that he spent a few weeks in prison. This is not a fiction... Just something to keep in mind.