PDA

View Full Version : Sunrise times and the winter solstice

Paul Beardsley
2011-Dec-27, 11:33 AM
Looking at the sunrise and sunset times for London, I see that immediately after the solstice (22 December this year) the days start to get longer. No surprise there, of course... but how come the sunrise times get later until 4 January? I'd imagined it would be symmetrical around midday.

Is it an issue with the Earth's tilt? Elliptical orbit? An arbitrary decision in how GMT works?

Tog
2011-Dec-27, 01:04 PM
One of the first two as I recall. I want to say it's the orbit specifically, but I'm not positive that's the reason. The little guy in my head says that it's because the aphelion and perihelion are really about two weeks behind the solstices right now.

See below for corrections. Possibly above as well.

Perikles
2011-Dec-27, 01:20 PM
It is all to do with the equation of time (http://en.wikipedia.org/wiki/Equation_of_time) and the fact that mean time is used, and not solar time. Apparently, the effect is much larger on Mars. I'm sure you wanted to know that.

I just love that use of equation meaning correction.

Paul Beardsley
2011-Dec-27, 02:02 PM
Thanks for the answers, guys. From the sound of it, it is symmetrical around noon - but only noon as read on a sundial, not necessarily noon according to clocks.

I shall look into this in due course.

ammer
2011-Dec-27, 03:46 PM
Thanks for the answers, guys. From the sound of it, it is symmetrical around noon - but only noon as read on a sundial, not necessarily noon according to clocks.
That is exactly what's going on. Check out Wikipedia's 'Equation of time' on how sundial noon varies during the year. There are two effects that cause this: the earth's tilt and the elliptical orbit. The elliptical orbit makes the earth go round the sun with varying angular velocity, going fastest at perihelion. And in addition to that, the apparent path of the sun along the tilted ecliptic becomes slightly distorted when mapped onto earth's equator along meridians. The total effect is most profound around now at some PLUS 0.5 minutes shift-of-noon per day.

a1call
2011-Dec-27, 03:58 PM
Accordingly the local noon is shifting by about one minute (Later) every day at my locale.
Since the length of daylight is more stable, it makes sense for the sunrise to shift later for a few days at this time of the year.

Hornblower
2011-Dec-27, 04:18 PM
I will attempt a verbal description of the phenomenon. Illustrating it visually on a globe or a chart is much easier.

Suppose the Sun moved along the celestial equator at constant speed. Then a constant rate clock could be adjusted to remain perfectly in sync with it. Let us imagine a fictitious body that moves in this way and call it the Mean Sun.

Now let's imaging the real Sun moving along the inclined ecliptic at a constant speed, as it would in the event of a circular orbit. At the September equinox its slantways motion makes its forward progress in right ascension slower than average compared with the Mean Sun. Thus the observer's local meridian passes it progressively earlier for a while. As it levels off toward the solstice, its motion across the longitude lines where they are closer together than at the equator makes its right ascension forward progress faster than average, so it overtakes the Mean Sun. The result now is that the observer's local meridian is progressively later in catching up with the Sun during this period. As it angles toward the next equinox, it transitions back to slow forward progress. This cycle repeats itself during the final six months. Thus sunrise, noon culmination, and sunset all tend to get later each day around solstice time, nearly in unison.

The Earth's orbital eccentricity makes the Sun's ecliptic speed faster around the December solstice and slower six months later. This exaggerates the effect in December and nearly cancels it in June.

Clear as mud? Once again, the limitations of a forum like this are frustrating.

ammer
2011-Dec-29, 03:25 PM
I will attempt a verbal description of the phenomenon. Illustrating it visually on a globe or a chart is much easier. ...
Thanks for explaining this topic with the tools at hand: just words. Tough topic, but a clear story imho.
As a newbie, I'm a bit struggling with the balance between explaining and referring to a good source (like Wikipedia). My post #5 touches the topic but barely explains anything. So thanks for filling in the dots I feel I shouldn't have left. I think I start to get the idea of the forum.

Tensor
2011-Dec-29, 04:14 PM
Thanks for explaining this topic with the tools at hand: just words. Tough topic, but a clear story imho.
As a newbie, I'm a bit struggling with the balance between explaining and referring to a good source (like Wikipedia). My post #5 touches the topic but barely explains anything. So thanks for filling in the dots I feel I shouldn't have left. I think I start to get the idea of the forum.

A way I think of it ammer, is to explain as much as you can, as much as is sensible (you don't want a solid wall of words in three parts), or as much as you feel comfortable with and then list your source(s) and maybe some other links. Or, you can just provide a link with a comment such as "I found this page to be a good explanation for me"

Paul Beardsley
2011-Dec-29, 04:22 PM
I found the Wikipedia reference in post 5 very helpful - and the post itself, of course!

What confused me (and what I found amusing later) was that the times given for sunrise and sunset were the actual times of those events, and so I assumed that 12 o'clock would also be the actual time of the sun crossing the meridian. Now I have a visual image of Apparent Noon and Mean Noon coinciding, but with Apparent Noon moving ahead into the afternoon as winter approaches, them moving back towards Mean Noon as spring approaches, then falling behind Mean Noon as summer approaches, then moving forward to meet back up with Mean Noon again as autumn approaches, and so on.

Chew
2011-Dec-29, 04:30 PM
I will attempt a verbal description of the phenomenon. Illustrating it visually on a globe or a chart is much easier.

Or with a picture.

http://i.imgur.com/nC9pV.png

This multiple exposure photograph was taken at the same time everyday (mean time, ignoring daylight saving time) every two weeks over the course of a year. The shutter was left open from sunrise to the normal time at the solstices and the point where the Analemma (http://en.wikipedia.org/wiki/Analemma) crossed. That funky figure 8 you see on some globes is an analemma.

The story of how this picture was taken: http://www.twanight.org/newTWAN/photos.asp?ID=3001422

Centaur
2011-Dec-30, 03:02 AM
Looking at the sunrise and sunset times for London, I see that immediately after the solstice (22 December this year) the days start to get longer. No surprise there, of course... but how come the sunrise times get later until 4 January? I'd imagined it would be symmetrical around midday.

Is it an issue with the Earth's tilt? Elliptical orbit? An arbitrary decision in how GMT works?

Paul – Hornblower in particular provides a fine and thorough explanation. The short way to describe the situation is to note as you suggest that with a sundial, the latest sunrise and earliest sunset occur on the day of the winter solstice. We moderns have developed a standardized way of keeping time so that each day from noon to noon is equally long as measured by an atomic clock, which is not the case with a sundial for the reasons given by Hornblower. Even at the meridian of Greenwich, the Sun does not cross the meridian at 12:00 each day, although on average it does. The result is that the earliest sunsets occur in early December, and the latest sunrises occur in early January, if measured by the modern standard for clocks. There is a similar although reversed and somewhat muted effect around the time of the summer solstice.