Gee, I hope nobody’s mad at me; I don’t see why they would be. Then again, I don’t see why anyone would be mad at Monty or chula either.
Monty’s post was a good point, and I’m sorry I didn’t address it more fully. When Terminus Est stated that 12h6m was “a bit more than 12 hours”, I misheard it as “quite a bit more than 12 hours”. I aimed to explain that the 6 minute difference was attributable to imperfections in the point-model concept of the Sun, not due to locales on the Equator getting any more or less than the average amount of sunlight throughout the year. That’s why I found the average.
According to chula, the average of the day-length in New York between Oct 07 and Apr 07 is 12h13m. I’m willing to bet that that value is within a couple of minutes of the average over the whole year. Now on Apr 07, New York’s day-length exceeds this value by 44m. On Apr 07, Quito exceeds its average by 0m. So I would guess that in general, Quito experiences no large variations in day-length (not more than a couple of minutes) whereas New York experiences quite a bit (at the very least 44 min).
Now it’s a good point that I picked a day close to the Equinox; Apr 07 is around 17 days after it. But that’s actually far enough away to get a good idea of what’s going on, because around the Equinox is when the day-length changes the most rapidly. If it hasn’t changed in the first 17 days, it’s not going to change much thereafter. I probably should have explained this fact earlier, though.
Okay, now I’ve made all these claims. Let me check it with some numbers. Here are the day-lengths for the equinoxes and solstices for the year 2002 (using Terminus Est’s link):
(Mar 21, Jun 21, Sep 21, Dec 21)
Quito: (12h6m, 12h6m, 12h6m, 12h8m)
New York: (12h10m, 15h5m, 12h13m, 9h15m)
That puzzled me for a moment. (Just a small hijack, here, I think.) The parallax effect mentioned above shows that the daylightlight length at the equator should be a bit shorter when the Earth is closer to the Sun. The Earth is closer to the Sun at the December solstice than the June solstice; so I expected the daylight period to be longer in Quito at the June solstice.
Had to look that up. Quito is actually at 0 degrees 9 minutes south latitude. Evidently the small decrease in distance at its winter solstice (in June) is overwhelmed by the small relative tilt toward the sun at its summer solstice (in December).
To the tune of two minutes, 1/720th of a day. Yeah, I know, whatever. I worry about stuff like that.
That’s a good point SCSimmons, but I should also point out that the precision on these figures is not that great. Since the times of sunrise and sunset are (I’m assuming) rounded to the nearest minute, they could be off by up to 30 sec. Since the day-lengths are the difference of two such times, they could be off by 60 sec. So the day-lengths of the solstices could in fact be within a couple of seconds - the difference is 2 minutes, give or take 2 minutes.
Horizontal parallax for the sun (radius of earth / distance to sun) would introduce a negative correction of about 30 seconds doubled so we have:
refraction: +2.26 min
semidiameter: +1 min
parallax : - 0.5 min
Total Added to sunlight time: 5.5 minutes
variation is the sun’s HP is negligible and would account for a fraction of a second at the most.
HP for the moon is quite a bit larger, almost a degree, and the variation is also much larger
The Earth being closer to the Sun in December means that it’s also moving quicker in its orbit, which will slightly increase the day length. This might be another conributing factor in Quito’s longer December solstice day.
Achernar, judging from dates near the Solstice for Quito (so rounding errors will cancel somewhat), it looks like 12h 8.5m . The long days seem to be centered on solstice, not perihelion, so the (small) latitude effect is apparently the dominant one, here.
One wonders, though, why the June Solstice doesn’t have a correspondingly shorter day.
Don’t foprget the Eq.oT. At any rate, for precise calculations one should go by geocentric coordinates and forget about refraction, semidiameter etc which just tend to cloud the issue. Then location and eq.o.T become more noticeable
