[chris (OP)]

We began discussing this the other day. Is there a consensus? And how long does a "day" actually last on the lunar surface?

[evan_au]

You have to distinguish "local" time and "global" time on the Moon.

For Local time on the Moon, Noon is when the Sun reaches its highest point overhead.
- One day is the interval between two adjacent Noon events.
- One Lunar Day = 29.53 Earth Days.
- See: https://en.wikipedia.org/wiki/Moon#Appearance_from_Earth

For Global time on the Moon, there is a Prime Meridian on the Moon, passing through the center of the visible disk of the Moon (similar to the Prime Meridian on Earth, which passes through Greenwich).
- In global terms, Noon is when the Sun is at its highest on the Prime Meridian.
https://en.wikipedia.org/wiki/Selenographic_coordinates

I am not aware of any specific timescale that is used on the Moon - to date, visits have been so short (a couple of Earth Days) that the Sun hardly moves in the sky. So astronauts would keep Earth time.
- If there is ever a permanent base on the Moon, they would need to decide on a suitable calendar and clock. Since the Moon is so inhospitable, they may just follow an Earth timezone (or 3 timezones, if they need to do shiftwork).
- The Lunar calendar could be marked in alternating 29 & 30 Earth-Day "Lunar Months", where you know that (where you are on the Moon), certain days are always too cold, certain days are always too hot, and a few in the morning and evening are relatively mild.

This is unlike Mars, where a permanent settlement is far from Earth, and has a clear incentive to change their clocks so that they register Mars Days rather than Earth Days: 24 hours and 37 minutes.

[alancalverd]

The time everywhere is UTC.

Only stupid politicians think the local azimuth of our nearest star has any significance for national pride, and even then, they mess about with "daylight saving" (an obvious lie) and "datelines" just to make work for the people who print timetables, and to make life difficult for those who depend on tide tables. 

20190701095638Z

[Janus]

You have to distinguish "local" time and "global" time on the Moon.

For Local time on the Moon, Noon is when the Sun reaches its highest point overhead.
- One day is the interval between two adjacent Noon events.
- One Lunar Day = 29.53 Earth Days.
- See: https://en.wikipedia.org/wiki/Moon#Appearance_from_Earth

Though to be accurate, that is the "Mean" solar day for the Moon.  Any given solar day can be from 29.13 to 29.93 Earth days long, varying over the course of the Earth's year.   This is due to the eccentricity of the Earth's orbit around the Sun and the slight variance this produces in its orbital velocity.

[evan_au]

20190701095638Z

- 2019: You need a starting date, sometime; in this case it was 4-6 years after the birth of Jesus (the monk who worked it out made an error in his calculations)
- 07: This is the month of July, created and named eponymously by the Roman emperor Julius Caesar. He gave it 31 days, so it wouldn't be less than any other month
- 01: We count daily periods, because that is a natural activity cycle for creatures on Earth, baked into the DNA of our cells. I expect that we would continue something approximating this to places outside Earth (unless we modified it by genetic engineering).
- 09: The day was divided into 12 hours on a sundial, and by extension with another 12 hours at night. This gives us a 24-hour day. But the light/dark cycle may be quite different in different parts of the universe (or absent altogether), so it is a bit parochial
- 56: The Babylonians used Base 60 for their calculations, because 60 has many factors, making it easier to calculate with fractions. These days, computers use binary very successfully. If you need to work with fractions, symbolic maths packages do it better.
- 38: Ditto. But at least we are getting to the fundamental unit of time (in the Metric system), and something that can be usefully reproduced by humans elsewhere in the universe. A Cesium clock (or it's optical successor) could measure time in seconds accurately, even if time flows differently due to velocity or gravitational potential.
- Z: The indicator for UTC=Zero hours, relative to the Greenwich Observatory*, on Earth. The only physical place that actually uses this time is Iceland (plus pilots and the military). But this includes leap seconds which are adjusted for the irregular and slowing rotation of the Earth. These are irrelevant to anyone elsewhere in the universe (and with the spread of GPS and equivalents as a backup, increasingly irrelevant on Earth).

*The French agreed to Greenwich as the Prime Meridian in return for the UK adopting the metric system. Last I heard, the UK government still hadn't ratified this agreement....

So I suggest that the only part of the comedy of accidents we call UTC which will be particularly significant to human scientists elsewhere in the universe (beyond the Moon, as a vassal state) is the seconds unit, and something approximating days.

[Bored chemist]

And any system that doesn't give me the weekend off can go +  **** itself.

However, I agree that  the world would be a better place if we stopped pretending that "daylight saving" achieves anything.

"… “Only the government would believe that if you could cut a foot off the top of a blanket, sew it to the bottom, and have a longer blanket.”"
More importantly

The time everywhere is UTC.

Guess again.
Yours sincerely
A Einstein

[alancalverd]

Not only pilots but ships' navigators, astronomers, polar explorers and all electronic transactional systems - in fact anyone to whom the time in two places matters - are based on UTC. You might display "local time" in the passenger lounge but that is sweet FA use if you have to divert from Delhi (UTC + 5:30) to Karachi  (UTC + 5)  or transit from Arizona (no summer time) to California (summer time) except some parts of Arizona use summer time. And of course south of the equator it's all different again, except that not all Australian states use summer time, so the TV schedule is more like a lottery.

The question occasionally asked in court, and indeed when looking for supraluminal particles at CERN, is "did A precede B?" It would be a lot easier to answer that question if everyone used the same clock.

Admittedly the origin of UTC is a bit arbitrary, and the counting of hours, days, months etc likewise is historic rather than logical, so I'd happily vote for a Julian date that incremented by 1 every  86,400 seconds - just like the Excel clock in everyone's computer! 

As for "scientists elsewhere in the universe", as far as we know, there are none. Those in earth orbit need to synchronise with those on earth, and any who get to Mars will still be talking to earth scientists for many years, so UTC will have to travel with them.

IIRC it's still delightfully arbitrary because the primary standard for navigational Zulu time is actually broadcast from Boulder, Colorado, not Greenwich.

[Bored chemist]

Those in earth orbit need to synchronise with those on earth,

Well, we don't have many people in orbit for long, but the GPS satellites don't run to UTC.
And, of course, they are the ones used by

Not only pilots but ships' navigators, astronomers, polar explorers

It's perfectly conceivable that the scientists in other solar systems are using Caesium clocks much like ours.
It matters little where you are- the properties of Cs that make it a handy reference here may also apply elsewhere.
But it's unlikely that they consider 9192631770 to be significant.

[chris (OP)]

It would be pretty miserable on the moon wouldn't it, with a sunny day going on for a fortnight...

[Bored chemist]

It would be pretty miserable on the moon wouldn't it, with a sunny day going on for a fortnight...

I opened bar there once but it didn't do well.
The patrons said it had no atmosphere.

[evan_au]

Some of the upcoming low-cost missions to the Moon will be running Android software, which is based on the UNIX operating system. This has a clock which counts seconds since 1 January 1970 (and does consider leap seconds).
- It uses a 32-bit counter, which will wrap around in the year 2038 - a mini "Y2K problem"
See: https://en.wikipedia.org/wiki/Unix_time

Time between computers on Earth is often synchronized using the Network Time Protocol (NTP), which has a 32-bit seconds counter, and a 32-bit fractional seconds counter, with 0=1 Jan,1900.
- This will work well for Earth and Lunar orbit up until the year 2036, when the seconds counter rolls around to 0
- Hopefully by then, everyone will have shifted to 64-bit seconds: NTPv4
- This protocol won't work well for Mars (or further in the solar system), because the time to ping a time server on Earth gets too long - the protocol would need to be tweaked to send multiple pings in parallel, rather than one at a time.
See: https://en.wikipedia.org/wiki/Network_Time_Protocol#Timestamps

The time everywhere is UTC.

It's not called that in English, French, or other languages. English speakers still call it "Coordinated Universal Time", which does not spell "UTC":

English speakers originally proposed CUT (for "coordinated universal time"), while French speakers proposed TUC (for "temps universel coordonné"). The compromise that emerged was UTC.

In fact, the whole thing is one big compromise, with errors, corrections (and corrections to the corrections) - but a compromise that works, at least on Earth.

Some other compromises/corrections:
- The number of daylight hours was fixed at 12, inscribed on a sundial. But the length of the day varies between summer and winter. So hours were longer in summer, and shorter in winter. How short depended on your latitude. It was too hard for mechanical clocks to emulate this pattern, so the hours were made of equal length.
- The number of months is 12, reflecting the way various politicians have played with it (Augustus Caesar gave us August), making up the 12 months in UTC
- Together, Julius and Augustus added 2 new months to the traditional Roman 10-month calendar. This produced some oddities in the English month names, like October (think octopus, Eighth month), November (9th month) and December (think decimal system, 10th month) are actually the 10, 11th & 12th months respectively.
- The number of days in each month has an odd pattern, with month 02 (February) being particularly bizarre. In years divisible by 4, it has an extra day (a correction introduced by Julius Caesar). But in years divisible by 100, it doesn't (mostly).
- This latter tweak was added by Pope Gregory in 1582. He achieved this by dropping 10 days out of the calendar.
- This ends up as a rational approximation to 365.256363... days in a year, a fact that is surely only of relevance to someone living on Earth or Moon.
- Because of a spat dating back several centuries before, Russia continued working on the older Julian calendar until 1918, when they dropped 13 days out of their calendar to catch up. This means that their "November Revolution" actually happened in October.

In reality, just as my smartphone currently displays time in Sydney, Zurich and Perth, I am sure that future smartphones will have no problems displaying the time at Greenwich (Reykjavik actually), on the Moon, Mars, or further afield (although  leap second adjustments from Proxima Centauri might take a while to catch up on Earth - about the same time that News will arrive).

[alancalverd]

I've never heard anyone say UTC or any other combination of the letters, but all yachtsmen, aviators and their ground stations with whom I have conversed, prefer Zulu time where local time can be confusing (i.e. pretty well anywhere!)

Z is indeed closest to L at Greenwich. Reykyavik happens to use Z because it is an important airport and seaport. Solar time in Iceland is about an hour behind Z but fairly irrelevant either side of the equinoxes.

The fatuity of quasi-local time is most apparent in western Spain. Thanks to European standardisation, the local time is currently Z + 2, whereas the solar position is Z - 1, and at that latitude, the position  of the sun is indeed significant.   

[evan_au]

western Spain. Thanks to European standardisation, the local time is currently Z + 2, whereas the solar position is Z - 1

When I first visited Spain, I found that the restaurants weren't yet open when I went looking for dinner.
And most people seem to go out for dinner around the time I would think of going to bed...

So maybe the Spanish people just live by the Sun, and not by the clock...

[Colin2B]

The time everywhere is UTC.

It's not called that in English, French, or other languages. English speakers still call it "Coordinated Universal Time", which does not spell "UTC":

English speaking originally proposed CUT (for "coordinated universal time"), while French speakers proposed TUC (for "temps universel coordonné"). The compromise that emerged was UTC.

I generally use UTC because English speaking instructors on all the sailing navigation courses and exams I’ve been on call it UTC (YOU TEE SEE) mainly because that is what is written on all the Admiralty and almanac tide tables. Times for various zones are then shown as UTC+1, etc.
Conversationally it’s GMT, BST or ‘local’, over the radio Zulu to coastguard etc, although a lot of recreational radio traffic is conversational.

[alancalverd]

western Spain. Thanks to European standardisation, the local time is currently Z + 2, whereas the solar position is Z - 1

When I first visited Spain, I found that the restaurants weren't yet open when I went looking for dinner.
And most people seem to go out for dinner around the time I would think of going to bed...

So maybe the Spanish people just live by the Sun, and not by the clock...

That's why Spain is such a great country!


No point starting work before 0900 - it's still dark.

Far too hot to work between 1200 and 1500.

No point in working after 1700 - the rest of Europe has closed down.

So what do we do after a 5-hour working day? It's still broad daylight, so the tapas bars open until it's cool and dark enough to eat dinner (not before 2000, often later).

I only work in Spain occasionally, but could quickly get used to it. 

[chris (OP)]

It would be pretty miserable on the moon wouldn't it, with a sunny day going on for a fortnight...

I opened bar there once but it didn't do well.
The patrons said it had no atmosphere.

Should have opened a Star-bucks instead...

[Bored chemist]

I've never heard anyone say UTC

I have.
I have also heard it called "Greenwich mean time,..  well, OK yes, GMT doesn't exist any more... but you know what I mean".

It's important to distinguish it from the time according to the GPS network (and also from TAI)

[alancalverd]

Anyway, to answer the original question, IIRC the time actually used on the moon was EST. Everyone who has ever been there had a mechanical wristwatch that was synchronised to local time before departing Cape Kennedy (as it then was).