Question about finding dates for full moons.

Any of the myriad almanacs of the time would have that info, as would any newspaper. Didn’t libraries in cities have lines you could call to ask general reference questions back in the day? I seem to remember an old movie about the staff of a library reference desk whose main job seemed to be to answer questions that people called in. Conflict/zaniness ensued when the library bought a computer to help/compete with/replace the operators.

Morgenstern, I totally thought “I’d watch a full moon if someone asked me.”

Or an orthodox Rabbi.

He’d just look at you funny and say “Full Moon is on the 15th of the month, of course!” :slight_smile:

(Because that’s how the Jewish calendar is built, from the ground up.)

Here is a nice site that has three different algorithms… The more complicated ones offer more accuracy. They all ultimately depend on the lunar period (“month”) but the more complicated ones encode that number to more decimal places. They go beyond merely dividing by 29.530589.

(As if more accuracy were needed? With that many decimal places, your error is less than a day, even going back thousands of years!)

The bigger difference is that the more complicated algorithms take into account seasonal and cyclical variation in the synodic period, to arrive at a more accurate full-moon date mid-cycle.

The figure of 29.53 days for the synodic period is a long-term average. The actual period from one new moon to the next oscillates about that value, as shown in the second graph from the top on this page.

The primary variation is seasonal, with the synodic period being longer near perihelion, when the moon must travel farther between new moons to compensate for the earth’s faster revolution about the Sun.

I minimized the impact of seasonal variation in my calculation by going December-to-December, from 2012 to 1923. Even if I hadn’t, it wouldn’t have been a big deal; I wouldn’t have been off by more than a day. (And the moon is just as bright, for all practical purposes, the night before or after full moon.) The synodic period varies over longer-term cycles which are of even less consequence (for a practical calculation) than the seasonal cycle.

Over very long time periods, simple algorithms break down because the average synodic period is changing, due to slowing in the rotation of the earth and greater distance of the moon from the earth. Needless to say, this isn’t a concern going back to 1923.

Very nifty graph! Thank you! Intriguing second-order effects!

Yes, and not just “back in the day”. The reference librarian at your local public library can still answer general reference questions. I’m a reference librarian in academia so I’m usually assisting professors and students with their research projects, but since I work at a state university we do serve the public as well and I have occasionally helped members of the local community find information on various topics.

I don’t think I’ve ever helped anyone with research for a novel, but I did once assist a tattoo artist in finding depictions of demons in traditional Japanese art to use as references for a design he was working on.

TelRef - NYPL