[damocles]

RD is correct I got it the wrong way around. But I think the point still stands. It may in fact appear to speed up at sunrise with reduction of the angle and slow down at sunset with the increase of the angle.

Sorry Anton, but the effect of atmospheric refraction is to slow the optical angular velocity at either end of the day.

[AntonMaeso (OP)]

No need to apologize.

Could you explain in more detail please. It has been a long time since I studied physics at school.

[damocles]

From AntonMaeso:

No need to apologize.

Could you explain in more detail please. It has been a long time since I studied physics at school.

Consider the situation in terms of the following figures relating to an equatorial location on the prime meridian at the equinox:

because the sun's rays are bent towards the normal by the atmosphere, we have optical sunrise at 5.55 a.m. Geometric sunrise is at 6.00 a.m.

When the sun's optical elevation is at 15° the time is 6.59 a.m, but for geometric elevation of 15° the time is 7.00 a.m.

For the next few hours the optical and geometric locations of the sun are close to agreement.

But when it comes close to sunset the geometric elevation of 15° at 5.00 p.m. comes a minute before the optical elevation of 15° at 5.01 p.m.

And geometric sunset occurs at 6.00 p.m. a full 5 minutes before optical sunset at 6.05 p.m.

The detail of these numbers may be a little bit out, but the fact of more than 12 hours of daylight at the equinox can easily be checked on any almanac -- sunrise and sunset times are based on the optical position of the sun, while the equinox date is determined by the geometric position of the sun.

Geometrically, the angular velocity of the sun is a constant 15°/hour along its path. Optically the sun moves its first (or last) 15° in 64 minutes, that is, its apparent motion would be slower near the horizon.