sun/moon phenomenon

Factual Questions
21

I’ve got my copy of Walker here. The biggest referene it gives is, as bonzer says, Minnaert. It also lists a set of articles, almost all of them entitled The Moon Illusion:
American Journal of Physics Vol. 11 p. 55 (1943) E. Boring

** Science** vol. 136, pp. 923, 1023 (1962) L. Kaufman and I. Rock

** Scientific American** Vol 207 p. 120 (July 1962) L. Kaufman and I. Rock

Science vol. 167 p. 1092 (1970 F. Restle

** Bulletin of the American Meteorological Sciety** vol. 26 p. 212 (1945) A. Miller and H. Neuberger

Have fun!

22

Oh wow. So many intelligent people with so many good replies. I hope I don’t sound dumb in comparison, but… I think Chronos is right. It’s a great circle problem. I didn’t say anything earlier because I’ve never noticed this phenomenon before, but tonight I did whilst out for a walk. I think that you can verify for yourself that there is indeed a straight line pointing out from the light side of the Moon to the Sun. Cabbage’s straight edge idea might work, but it could be rather unweildy.

Try this instead. Get a length of string, say 5 feet or so. (I used my shoelace, since I was away from my room at the time, but I don’t suggest this.) Hold it taught, one end in each hand. Verify that no matter how you hold it, it appears to be a straight line. Now go out some time when the Sun and the Moon are both in the sky. (You actually need to get moving on this - it won’t be a waxing gibbous for long.) Close one eye. Hold one end between your open eye and the Moon, and hold the other end between your eye and the Sun. Be sure you’re holding it taut. What you should have, if I did it right, is a line traced out in the sky connecting the Sun and the Moon, and the light side of the Moon should be pointing along this line. That is, this line will be perpendicular to the line segment which connects the two tips of the dark crescent. This is the perpendicular bisector that was mentioned in the OP, and as you can see, it does not miss the Sun.

Now, why does this happen? Like I said, it’s a great circle problem. It’s not so much the great circle of the sky as the great circle of your perception. Everything you see is a projection onto a sphere. And I know how irritating spherical projections can be. If you’re looking over a small arc, then your concepts of Euclidean geometry hold fairly well. However, if you’re looking over a large arc, they’re not so good. The arc connecting the gibbous Moon and the Sun is more than 90°, so it’s too large to be intuitive.

We see the horizon, and we think of it as a straight line, which it is. We think of a ray coming out from the setting Sun, which should also be a straight line. On a Euclidean plane, if two rays are going apart from each other, they’ll always be going apart from each other. On a spherical plane, that’s only true for the first 90° - after that they start coming back together. So, when the setting Sun’s ray hits the gibbous Moon, it should be going toward the horizon, not away from it. And this is exactly what you see.

I’m sorry, Podkayne and Irishman, but I don’t think your explanation holds. I think the Sun can be thought of as a point object for this. Also, relative distances don’t matter either. If the Moon were 10 miles or 10 Billion miles away, the light side would still point in the same direction in the sky, although the appearance of the Moon would be different.

Sorry if I explain things badly; I’ll try to come up with something better. In the meantime, feel free to refute me. But do try the string experiment - it just might work.

23

I checked out the gibbous moon last night, and I agree that there’s definitely a great-circle illusion happening because the Sun and the Moon are so far apart in the sky. Also, I was in error before; the “lip” of the moon pointed down relative to the horizon. (Or at least it looked like it; I didn’t use a plumb line, but I should have.)

But I still maintain that the great-circle illusion occurs in addition to the parallel-rays thing. It’s just a simple matter of geometry to show that the terminator is not perpendicular to the Sun-Moon line. I encourage everyone check out the crescent moon, which will be closer to the Sun in the sky. I think you’ll find that the great-circle problem is greatly reduced.

I’ll have to see if I can get my hands on a copy of the book bonser mentioned.

24

Amazon carry it here.

25

Thanks for all the replies. I think I understand why it’s happening now, although I’m not going to attempt to put it in my own words.

Podkayne and Irishman definitely had a key point about things being parallel. Cabbage’s plane (as opposed to line) comes into play as well, as did several other people’s responses.

That’s very important as well. I spent some time playing around with spheres, and I came up with a moon-sun-earth position that would give the effect I was seeing.

I used a string to check the other day, and of course it was perpendicular. It was very surprising though, in that it really didn’t look like it should be working :slight_smile:

Again, thank you all for clearing this up. It’s a very strange illusion.

-Tim

26

I think this question has been answered pretty well, and I’m glad we all have a better understanding of it now. At least, I know I do. I’m going to wait and take a look at the crescent Moon before I claim to know what I’m talking about, but I have to say that I still don’t fully buy the parallel lines idea. I would like to point out that on a sphere, there is no such thing as parallel lines. If a line seems to stay parallel with the horizon, in fact it is not a line and all, and is a curve. Just like the lines of latitude on Earth that are not the Equator. And I think that, whether we agree that the Sun’s rays are parallel or not, we can at least agree that they’re straight, right? Anyway, as I said, I am eagerly awaiting the crescent Moon…

27

Achernar: For a discussion like this, you need to distinguish between Euclidian and non-Euclidian geometry. When projecting onto a sphere (as we do when we look at the sky), the rules change. Parallel lines cross, and all lines curve in threespace. Applying Euclidian geometric rules to this will only confuse you.

28

I read this thread with only passing interest when it came up but a week or so ago we had close to a half moon as I was walking with a friend and saw it (and the sun) I took some crude ‘eyeball’ measurements and sure enough it did not line up.
Being an artist, the next night I took my T-square out side and guess what? A perpendicular line from the center of the half moon (in relationship to its shadow [or ‘sunny side’]caused by the sun) came no where near the sun. After the sun set, the perpendicular line (guaged by the T-square and a clear drawing triangle) still shot upwards from the horizon at about 16 degrees (it was an adjustable tri-angle).
Not to be a trouble-maker, because I truly admire the knowlege of the posters here, but simply citing someone who said it was an illusion doesn’t do it for me. This thread dealt with a gibbous moon so I get the problems mentioned by galt in the post previous, but a half moon (yes, I realise this is still a curved line) puts us in a more direct line viewing position. Any way you cut it the moon and the sun are big ‘balls’ floating in space and the light from one will shine straight and directly on the other - big circles or no!
Based on my direct and semi-scientific observations, I think this phenom has not been sufficiantly answered.
Any chance of bringing in the Master? We are fighting ignorance after all. No?

29

warmgun, how long is your T-square? It sounds like you’re doing exactly the same thing that Rufftim did with a pen. I don’t think you can get an accurate picture like that, because your straight line does not subtend a large enough arc in the sky. The perpendicular line should indeed point up, so there’s nothing wrong there. The actual angle that it should, right at sunset, I calculated somewhere…here:
http://www51.homepage.villanova.edu/christopher.pilman/image/SUN-MOON.GIF

Well, I’ll try to explain better later.

Wouldn’t it be cool if this thread resurfaced for one week every month?

30

This thing is perplexing sho 'nuff.
The question is not how long my T-square is but how close you hold it to you. In my experiment, it reached from the moon to where the sun should have been were it to have been perp to the shadow line on the moon. I followed the provided equation, but I was very careful in taking ‘measurements’.
If we are looking at a half moon, aren’t we standing in a way as to be, in line of site, perp to a line perp from the shadow line on the moon to the sun? (hope that made sense)
If so, why the illusion?
Waiting to hear more…

31

I hate to be picky (well, not really) and a month late, but I just came across this thread. I think you meant “latitude,” Chronos.

32

Let me try and sum up what seems to be the problem.

In an earlier post Chronos claimed it was a great circle problem, and I said something to the effect that it didn’t seem to me to be so. After giving it some more thought, I realize that it basically is a great circle problem.

When we look at the sky, we do tend to perceive it as the interior of a sphere (the sphere, of course, is nonexistent, but the perception is certainly there). For that matter, imagine yourself standing at the center of a sphere, and say you see an (infinite) line somewhere in space–How do you perceive it? You perceive it as an arc of a great circle. This goes back to a previous post of mine (7/31, 1:55PM EDT) where I tried to describe what an infinite line that is directly above you looks like–in one direction, you’ll see the line approaching some point on the horizon; in the opposite direction, you’ll see the line approaching some point on the horizon directly opposite the previous point. The line will span a full 180[sup]o[/sup] of your field of vision–failure to realize this is, I think, a big contributor to the illusion of the moon not pointing at the sun.

So, with that imaginary line (stretching from horizon to opposite horizon (and remember that it is a straight line)) in mind, picture the sun at one of the two points on the horizon where the line meets. Picture the moon somewhere else on the line, say somewhere between the opposite point on the horizon and the point directly overhead you (I’m picking this case in particular because it seems to be the confusing case). What do you see? The moon will appear to point up and away from the sun. Remember, however, that to get an accurate picture of what the moon is pointing to, we have to follow along that imaginary line–the imaginary line comes up through the moon, passes through the point directly over head, then continues down to the location of the sun.

So the moon does point to the sun, it’s just a matter of keeping track of what straight, infinite lines through space look like to us from the earth. I hope that’s clear, but let me know if it ain’t.

33

warmgun, you’re right, you can use an arbitrarily short stick to do this, and it should work, but I imagine that you have to hold it pretty close to your face. Could you do me a favor tonight? Hold your T-square so that the line from the moon is in fact pointing at the sun. I mean hold it so that the edge is actually on the sun and actually on the moon. If the “perp” does not line up with the shadow-line of the moon, as you seem to be saying it wouldn’t, try to figure out about how much it’s off by.

I’m with you, Cabbage, but I think you can adequately visualize the infinte line by using the string experiment I described earlier.

34

Achernar, I also have a 60" straight edge which I will afix to my T-square which will hopefully yield more accurate results. I will also use the adjustable tri-angle to check ‘out-of-squareness’ (so to speak) if it exists at all. However, I am in Monterey, CA and there might be fog tonight. Regardless, I will check in asap.
Cabbage, I understand what you are saying and being an artist I am aware those types of opticle illusions. It makes sense. But I still have my first field experiment in my head.
As Bugs Bunny would say, " I’ll take another wack at it."
But this time I will be as diligent as possible.
I’ll check in soon…
Steve

35

Lotsa good answers here…just different ways of looking at the same phenomenon. Personally, I like Chronos answer…as well as some others. Maybe this will help, if I can make sense.

Go to Your Sky, and lets go back in time.

  1. For “date and time”, select Universal Time and type in "2001/07/27 22:00

  2. For “Observing Site”, type in 32° North Latitude and 96° West Longitude.

  3. For “Display Options”, select Ecliptic and Equator, Moon and Planets, and Show Stars Brighter Than Magnitude 0.0.

That’s all you need, so click on “Update”.

The star map you should get will show the sky as it was on July 27 at 1700. The red arc is the ecliptic and the white arc is the celestial equator. You can spot the Sun and Moon as well as some of the planets. Note that the bisection of the Moon’s terminator won’t intersect the Sun…it will fall below it. (south of it) That’s because if you project a straight line on the inside (or outside) of a 3 dimensional sphere, it will appear as the arc of a circle, and that’s why the straight edge experiment won’t work. Note, though, if you follow the ecliptic from the moon to the Sun, they will intersect.

I hope that makes sense. :slight_smile:

00

36

That’s a neat program, 00 Buckshot, but I don’t think it’ll help with this problem. The image of the moon they use, though it does show the phases, does not show the direction correctly. The light side always faces directly right or left. For an extreme example of the moon not pointing toward the sun in this program, check out JD 2452169.17375 at 47°N, 7°E.

Now, even if it did reproduce it faithfully, it wouldn’t help. There’s a difference between drawing a line on your flat computer screen and holding up a straight edge to the sky. The first one is not a projection of a straight line on the sphere; if that were the case, the equator and the ecliptic, which are straight lines projected on the sphere, would appear straight in the Your Sky program, which they don’t.

37

I believe that your problem can be solved by realizing that the Moon’s orbital plane is at a considerable angle to the ecliptic plane. Thus tha angle that you expected can be seen only twice a year, at the times that the Moon is at the point in its orbit where it crosses the ecliptic. Interesting that visualizing this takes some conceptual gymnastics, as many of you have been experiencing.
George

38

Oh my, is it First Quarter already? No? Well then what the heck. :slight_smile:

George, you’re right that the Moon’s orbital plane is not the same as the ecliptic. But the inclination angle is only five degrees. I believe that the phenomenon we have on our hands here is more than five degrees—otherwise we wouldn’t notice it. This inclination, I believe, does not contribute to the phenomenon at all, and even if the inclination were 0deg;, 45°, or 85° we would still see the same thing happening, even though the moon would be in a different position in the sky. Also, you state that the angle you expect will occur only twice a year. However, I contend that a waxing gibbous Moon will never, ever, ever, point down at sunset.

39

I’m a big fan of sky phenomena and I have to confess that I have never seen the angle of the Moon to the Sun to be anything other than perpendicular. I guess I just have good vision.