File talk:Hours of daylight vs latitude vs day of year with tropical and polar circles.svg

The tropics and the polar circles are incorrectly labeled. --Ianuslorenzo (talk) 01:00, 16 January 2020 (UTC)

Thanks, I swapped “Can.” and “Cap.” ◄ SebastianHelm (talk) 12:09, 16 January 2020 (UTC)

Good. Thanks. What about the polar circles? --Kupirijo (talk) 09:15, 1 March 2020 (UTC)

Thank you for checking. I swapped those, too, now. ◀ SebastianHelm (talk) 20:44, 10 March 2020 (UTC)

Why are the borders to constant night and constant day labelled with ‘1 hour’ and ‘23 hours’, respectively?

The curve for constant day says 23 hours, but actually the curve shows 24 hours of dayligt. Likewise the curve for constant night says 1 hr, but it actually is zero hours of daylight. (Furhermore the "equal daylenghth"-latitude in not exactly at the equator, but a little south of it, due to the excenticity of the earth's orbit). Episcophagus (talk) 20:19, 21 March 2024 (UTC)

Let's separate the two questions: The first question - before the parenthesis - is clearly worded and roughly fits the original headline of this section “23 hrs?”. So let's keep this section for this question, and allow me to clarify the headline to “Why are the borders to constant night and constant day labelled with ‘1 hour’ and ‘23 hours’, respectively?”. I see the same problem. The best person to answer this would be Jalanpalmer, the creator of the original diagram File:Hours of daylight vs latitude vs day of year.svg, from which this (as well as File:Hours of daylight vs latitude vs day of year cmglee.svg by Cmglee) is derived. Unfortunately, Jalanpalmer is not around anymore. So maybe we can figure it out ourselves by studying using the article w:Sunrise equation referred to in the original diagram's description. I haven't been able to get the dependence of the day from that, though. An explanation might be that the two values are very close together. If we had a formula or a diagram for the day length for a given day as a function of latitude, we could test that. ◅ Sebastian Helm 🗨 10:13, 22 March 2024 (UTC)

Obviously the 12 hours line is "correct". The length of the day along the equator is somewhere close to 12h 7m (12h 6m 40s) all year long. But the 23 (and 1) hour lines should be labeled 24 (and zero). The "24 hours (constant day)" and "0 hours (constant night)" areas ought, in my opinion, to be relabeled "Constant day" and "Constant night" respectively - as the 24 hours and 0 hours lines already indicates the length of the day. Episcophagus (talk) 15:46, 22 March 2024 (UTC)

@Episcophagus: I've asked on http://en.wikipedia.org/w/index.php?title=Wikipedia:Reference_desk/Science&diff=prev&oldid=1215149171 cmɢʟee ⋅τaʟκ 12:14, 23 March 2024 (UTC)

I replied above and below, and between above and below.... Episcophagus (talk) 16:59, 23 March 2024 (UTC)

‘equal daylenghth’-latitude should be south of the equator

This statement was made parenthetically by Episcophagus in the previous section. I don't understand it, since there is nothing in the picture labelled as “‘equal daylenghth’-latitude”. But this statement seems to be unrelated to the former question, so I'm moving it here in its own section. ◅ Sebastian Helm 🗨 10:13, 22 March 2024 (UTC)

Sorry! I was tired when I wrote it and managed to confuse it with the effect of the equation of time. I was wrong! Of course! Episcophagus (talk) 15:37, 22 March 2024 (UTC)

Well, actually i was not completely wrong, as the rate of change in the equation of time affects the daylength (the steepness of the ecliptic at the equinoxes make the sun to move slower in rectascension and thus the daylength is shorter than at the solstices), but now we are down to seconds and fractions thereof. Episcophagus (talk) 16:22, 22 March 2024 (UTC)

@Episcophagus: I've asked on http://en.wikipedia.org/w/index.php?title=Wikipedia:Reference_desk/Science&diff=prev&oldid=1215149171 cmɢʟee ⋅τaʟκ 12:13, 23 March 2024 (UTC)

Why does the diagram show areas around both poles illuminated at the equinoxes?

If we conclude above that we should change the label of the borders to constant night and constant day to ‘0 hours’ and ‘24 hours’, respectively, then we are faced with another question: Why don't these curves convene at the equinoxes at both poles? Or asked differently: How can it be that at the equinoxes there are whole areas in sunlight at both poles, as shown by the blue lines running over white areas on either end? My guess is that this is a consequence of atmospheric refraction, as described at w:Sunrise equation#Generalized equation. If that is so, we should mention it in the description. ◅ Sebastian Helm 🗨 10:13, 22 March 2024 (UTC)

It is an effect of the refraction (34 minutes of arc) and that the sun is seen as a disc (radius 16 minutes of arc) and not a point. The sun thus illuminates some 50 minutes of arc more than 180 degrees on either side. 50 minutes of arc at the equator (where the sun rises/sets perpendicular to the horizon) means 3m 20s (the earth rotates 1 arcminute in four seconds - divide 360 degrees with 24 hours - and add this amount both in the morning and in the evening) - the further north or south you go, the rising-angle slants more and more and thus the time for the sun to rise or set increases - at 60 degrees north or south it has doubled (pure trigonometry!) and thereafter the increase is BIG - at the poles the sun rise and set once a year, but it rises and sets before respectively after the equinoxes due to the 50 arcminutes that results from refraction and the sun's radius. The sunrise and sunset occurs when the upper edge of the sun's disc breaks the "virtual" (due to refraction) horizon (the "real" horizon is the tangential plane of the earth's surface). I hope I disconfused you a little! Episcophagus (talk) 16:00, 22 March 2024 (UTC)