h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

By my calculations, the evenings are already starting to get longer again. The day of earliest sunset is 12 December. That's in Dublin at 52.5 degrees North. Further south, such as London, Lyons or New York, it is earlier.

So we should start to notice the evenings getting longer soon.

Morning Gnomon

I always reckon on 21 December being the shortest day. Have I been misleading myself all these years?

F/b

You're right, but so am I. I wasn't talking about the length of the day. I was talking about sunset.

Sunrise is a different thing entirely. It'll continue getting darker in the mornings until about 30 December. The day length is of course the difference between sunrise and sunset. That is least on the 21st.

Gnomon, "by your calculations" - did you calculate this yourself? Sounds complex!

No.

I just looked up some tables and did a bit of interpolation.

Shucks! Ruined the myth! I had this idea of you sitting up in a lonely tower with a telescope and a sextant and lots of paper...

You'll have to explain this to me.

If noon is regular at 24 hour intervals, how does the sunrise vary differently from the sunset? I visualise them as expanding and contracting symmetrically about noon. How can the sun start setting later without rising earlier?

Noon isn't really at regular intervals of 24 hours. It varies with the seasons. The earth spins at a constant rate, but the motion around the sun makes things more complicated. I don't fully understand it.

We impose a standard 24-hour time on our day and insist that noon is 12:00, which makes it appear that sunrise and sunset are not symmetrical about noon.

So daylight shifts to earlier in the day, even though sunrise continues to get later?

If sunset is getting earlier, but the days are still getting shorter, sunrise must be getting much later every day, to make up the difference.

Is that right?


F/b

"Noon isn't really at regular intervals of 24 hours. It varies with the seasons."

Help. I've been in this conversation before. If the above is true, how did sailors dead-reckon by the sextant?

How do sundials work?

Please say it ain't so.

If noon was at regular intervals in the 24 hours, how come we need a leap year every 4 years?

Isn't it because the year is really 365.25 days long. If that's the case, then noon must move.



F/b

The length of the year and the day are generated by separate movements: the orbit and the spin of the earth. Not related, and therefore not surprisingly incommensurate. But I'll eat my hat if Gnomon can show that noon is irregular. (Prepares pastry hat, knowing Gnomon's record)

Yes, be careful Recumbantman! Perhaps a cake hat would be tastier?

Long time since I did physics foundation at Uni, so thank you for the year / day reminder

I still don't quite get the day length thing though Gnomon...


Fb

Now, I'm only guessing here, but if it's getting lighter at night, but we haven't reached the shortest day yet, then dawn must be getting later. Until we reach the shortest day at least, and then it can (though I'm not saying it will) get earlier.

And if noon is the mid-point between sunrise and sunset (I'm not sure on that one, but it would make sense), then that must be getting later too.

Sundails would show noon. But noon is not necessarily 12 o'clock. After all, clock are tricky things. We've set up these number systems to organise our day, but that system only pays scant relation to the reality of nature. They was a time when different places in England consider 12 o'clock to happen after different times, for the simple reason that noon occurs at different times depending on where you are (different latitudes, I think, though I may be wrong). Times were standardised so that train time-tables would make sense.


Geggs

Noon is constant for all latitudes but varies with longitude (hence its importance in navigation before there were reliable clocks). Here in Dublin we are that much west of Greenwich that noon comes twenty-five minutes after noon GMT. That's not the point. I'm deeply upset by the notion of noon varying with the seasons (apart from the Summer Time debacle). Gnomon is ominously quiet. Research.

Having driven home from school between 3:45 and 4:30 for the last three weeks, I could have told you that, Gnomon.

B

Life was much easier on the Roman/sundial system. At the time hours were not set in length (60 mins is a strange number anyway) but were instead defined as a fraction (1/6) of the daylight. Thus an hour lunch during the summer was a treat, while an hour lunch in December would not be worth getting up from the desk.
If you're concerned about noon in Dublin, think about the situation in China, where they actually cover (I think) 5 time zones but all use the same time. In that case the government 'noon' in some areas of the country is as much as 3 hours off the 'astronomical noon' which is the median point between sunrise and sunset.
-MotDoc

Sure, but what we're talking about here is sunset getting later before the winter solstice. Gnomon or Master B please explain.

"I visualise them as expanding and contracting symmetrically about noon."

Quite simply, it doesn't do that. I suspect it's all something to with the fact that the Earth is elliptical rather than perfectly spherical.

B

It may have something to do with Earth's elliptical movement around the Sun, and the fact that the Earth moves faster in its orbit the closer to the sun it gets.

It's also a fact that the Earth does not spin around on its axis once every 24 hours. In fact, it spins right around in 23 hours 56 minutes. However in that time, the Earth's position relative to the Sun has moved, so the extra four minutes is required to catch up with the *new* position of the Earth with respect to the Sun.

Still with me?

Now because the Earth moves faster in its orbit when it is close to the Sun, and slower in it's orbit when it it is further out from the Sun, this "catch-up" time is not constant. The 4 minutes allowed is an average. So at some times of the year it will catch up faster than expected, crossing the meridian in less than 24 hours. At other times of the year it will be slower, taking more than 4 minutes to catch up.

I have some astronomy software at home, and to my recollection, if you went out at 12.00 each day for a year, and plotted the position of the sun in the sky at that time each day, at the end of the year you would be left with a design that looks something like a figure of eight. So, for 3 months of the year the sun is left of the meridian, then it goes right, then back left and finally right again.

The time zone issue does not explain the anomaly. It will just shift the "8" right or left. You will see the same figure of eight no matter where you are (apart from the arctic/antarctic cirles of course).

Woodpigeon