The earth is tilted at 23.5 degrees which means that in the solstices the north/south hemisphere is more radiated by the sun and the tilt to the sun is maximum.

In the equinoxes, the tilt is not towards the sun but to the "side" so the equator has direct exposure to the sun. As a result, we have 12 hours on day and night on the equinoxes.

What I don't understand is why is it only for these 2 days (equinoxes) that this happens. In the next day(s) I think the equator should have the same exposure to the sun as the equinoxes. Has the tilt changed so much?
I couldn't find any animation that shows the exact path of the tilt throughout the year.
Can someone please help me understand this?

  • $\begingroup$ @BillDOe: I guess you could think of the solstices you mean equinoxes right? $\endgroup$ – Jim May 5 '19 at 9:24
  • $\begingroup$ Yeah, I meant equinoxes. (Hate when I do that.) I deleted it and added new corrected comment. My apologies. $\endgroup$ – BillDOe May 5 '19 at 23:37
  • $\begingroup$ I guess you could think of the equinoxes as that moment when the plane of the poles is exactly at right angles to a line drawn from the Sun to the Earth. "Plane of the poles" being a plane defined by the poles of the Earth and a line drawn through the Earth perpendicular to the Plane of the Ecliptic. $\endgroup$ – BillDOe May 5 '19 at 23:38

First, don't think that equinox is when the day and night are the same. It is the moment, when the sun's declination is zero, or when the sun is directly over the equator. Let me explain ...

Earth is always rotating around the Sun, so declination of the Sun is always changing (except at the solstices when it stops for a minute and goes in the other way). As @JohnHolst has pointed out, Sun's declination is changing for 1 angular minute per hour (mark 1 mm on the 3 m of distance – this is nearly 1 angular minute). An equinox doesn't last for a day. It lasts for a moment when the Earth is totally aligned and straight with the sun: (this figure shows the geocentric model, which is easier to represent and imagine)

declination = 0

Also, Wikipedia says:

An equinox is commonly regarded as the instant of time when the plane (extended indefinitely in all directions) of Earth's equator passes through the center of the Sun. This occurs twice each year, around 20 March and 23 September. In other words, it is the moment at which the center of the visible Sun is directly above the equator.

P. S.

You can look this simulation.

  • 3
    $\begingroup$ @Jim. You are correct that the change from one day to another is not that large. But the definition of the equinox is not based on the length of daylight, or the sun being approximately over the equator. The definition is the Sun's declination is 0 degrees 0 arcminutes 0 arcseconds. At the time of the March equinox, the declination changes by approximately 1 arcminute per hour (or 1 arcsecond per minute of time). Thus, the time of the equinox can be stated very precisely. $\endgroup$ – JohnHoltz May 5 '19 at 1:10
  • 1
    $\begingroup$ @User123: when the Earth is totally aligned and straight. but the earth is always tilted. $\endgroup$ – Jim May 5 '19 at 9:25
  • 1
    $\begingroup$ @Jim. You're correct that equinox originally meant "equal day and night". If this definition is used, and if the length of day is sunrise to sunset, then the date of the equinox would be different (by a few days) depend on your latitude, and the March equinox would occur before Mar 20. At some time in the past, the definition was changed to when the Sun is at 0 declination. (Declination is one of 2 celestial coordinates, and 0 declination is directly over the Earth's equator.) $\endgroup$ – JohnHoltz May 5 '19 at 21:21
  • 1
    $\begingroup$ +1 for 'equinoxes last for a moment'. They are the instantaneous point when the change occurs $\endgroup$ – Dave Gremlin May 6 '19 at 9:12
  • 2
    $\begingroup$ @Jim You can measure the Sun's declination with a well-built sundial. You can get a more accurate measurement if you have a telescope, and we've had those for over 400 years. $\endgroup$ – PM 2Ring May 10 '19 at 8:16

The easiest graph to consider is the graph of day-length. If the day-length varies from 8 to 16 hours, there can only be two equinoxes: enter image description here

At the time of equinox, the day length changes at the fastest rate, by a few minutes in temperate regions, compared to the solstices, which change by a few seconds every day. So, the equinox doesn't exist perfectly, the night time will be 11.57, 12.01, 12.04... It can be off by a minute, but the next day, it is changed by a few minutes... march 20th: https://www.timeanddate.com/sun/france/paris?month=3&year=2019

Don't try to understand with words, try to understand with images and objects in your hands: This image shows you. On the equinox, does the north pole and the south pole face away from the sun? https://youtu.be/tX3Y5bzNDiU?t=35

Don't over think it, rotations are confusing, just have a basic idea that the north pole becomes parallel with the sun on the equinox, and points away from the sun in winter, and towards the sun in summer.

Perhaps you can pick up an apple or a fruit and put it near a lightbulb / lamp and do the movement mechanics for yourself. try youtube for "seasons earth animation" https://youtu.be/WLRA87TKXLM?t=5

  • $\begingroup$ I have seen that video. I don't get how in 1:17 we see that the sun is 90 degrees of the equator but as it shows to move, I don't understand why for the next 24 hours there is such a difference in the angle $\endgroup$ – Jim May 4 '19 at 16:50
  • $\begingroup$ Can you elaborate on the graph a bit? $\endgroup$ – Jim May 4 '19 at 17:13
  • 1
    $\begingroup$ Ok I added some info. you are right by the way, the equinox changes changes slowly and it can be stuck in between two days, for example, one night will last 11.58 hours, and the next day can last 12.02 hours. with the day being about 12 hours but almost never to the very second, and it can be either the day or the night which is closest to being 12.00 hours, it depends where you are on the world when the pole is half way between facing towards and away from the sun. $\endgroup$ – predatflaps May 4 '19 at 19:10
  • $\begingroup$ Thanks for the update. The sun is a sphere so the light rays emitted should be the same in all directions. Hence if the tilt of the earth on the equinox moves the next day since it won't move towards the sun, the rays should be 90 degrees to the equator as well after the "official" equinox. I'll see the videos you pasted again in slow motion $\endgroup$ – Jim May 4 '19 at 19:18
  • $\begingroup$ E.g in the video youtube.com/watch?v=tX3Y5bzNDiU&feature=youtu.be&t=35 what is the difference of the angle of the sun to the equator from 0:35-0:36? It seemds to me it should be 90 degrees till 0:36 even though the earth has moved a lot from 0:35. This is the part that is confusing to me $\endgroup$ – Jim May 4 '19 at 19:24

Two further points to make are that:

(1) because of the way we define sunrise and sunset as being when the entire Sun is below the horizon not just the centre of the Sun.

(2) When the Sun is close to the horizon the Sun’s ray are refracted (bent) by the Earth’s atmosphere so that the Sun appears to be slightly higher in the sky that it would be if the Earth had no atmosphere. For more details see https://en.wikipedia.org/wiki/Atmospheric_refraction. This means that the Sun appears to be above the horizon when it is actually just below it.

Because of these facts there are always more than 12 hours of daylight and thus slightly less than 12 hours of night on the dates of the equinox.

The dates when there are almost exactly 12 hours between sunrise and sunset are know as the equiluxes and vary according to latitude.

For more details see my recent blog post of this topic. (Apologies if this seems like self promotion, it just pulls it all the information together about equinoxes and equiluxes into one single place.)


  • $\begingroup$ Can you rephrase/elaborate the ` Due to the refraction of the near horizontal rays of the Sun near the horizon, `? I don't understand what this means $\endgroup$ – Jim May 16 '19 at 16:54
  • $\begingroup$ @jim I have done so. I hope this helps $\endgroup$ – Science Geek May 17 '19 at 11:46
  • $\begingroup$ amazing facts !! thanks !!!!!!!!!!! $\endgroup$ – Fattie Jun 21 '20 at 19:48
  • $\begingroup$ @ScienceGeek sent a bounty, thanks :) $\endgroup$ – Fattie Jun 25 '20 at 20:15

The earth is smoothly and constantly moving around the sun. At the moment of the equinox, a line from the center of the sun to the center of the earth passes through the equator. That only lasts an instant. Before that instant, the line is on one side of the tilted equator, and after that, it's on the other side. The day labeled "the equinox" is just the day on which that instant happens.

In any case, a single day is enough to make a measurable difference in any definition of "exposure to the sun". For instance, depending on the latitude, the length of time from sunrise to sunset can change by several minutes from one day to the next, near the equinoxes.

  • $\begingroup$ a line from the center of the sun to the center of the earth passes through the equator. this sentence is what is confusing to me as in my mind this line should exist always. I suspect that the meaning is different than an arbitrary straight line from the center one sphere (sun) to another (earth) $\endgroup$ – Jim Jun 26 '20 at 7:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.