Sunrise and sunset times when compared between the vernal and autumnal equinoxes

Question

Why is it that sunrise and sunset occur at different times when compared between the vernal and autumnal equinoxes?

Consider Pontianak, Indonesia, 0 deg N, for simplicity. Timeanddate.com gives sunrise and sunset times 5:46 AM, 5:53 PM at 3/21 (vernal equinox) and 5:31 AM, 5:38 PM at 9/23 (autumnal equinox). Given the symmetry of the problem, I would expect that sunrise would occur at the exact same time during the two events. Same for sunset. What is the reason for the 15 minute difference between the time sunrise occurs on the vernal equinox and the time sunrise occurs on the autumnal equinox?

Answer 1

Instead of looking at sunset/sunrise times, it's easier to look at noon, which is right in the middle between sunset and sunrise: 11:50am in March and 11:35am in September.

The fact that noon doesn't always happen at the same time of the day (when measured with steady clocks) is expressed by the Equation of Time. Two factors play a role here:

• The plane in which Earth rotates around the Sun (the ecliptic) is tilted with respect to the one it rotates around its axis (the equator)
• The Earth's orbit around the Sun is an ellipse, not a circle, and it moves faster than average in its orbit from (roughly) October up to and including March, and slower in the other months

Answer 2

Given the symmetry of the problem...

That's the key to the answer. While we like to imagine motion in the system as having some symmetries, it's all a bit wobbly.

As @Glorfindel points out the particular asymmetry is connected to the direction that Earth's tilted axis points relative to the shape of Earth's elliptical orbit.

If the Earth's orbit were a circle and you were using local sidereal time, then maybe you would have your symmetry.

But it's not, it's an ellipse.

But if Earth reached it's perihelion on January 1 or thereabouts every year, you might still have most of the symmetry you're expecting.

But it doesn't. It tends to wobble around in the first week of January by several days.

So since these two directions don't quite match up (and there's no reason to expect them to either) the imperfect symmetry yields times that are close, but don't quite match.