NavList:

When the Moon eclipses another celestial body, it's an opportunity to determine GMT/UT and longitude. But when the Moon eclipses the Sun or eclipses a distant star (normally then called an "occultation"), the math, both today and historically, was fairly complicated. But if we turn it around and consider a lunar eclipse, it's all much simpler. The Moon is "up there". The Earth's shadow is cast onto the Moon blocking the light of the Sun. And the events on the lunar surface are observed at the same instant of absolute time by everyone on Earth who can see the Moon. Like the eclipses of the moons of Jupiter, all we have to do is look...

There's a "not quite total" lunar eclipse happening on Friday 28 August 2026 UT date (starting on Thursday 27 August by local time in much of the western hemisphere).  Penumbral phases are not worth attention. The real eclipse --the partial lunar elipse-- begins at 2:33 UT and ends at 5:52 UT. These events can be timed to the nearest minute or two. But so what? The local zone times will match the predicted UT of the events offset by the normal Zone Descriptor. So that tells you nothing! Instead the observer needs to take note of the Local Mean Time (or some equivalent or surrogate) at the instant of the partial eclipse begin and end events.

So how do you do that?? It's an open question, usually glossed over in modern accounts of this process... how would you do it? How with minimal instrumentation? How with a sextant?? How might an interested young student do it? Or imagine yourself as Ptolemy, about 1900 years ago, writing up methods for observers across the known world back then, preparing to watch a lunar eclipse. What should they record to convert their observations into a map of the longitudes of the world??

Frank Reed