So, the Old Farmer’s Almanac, which I’ve known and more or less trusted over the last 25 years, informed me that autumn would begin in the Northern Hemisphere early on the morning of 23 Sept. “Fine,” figure I, “that means there will be exactly 12 hours of daylight that day.”

But I was wrong. The sun crossed the celestial equator on the 23rd, but the period of daylight was still a few minutes over 12 hours and didn’t hit that mark until about now.

I’d always understood that when the sun was directly above the equator, day and night were always of equal length. Now I’m confused. The sun actually has to go south a couple of degrees before that happens.

So what’s the deal? Is precession a factor in all of this or have my grade-school astronomy books led me astray again?

How are you measuring sunset?

I could be totally wrong, but isn’t daylight always going to be a bit longer than night at that period, due to atmospheric lensing, scattering, etc.?

I could be totally wrong, but isn’t daylight always going to be a bit longer than night at that period, due to atmospheric lensing, scattering, etc.?

I could be totally wrong, but isn’t daylight always going to be a bit longer than night at that period, due to atmospheric lensing, scattering, etc.?

Verturing a WAG…

The sun would only be exactly over the equator for an instant as the change in the earth’s angle is constant. Because of this the only spot where the day would be exactly 12 hours is on that exact spot. Any spot east or west of that spot would have a day that is slightly shorter or longer than 12 hours. I could explain this better with a java applet like NASA uses (if I knew java (and if I worked for NASA)).

Procession might figure into this somehow but I think the numbers would be much smaller. Anything divided by 21,000 is going to be pretty small. only possibly related

Jiminy jillickers, that’s weird. I posted, nothing happened, then I went to GQ, opened the thread, saw Padeye’s response had been posted but not mine, then posted again…

Must be the equinox.

Basically, it’s because while the sun’s center is on the horizon eaxctly 12 hours apart at the equinox, the time of sunrise and sunset is calculated when the sun’s edge is on the horizon. Refraction accounts for the additional time.

No, this has nothing to do with precession.

If there were no atmosphere and the sun were a point, it would be above the horizon for exactly 12 of the 24 hours surrounding the equinox, everywhere on earth.

The sun is not a point. It is a disk half a degree in diameter. It takes time to rise and set. Sunrise is defined as when the leading edge of the sun rises above the horizon, and sunset is when the trailing edge sets. This adds a few minutes of “daylight”.

In addition, the lofting effect of atmospheric refraction keeps the sun above the horizon even longer. The atmosphere acts like a giant, weak fish-eyed lens.

The combination of these two effects makes daylight several minutes longer than 12 hours at the equator on the equinoxes, and adds even more day length as you move closer to the poles.

Oh.