Talk:Equinox

Incorrect to compare dates observed in Rome to proleptic Julian calendar

In this edit Eric Kvaalen replaces a statement that Julius Caesar set the date of the spring equinox to March 25 with arguments about when it occurred astronomically. These arguments are invalid and inapplicable because Julius Caesar would have made decisions on the basis of how the Julian calendar was actually observed in Rome during his lifetime, or if making decisions before the calendar went in to force, how it was planned to be observed. But after his death, the rule he made about leap years was not correctly observed, and the surviving documents are insufficient for historians to determine with certainty the conversion between the observed Julian calendar and the Proleptic Julian calendar. The latter is used by astronomers when presenting dates before AD 8. See Julian calendar#Leap year error for more information. Jc3s5h (talk) 16:33, 27 July 2022 (UTC) (Link fixed 18:20 5 December 2022 UTC by Jc3s5h.)

@Jc3s5h: Once again you have reverted a whole edit just because of one little thing. Yes, there is some uncertainty about whether the Kalends of January of 709 AUC was really January 1 in the proleptic Julian calendar. It might have been January 2 or December 31 in the proleptic Julian calendar (a difference of one day from January 1). But the way you put it (by reverting) it says unambiguously that Caesar set March 25 as the equinox, with no reference at all, let alone a reliable reference. That would only be true if his calendar in 45 BC was two or three days different from the proleptic Julian calendar, not just one day, as I demonstrated in my footnote with a simple calculation and references for the inputs. So our article is wrong after your reversion. Let's restore my edit but reword that sentence a little bit. Eric Kvaalen (talk) 14:34, 2 August 2022 (UTC)

If you read the article you will see the part of your comment that says "with no reference at all, let alone a reliable reference" is not true. There is a reference to page 135 of Blackburn and Holford-Strevens. That source states, in a section devoted to 25 March,

In the Roman calendar as reformed by Caesar this was counted as the vernal equinox, being the normal date in the years following the reform...It was also the date of the equinox in the rules devised by the church at Rome for finding Easter, and in the somewhat different system used by Celtic Christians

When the equinox actually occurred is a separate issue from what date Julius Caesar set as the equinox date.

As for when it actually occurred, the gist of your calculations seems to be to find a way to smooth out the year-to-year variations due to leap years and other less important factors. I am not aware of a reliable source that has put forth a method for doing this smoothing. Calculations by a Wikipedia editor are unacceptable original research. You would have to find a reliable source that sets out a generally accepted method for smoothing out the date of the equinox, either in general, or for years near 45 BC. Jc3s5h (talk) 15:18, 2 August 2022 (UTC)

@Jc3s5h: I don't see any reference to "Blackburn and Holford-Strevens". And why does your quote say that March 25 was the date in the rules devised by the church at Rome? The Computus uses March 21. Eric Kvaalen (talk) 18:22, 24 August 2022 (UTC)

The work by Blackburn and Holford-Strevens is footnote number 11 in the article. Julius Caesar, in conjunction with whoever advised him, set the March Equinox on March 25. That was 45 BC. Among Christians, the computus commemorates the death and resurrection of Jesus, which didn't happen until about 77 years later in about AD 33. A lifetime or more passed before Christians decided to commemorate Easter. The rules for the computus differed between the Church in Rome and the Church in the East until Dionysius Exiguus extended a set of Easter tables that the Church in Rome had received from the Church in Alexandria; the first year in Dionysius's table was 532. Before that, Rome used March 25 while Alexandria used March 21. Dionysius used March 21. The computus of Dionysius, after a few hundred years, was used throughout Christianity until the Gregorian reform of the calendar in 1582. Jc3s5h (talk) 19:35, 24 August 2022 (UTC)

@Jc3s5h: I don't understand what you (or Blackburn and Holford-Strevens) mean by saying that Caesar set the equinox to March 25. If in his calendar it occurred on March 23, then in what sense did he set it to March 25? I don't think your reference is reliable. I have now used Stellarium to check that the equinox occurred on March 21, AD 3, at 9:26 AM in the proleptic Gregorian calendar. That is March 23 in the Julian calendar. (Stallarium doesn't allow me to try years before 1 BC.) Eric Kvaalen (talk) 09:06, 11 December 2022 (UTC)

I suggest using JPL Horizons On-Line Ephemeris System which does cover BC years as early as 9999 BC. It does agree that in AD 3, Proleptic Julian Calendar, the equinox occurred on March 23 between 9 and 12 in the morning, UT. I set the hour step size to 3 hours.

If you look at our Julian calendar article (which has many reliable sources), you will see that Julius ordered the year 46 BC to be 455 days long. Roman years were normally around 355 days if a common year, and 378 days if an intercalary month was added. But 46 BC had two extraordinary months to make it 67 days longer than if it were a typical year containing an intercalary month. So we can certainly say Julius "set" the calendar to align to whatever he wanted to align it to. The note in our article claims "it is not known why he decided that 67 was the correct number of days to add, nor whether he intended to align the calendar to a specific astronomical event such as the winter solstice."

To determine what date it did align to, it would be necessary to find the date for several years surrounding 45 BC, because the leap years cause the date to shift around. Even after doing that, we would only have the date in the proleptic Julian calendar. There is some uncertainty, of a few days, about how the calendar was observed in Rome before AD 8. And if it turns out to be a close matter, we have to allow for the fact that the time in Rome is about 49 minutes earlier than Universal Time. Jc3s5h (talk) 16:43, 11 December 2022 (UTC)

I looked up the values in the Horizons System at one day intervals around March 23, and interpolated to find the date and time of the equinox. Then I averaged the years 50 BC to 39 BC inclusive (12 years). The average Universal Time of the equinox was March 23, Julian proleptic calendar, at 9:13 AM (All times were truncated to the minutes.) This would be about 8:24 AM Rome time. Jc3s5h (talk) 20:17, 11 December 2022 (UTC)

The statement in question was added in the form "When Julius Cesar established its calendar, he fixed the Spring equinox on March 25" by 95.237.151.2 (talk · contribs) on 7 August 2011 in this edit. No sources were cited in that edit.Jc3s5h (talk) 16:18, 12 December 2022 (UTC)

Sst

name the line of latitude where equinox is experienced 102.222.234.44 (talk) 12:57, 9 January 2023 (UTC)

All of them. If you mean where the Sun is overhead, it would be the equator. --Lasunncty (talk) 02:21, 11 January 2023 (UTC)

A question for you all: What is a spring equinox?

Answer the question with the date of this equinox and also give another name of this equinox. (Hint: The another name of spring equinox starts form letter V.) 2405:201:3009:4DDB:15EF:22AB:EEC1:CA98 (talk) 15:05, 18 March 2023 (UTC)

The spring equinox in the northern hemisphere occurs on or near 21 March. The spring equinox in the southern hemisphere occurs on or near 21 September. Jc3s5h (talk) 15:41, 18 March 2023 (UTC)

The spring equinox in the Southern hemisphere occurs most often on September 22 or 23. (Yes, the March date is mostly 21 or 22.) It depends, of course, on your time zone.

The other name for “spring” is taken from the Latin word for “spring”. TomS TDotO (talk) 16:58, 18 March 2023 (UTC)

Correction: The March date is usually 20. TomS TDotO (talk) 14:50, 19 March 2023 (UTC)

According to NASA/JPL DE422 ephemerides (and DE405 also), the exact time this year, when Sun is at 180°, or when crossing the line of Earth equator in Y direction, is 2023-03-21 05:03:00 (or 05:02:59) TAI... There may be spread by 8 minutes of light-time distance from Earth and by 37 seconds of difference between TAI and UTC... Wondering, how USNO came to earlier time by more than 7 hours ? 85.70.204.171 (talk) 19:19, 19 March 2023 (UTC)

The JPL DP422 and DE405 ephemerides do not provide this kind of information directly. Software must be used to extract the information. Some software is well-written, and some is not.

Looking at other sources, I see that the USNO at this link gives 2023 March 20 21:24 Universal Time. A computer program provided by USNO but no longer available is the Multiyear Computer Interactive Almanac (MICA), under Phenomena, gives 23 March 20 21:26 Terrestrial Time. MICA, when ordered to calculate apparent geocentric positions, true equinox and ecliptic of date, Terrestrial Time, for the Sun, gives 2023 Mar 20 21:25:36.0.

According to the most recent IMCCE Newsletter issued by the French Bureau des Longitudes the computed moment of the spring equinox for 2023 is 20 March 21h 24m 27.74s UTC. AstroLynx (talk) 20:11, 19 March 2023 (UTC)

Using JPL Horizons system at https://ssd.jpl.nasa.gov/horizons/app.html , setting Ephemeris type "Vector table", Target body "Sun", Coordinate center 0°N,0°E,altitude 0 (it's a pity there cannot be specified Earth center), time 2023-03-20 to 2023-03-22 in 1 hour intervals also gives Y component traversing from negative to positive after 2023-03-21 05:00:00 and before 2023-03-21 06:00:00 ... It's not a problem of "not-well-written" software... Maybe their coordinate system is not "equatorial" any more ? Did the Earth equator inclined ?? 85.70.204.171 (talk) 02:15, 21 March 2023 (UTC)

Using vector table results in x, y, z coordinates. The northward equinox is not defined in those terms, it is defined when the apparent ecliptic longitude of the Sun, observed from the center of the Earth, passes through 0 degrees. If you choose "observer table" rather than "vector table" you can choose the location of the observer as geocentric, and the UT time of the equinox is on 2023 Mar 20 between 21:00 and 22:00 UT. Jc3s5h (talk) 02:55, 21 March 2023 (UTC)

Then the text in "Equinox" article heading is wrong? Because there is nothing about ecliptic at first... At present, the article heading reads: "A solar equinox is a moment in time when the Sun crosses the Earth's equator, which is to say, appears directly above the equator, rather than north or south of the equator. ... More precisely, an equinox is traditionally defined as the time when the plane of Earth's equator passes through the geometric center of the Sun's disk ". (emphasis mine)

"Ecliptic" is a mere theoretical concept, or an average of Earth position, ignoring the fact, that the Earth position on the orbit is not "smooth", but vibrates primarily due to Moon counterbalance...

Then the article heading reads: "In modern times[when?], since the Moon (and to a lesser extent the planets) causes Earth's orbit to vary slightly from a perfect ellipse, the equinox is officially defined by the Sun's more regular ecliptic longitude rather than by its declination." (emphasis mine)

The article should mention, who and when redefined the "Equinox" to use average Ecliptic instead of true Earth and Sun positions - possibly in "History" section, and maybe even list the times according to true body positions beside the "ecliptic" average ones...? 85.70.204.171 (talk) 03:35, 21 March 2023 (UTC)

I don't know when the present definition was first adopted. It may have been long enough ago that it wasn't possible to measure the difference between when the ecliptic longitude was 0 and when the declination was zero. We shouldn't be listing the time of the equinox according to some obsolete definition. Jc3s5h (talk) 04:36, 21 March 2023 (UTC)

I already added a sentence to article heading, but reverted it myself... The problem is, that "Equatorial" coordinates in JPL Ephemerides drift from true Equator by precession of equinoxes, approximately 20 minutes each year makes one year difference in approximately 26000 years... 85.70.204.171 (talk) 08:19, 21 March 2023 (UTC)

The current definition (apparent geocentric ecliptic longitude of the centre of the Sun = 0° or 180°) has been the standard definition since Ptolemy's time and there is no pressing reason to change this. When trying to compute its occurrence with online software such as JPL Horizons set the Observer Location to "Geocentric [code: 500]" and in Table Settings only select boxes 30 & 31.

This results in:

 Date__(UT)__HR:MN           TDB-UT     ObsEcLon    ObsEcLat
  2023-Mar-20 21:24        69.185594  359.9996820  -0.0001516
  2023-Mar-20 21:25        69.185594    0.0003723  -0.0001515
  2023-Mar-20 21:26        69.185594    0.0010627  -0.0001515

from which the time of the equinox can be determined as occurring at 21h 24m 27.64s UT. AstroLynx (talk) 10:55, 21 March 2023 (UTC)

I think the question is pretty well answered, but I was able to reproduce 85.70.204.171's result of the y coordinate changing from negative to positive on 21 March between 05:00 and 06:00 by using the default table settings, which results in using the ICRF reference frame, which is essentially where the equator and ecliptic were on 1 January 2000. As far as I can tell, Horizons does not offer the ability to order Horizons to use the equator and equinox of date when using a vector table. Jc3s5h (talk) 15:14, 21 March 2023 (UTC)

Illumination of Earth by the Sun at the equinox

The figure seems to be false. 2A01:CB04:4D7:1B00:59DA:7FD9:FF11:C77C (talk) 07:05, 13 September 2023 (UTC)

In what way? Lithopsian (talk) 14:17, 13 September 2023 (UTC)

The figure suggests that the sun is at the Zenith all over the equator (it is th cqse only at one point)

The earth is inclined 2A01:CB04:4D7:1B00:A1DD:6648:D396:FBD3 (talk) 06:28, 24 September 2023 (UTC)

I don't think your statement about the equator makes much sense, but I think you want an image more like the one on this page? See if you can find one that is free-to-use or at least creative commons, and we can substitute it. Lithopsian (talk) 10:02, 24 September 2023 (UTC)

Thank you for your answer

The link's image is more better

I like it !! 2A01:CB04:4D7:1B00:C8DA:68AD:217C:628 (talk) 07:12, 25 September 2023 (UTC)

Is the spring equinox at 3:07 UT or 3:06 UT?

According to this (https://www.imcce.fr/newsletter/docs/Equinoxe_printemps_1583_2999.pdf) article, the 2024 March (northern hemisphere spring) equinox happens at 20/03/2024 3:06 UT, but according to another page, (http://www.astropixels.com/ephemeris/soleq2001.html), the spring equinox begins at 3:07 UT, note that both articles are sited as sources for the equinox table in WP's article. So I wanted to know which source is correct on this. Thank you! :) Woomerrrr (talk) 15:29, 1 January 2024 (UTC)

The US Naval Observatory site agrees with www.imcce.fr the time is 03:06 UT. Jc3s5h (talk) 17:15, 1 January 2024 (UTC)