Talk:Twilight/Archive 1

This is an archive of past discussions about Twilight. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 1

Archive 2

Archive 3

Additions

I would like to see the addition of calculating all twilights (civil/nautical/astronomical) by hand given lat/long/altitude. I am not familiar enough with LaTex to do this myself at this point. —Preceding unsigned comment added by 82.127.87.189 (talk • contribs)

I am not sure why you would like to calculate this yourself. There are lots of astronomical and planetarium software that works out the timing automatically depending on location. I know of Astronomy Lab 2, Cybersky and Orbitron that generate this information. There are also a number of astronomical websites that can tell you this information eg. Heavens Above. - Shiftchange 00:52, 16 February 2006 (UTC)

There is one such calculator here [1]. Slowmover 19:36, 11 April 2006 (UTC)

Whether you can see a reason why someone would want to calculate it isn't relevant to whether or not it's an appropriate addition to the article. Given that up until such programs and websites became easily available, calculating the time for twilight was the only way to determine it. And if someone is writing a program that includes twilight calculations, an encyclopedia article on twilight is a reasonable place to expect to find the formula for doing so. CruiserBob (talk) 18:54, 27 November 2007 (UTC)

It looks like the Civil Twilight page material has been merged with Twilight, but the page still exists and the diambiguation page does not list it. I don't know how to fix this.--Fitzaubrey 04:18, 13 July 2006 (UTC)

I changed Civil twilight to be a redirect to this page. --Usgnus 05:00, 13 July 2006 (UTC)

Merge

Should this article be merged with dawn?

Oppose. Twilight and dawn are two different things, though related (provide links). However, I am for merging twilight with dusk. --HereToHelp (talk) 13:38, 27 November 2005 (UTC)

Oppose. Dusk and dawn are the times that evening twilight begins and morning twilight ends with. I would oppose twilight being merged into either dusk or dawn. Merging both of these topics into twilight would be more appropriate. - Shiftchange 10:07, 29 November 2005 (UTC)

Oppose. I agree with Shiftchange. I have just used Wikipedia to look up the definition of civil twilight. I would have been most dissapointed had I been redirected to dawn or dusk. Dawn and dusk refer to specific times. Twilight refers to a period of time. - Allen Oliver 18:42, 19 February 2006 (UTC)

Oppose per Allen Oliver. Bad idea. Civil twilight, nautical twilight, and astronomical twilight are specific terms with distinct meanings which would be lost with a merge into "dusk" or "dawn". - Sensor 03:42, 1 March 2006 (UTC)

Oppose per Shiftchange above. Dusk and dawn are terms for evening twilight and morning twilight respectively. Twilight is the usual scientific term. Dusk and dawn should be merged into Twilight. 204.101.243.169 16:36, 11 April 2006 (UTC) (logging in again to sign) Slowmover 16:37, 11 April 2006 (UTC)

Oppose Dawn and dusk are both twilight - before the sun rises, and after the sun sets respectively. Merging all three articles (dawn, dusk, and twilight) as a single article under "twilight" would be a good idea. 203.122.108.171 (talk) 17:17, 2 January 2008 (UTC)

Length of twilight

The length of twilight is also influenced by the time of the year as well (longer near the solstices). I'd like to see this topic treated better in this article.

68.183.119.26 17:52, 27 December 2006 (UTC) ngur

So would I, and I'd like a clearer explanation of why the length of twilight changes with latitude. The concept of observer's horizon only distracted me from thinking about the relative motions of the sun and earth. This cartoon about global twilight durations sorted it out for me:

http://www.mangobay.cc/users/moonfinder/sep-99.htm

and this one about seasonal twilight durations was also helpful:

http://www.mangobay.cc/users/moonfinder/aug-98.htm

but I don't know if it's appropriate to add these links to Wikipedia (I'm new here). The cartoons originally appeared in Sky & Telescope magazine.

--ELefty 19:13, 3 July 2007 (UTC)

The graph currently accompanying this section does not show duration of twilight. It shows total duration of daylight (from dusk to dawn). 68.122.104.16 (talk) 01:22, 25 October 2008 (UTC)

Ecclesiastical uses of twilight

The article as it stands mentions twilight in the context of the day's fasting during Ramadan. My understanding (which I don't trust enough to add to the article; it contradicts the information in day, for one thing) is that in Jewish ecclesiastical practice, the end of one day/beginning of the next happens in evening twilight, when it is dark enough that one cannot distinguish a blue thread from a black thread when they are held in the hand at arm's length. Expand the "other uses" section, perhaps renaming it, to include these and related points? BSVulturis 19:45, 29 December 2006 (UTC)

twilight before dusk?

is twilight bfore dusk or is it the ther way around?--Hicups0002 08:19, 27 January 2007 (UTC)

Neither, evening twilight is dusk; morning twilight is dawn (in the broad sense, not to be confused with sunrise, which is the end of morning twilight). Refs: the USNO pages, etc. 193.122.47.170 18:58, 8 March 2007 (UTC)

Mention of burglary

The article mentions night-time burglary carrying a lesser penalty. Am I correct in thinking this is something that applies only in some US states and should be noted in this article? 82.151.234.75 12:14, 16 February 2007 (UTC)

This concept applies to Australian Law as well. As it is not univeral but is international I would prefer the wording "some juristictions". 202.134.253.55 13:25, 7 November 2007 (UTC)

Best time for photography

This article should contain what many photographic lessons say -- "Twilight" is the best time to take photos. Also, unrelated, but still a question -- should it be Nautical or Civil Twilight? —Preceding unsigned comment added by 125.17.142.146 (talk) 09:51, 25 October 2007 (UTC)

Sunsets?

Why is this page polluted with sunset photos? As the article points out, twilight refers to the light by which we can see once the sun has set (or before it rises). It needs illustrating with photographs showing this, not some incidental and gratuitous collection of pretty skies. Assuming no-one can provide a justification for them, I'll replace them with some more relevant images in the course of copyediting the page. --mikaul^talk 14:14, 19 November 2007 (UTC)

I agree, especially with the silo photograph: twilight is when one can still see *without the aid of artificial light*, which is prominently shown in this photograph, despite the pretty evening star. Delete? --Stefankamph 08:48, 1 December 2007 (UTC)

Yes, I'd like to replace both those bottom two images, but at least the silos one shows a visible star (mentioned in adjacent text). The last photo needs to illustrate astronomical twilight (almost dark) so it's really there as a placeholder for now. I've uploaded one for civil twilight and hope to have replacements for these two soon. --mikaul^talk 18:54, 1 December 2007 (UTC)

In response to the statement above, "twilight refers to the light", I would like to point out that the article indicates twilight is a period of time, not a kind of light. Therefore showing photos taken at this time that support related text in the article is not incidental. Maybe there were too many sunset pics, but consider this clarification when evaluating the reason some of the photos appeared on this page. - Shiftchange 21:57, 1 December 2007 (UTC)

I take your point, but technically, it's neither: it refers to the relative position of the sun. What the sky looks like is only borderline relevant, IMO. There's no picture of the sky at daylight, for example. Like daylight, it's more descriptive and easy to understand if it's partly defined by what it does, as much as what it is. It does kind of make the sky look pretty, but then that's (amply) dealt with at sunset. A good part of its definition is the practical relevance of twilight; basically, what you can and can't do at various stages just before daybreak or after sunset, which means including images of (or at) typical light levels.--mikaul^talk 00:20, 2 December 2007 (UTC)

rejig

In fairness to those sunset images, I've removed the monument one for the same reason: we need to illustrate the light the sky gives off at these times, rather than the light remaining in the sky. A silhouette is obviously not giving that impression. The rest of the rejig was just to lay out the images better. The duration chart is awesome, btw :o) --mikaul^talk 23:33, 28 November 2007 (UTC)

Headline text

The twilight zone is one of the most amazing places. It is DARK and thats awesome! hehehehehehehehe —Preceding unsigned comment added by 134.197.171.254 (talk) 20:26, 23 September 2008 (UTC)

Uht

I don't know where this could be put in, but the Anglo-Saxons had a concept of what they called 'uht,' which was the time just before twilight in the morning when supposedly the sky is blackest. 140.247.44.14 (talk) 07:41, 23 April 2008 (UTC)

Twilight versus daylight

The chart Image:TwilightLength.png seems to be confusing twilight with daylight. Twilight does not last for 12 or 14 hours a day in a typical region. - SimonP (talk) 13:51, 25 April 2008 (UTC)

No Simon the chart seems correct. Take a look at the discussion re: Civil, Nautical and Astronomical Twilight definitions below. - AdeBarkah (talk) 02:35, 5 May 2008 (UTC)

WRONG definitions of Civil, Nautical, and Astronomical Twilights!

It's very important to make a distinction between the casual concept of twilight and the technical terms Civil Twilight, Nautical Twilight and Astronomical Twilight which have counter-intuitive meanings!! In particular the main article got the definitions of Nautical Twilight and Astronomical Twilight completely wrong!

By definition, Civil Twilight begins in the morning when the sun is 6 degrees below the horizon, lasts through the day, and ends in the evening (when the sun is again 6 degrees below the horizon.) During this entire period, outdoor activities can generally be undertaken without the assistance of artificial lighting.

So there is really no "morning civil twilight" which ends at sunrise, or an "evening civil twilight" which starts at sunset. There is only one Civil Twilight: it begins in the morning and ends in the evening. For example, today in New York City Civil Twilight began at 5:20 AM and ended at 8:26 PM.

Similarly there is only one Nautical Twilight. Like Civil Twilight, it begins in the morning(!) (when the sun is 12 degrees below the horizon), lasts through the day, and ends in the evening (when the sun is again 12 degrees below the horizon.) Sailors generally cannot "take reliable star sights of well known stars" during Nautical Twilight as noted in the main article. They can do so either before or after Nautical Twilight, not during.

And you might have guessed it by now, there is only one Astronomical Twilight, which begins in the morning(!), lasts through the day, and ends in the evening. Astronomical Twilight begins and ends when the sun is 18 degrees below the horizon.

All of this is explained in detail at the US Navy page referenced by the main article: http://aa.usno.navy.mil/faq/docs/RST_defs.php.

- AdeBarkah (talk) 02:30, 5 May 2008 (UTC)

Okay, I don't know where you learned your twilight from but that's not how it is. If you look closely, it says for computational purposes, meaning that that's almost a spherical cow. Notice how every other twilight definition is ambiguous. Any other source should show morning or evening civil, naut etc. Sailors generally do take star sights during nautical twilight because if they did it brighter the bright stars couldn't be seen and if they did it darker the sky'd be too dim to stand out against the sea well and see the horizon's (to measure their altitude). Notice that what is said about stars at end of civil twilight, EOCT=BONT. Full Moons, or Scottish midsummer midnights are equivalent to nautical twilight. Cities are nautical twilight or greater forever. Sagittarian Milky Way (talk) 14:18, 7 May 2008 (UTC)

Actually AdeBarkah has succinctly explained the difference in terms, the image that raised the contention is accurate and the article needs to be clarify the terms more accurately. Civil, nautical and astronomical twilights all begin in the morning and end in the evening. If you run a twilight report in the program called Astronomy Lab 2, for example, it will confirms this. - Shiftchange (talk) 23:11, 7 May 2008 (UTC)

Hmm this is interesting and unexpected. I was looking for Chicago twilight times on the web and noticed this too. They begin in the morning and end in the evening. I find it strange that astronomical twilight starts in the morning but I guess it does. Eg. http://www.timeanddate.com/worldclock/city.html?n=64 for Chicago. I'm a pilot and we do say "morning civil twilight" but the FAA regs define the three twilights exactly the same as the Navy definitions AdeBarkah refers to (word per word.)

I would say that this is a technicality in the definitions. It is correct that civil twilight begins in the morning and ends in the evening according to the astronomical definitions. The same goes for nautical twilight and astronomical twilight. Thus, during the day - technically - it is civil, nautical and astronomical twilight at the same time. However, daylight takes precedence over civil twilight, civil twilight over nautical twilight etc. for all practical purposes. Therefore (in the morning) the astronomical twilight is "suspended" by the onset of nautical twilight, which in turns is "suspended" by the onset of civil twilight. I think it is better to stay with the definitions as they are and mention this technicality in a side note? 90.184.242.141 (talk) 17:04, 5 June 2008 (UTC)

With all the pedantry, the article has been edited so that it's practically meaningless unless one reads the discussion on this talk page. Good job, Wikipedia. <sigh> —Preceding unsigned comment added by 67.185.156.172 (talk) 00:51, 15 December 2008 (UTC)

There is little point in supplying the technical definitions of Civil, Nautical, and Astronomical Twilights if one isn't going to apply the primary definition of twilight, which, from the USNO webpage, states, "Twilight: Before sunrise and again after sunset there are intervals of time, twilight, during which there is natural light provided by the upper atmosphere, which does receive direct sunlight and reflects part of it toward the Earth's surface" (http://aa.usno.navy.mil/faq/docs/RST_defs.php). Therefore, these three distinct types of twilight happen before sunrise and after sunset. They do not extend through the day. The language is abbreviated in each case, "is defined to begin in the morning, and to end in the evening", because it has already been stated that twilight ends in the morning at sunrise and begins in the evening at sunset. The idea that twilight contiunues through the day is preposterous and incorrect. Underwriter (talk) 16:26, 2 March 2009 (UTC)

Twilight Zone

Should there be a reference to the TV show The Twilight Zone on this page? The TV show takes its name from the natural phenomenon. Theneogon (talk) 02:18, 29 May 2008 (UTC)

Hmmm, the 9th edition of the OED gives twilight as the period after sunset, and dusk as the latter and darker part of this. I realise that the OED gives commonly accepted usage and not always the rigorous definitions by supposedly authoritative bodies, but, shouldn't this definition be mentioned? 68.228.208.191 (talk) 01:19, 26 September 2008 (UTC)

Addendum to above comment............

Shouldn't the 'authorities' who define twilight as per the article be mentioned? Also, in an encyclopedic article, it might be helpful to state where they derive their authority from.

In the UK, twilight is generally accepted as the period after sunset; hence, such expressions as "twilight of the Gods", or "the twilight of one's life", both expressions indicative of the concept of "day's end". :)

68.228.208.191 (talk) 01:51, 26 September 2008 (UTC)

Conflict

I'm not a native english speaker, so I dont actually know the precise meaning of twilight, but there is some conflict between 2 articles:

Dusk states:

"Dusk refers to the period of time following sunset. Although commonly confused with twilight, dusk is the time frame that occurs either before or after a twilight..."

Twilight states:

"Twilight is the time before sunrise (dawn), and the time after sunset (dusk)..."

Which one is correct? Almighty11 (talk) 20:33, 30 November 2008 (UTC)

Twilight. - Shiftchange (talk) 21:36, 30 November 2008 (UTC)

Reversion too broad

The reversion swept away corrections to the text as well as the image placement.

The article says:

Twilight is specifically defined as the period before daytime or after nighttime during which it is possible to conduct outdoor activities without the aid of artificial light.

The link given, as well as other discussion on this talk page, says otherwise. Twilight includes all times "during which it is possible to conduct outdoor activities ...", including broad daylight. That is not the definition I am accustomed to, but it is the technical definition.

The edit summary in the reversion says "please place diagrams elsewhere." Such as ...? Could it not have been moved to that suggested location instead of reverted? - Wmc824 (talk) 20:44, 31 December 2008 (UTC)

The existing lead image seems to be mislabeled in the caption.

The actual title of the painting is "The Evening" according to the information on the original image:

http://en.wikipedia.org/wiki/File:Caspar_David_Friedrich_008.jpg

The same title is used here:

http://www.canvaz.com/painters/friedrich1.htm

Google's translator agrees:

http://translate.google.com/translate_t?hl=en#de%7Cen%7CDer%20Abend

Der Abend = The evening

One web page says that "dammerlicht" is German for "twilight".

No form of the word "abend" appears in Götterdämmerung, which is translated "Twilight of the Gods".

The article contains five images. The captions on two of them are misleading. The other three pictures are eye candy. I created an image that was specifically relevant to a point of confusion discussed on the talk page and it was reverted. I don't get it. -Wmc824 (talk) 21:58, 31 December 2008 (UTC)

I restored my previous edit at 18:34, 31 December 2008, which was reverted for undefended reasons. It was not merely a "good faith edit"; it was simply a "good edit". I also preserved the edit by 92.20.46.130 at 16:01, 3 January 2009.

Three days ago, an invitation was left on my talk page to discuss the reversion here. I began the discussion here at that time. There has been no reply.

The editor who made the reversion previously commented on this talk page:

[Twilight] refers to the relative position of the sun. What the sky looks like is only borderline relevant, IMO.

- 00:20, 2 December 2007 (UTC)

The image I created is specifically about "the relative position of the sun" for various interpretations of the word "twilight". The illustration was placed next to definitions which differ. That seems like a good place to show interpretations of both definitions.

The photo of a painting of a dim horizon (a twice-removed representation of a symptom of solar position) by Caspar David Friedrich was mislabeled. The implied name of the painting in the caption matched the name of this twilight article, not the name of the painting.

Even after changing the lead image to one that is quite beyond "borderline relevant", three of the five images in this article continue to show dimly lit horizons. Do we need a fourth picture to show yet another dimly lit horizon?

An illustrative "diagram" seems like a pretty good choice for the lead image after all. - Wmc824 (talk) 04:35, 4 January 2009 (UTC)

Point conceded. For the record, I have no idea where the Friedrich painting's caption name came from, although I'm fairly sure I'd was me who posted it. I've always hated the sunset pics on this page and that painting replaced a sunset pic as a much more accurate illustration of twilight. Your diagram, from this descriptive perspective, belonged in a more explanatory section, hence my rather hasty reversion. My apologies for that, it's clearly excellent frm either perspective and a worthy lead image. mikaul^talk 12:26, 8 January 2009 (UTC) f

I'd prefer that painting to those other two sunset pics, which IMO exist only as incidental eye candy. They survive here mainly because the text mentions twilight a an inspiration to artistic works and (IMO) the painting serves that purpose better than either of them. mikaul^talk 12:34, 8 January 2009 (UTC)

Twilight All Night: Astronomical 48° 34’ Nautical 54° 34’ Civil 60° 34’

This article claims that in Oslo 59° 57’ (59.95°), Stockholm 59° 21’ (59.35°), Helsinki 60° 10’ (60.17°), Tallin 59° 26’ (59.43°)and Saint Petersburg 59° 56’ (59.93°); "civil twilight lasts allnight in midsummer". Furthermore, that in Hamburg 53° 35’ (53.58°), Gdansk 54° 22’ (54.37°) and Edmonton 53° 32’ (53.54°); "nautical twilight lasts allnight in midsummer". These are blatantly false. Received this in an e-mail from the U.S. Naval Observatory, "Nautical twilight begins and ends when the center of the sun is 12 degrees below the horizon. Therefore the most extreme latitude (north or south) that Nautical Twilight can last all night is 90 - 12 - obliquity of the ecliptic. The obliquity is now 23 degrees 26 minutes, which makes the most extreme latitude indeed 54 degrees 34 minutes" (54.56667°). Therefore, Astronomical Twilight is 90 degrees - 18 degrees = 72 degrees - 23 degrees 26 minutes = 48 degrees 34 minutes (48.56667°). Civil Twilight is 90 degrees - 6 degrees = 84 degrees - 23 degrees 26 minutes = 60 degrees 34 minutes (60.56667°). The websites below will confirm this, by imputing coordinates. Impute 60° 34’ (60.56667°) with a longitude and the correct time zone. You will see that Civil twilight does occur on the longest days. However, impute 60° 33’ (60.55°) and you will find that Civil twilight does not occur, even on the longest days. The same will occur with Nautical twilight, 54° 34’ (54.56667°) does occur, 54° 33’ (54.55°) does not occur. Astronomical twilight, 48° 34’ (48.56667°) does occur, 48° 33’ (48.55°) does not occur: HM Nautical Almanac Office, Websurf: http://websurf.nao.rl.ac.uk/ Geoscience Australia: http://www.ga.gov.au/geodesy/astro/sunrise.jsp Federal Aviation Administration, Sunrise/Sunset/Twilight/Calculator: http://akweathercams.faa.gov/srsscalc.php In June 2008, I received this e-mail from the U.S. Naval Observatory, "Further testing we have done seems to indicate that our online program may be missing some phenomena at this near-degenerate case; that is at the longitudes you tested, below N54 34 limit, the program should have found that there was a specific beginning and end to nautical twilight and not shown ////. We are investigating now to most simply modify the program to find these cases. The repair will take several weeks at best. Thanks for your diligence in testing our application and for your e-mail calling attention to the problem." However, I have just tested the coordinates, that I found were incorrect in June 2008. They were in the UK, Germany, Russia and Canada. All with exception of the Canadian coordinates, are still producing errors regarding twilight. The above latitudes for twilight are approximate. So, with the precise figure for the obliquity of the ecliptic; the precise latitudes could be obtained, for that date. They are in my opinion, very slightly below those I have shown. Sulasgeir (talk) 23:29, 15 January 2009 (UTC)

Vandalism

I'm the unlisted editor who recently made changes.... [in deleting] ... obvious idiocy and irrelevance.

CheshiresMasquerade (talk) 16:02, 7 January 2009 (UTC)

Credit for appropriate action against vandalism duly noted, though it seems unnecessary to be so meticulous about the "paper trail" in this case. I deleted the copy of the vandalism that was included in the remarks above. As it was "obvious idiocy and irrelevance", there would seem to be no need to repeat it here. A history of changes to the article is visible here. Anyone wishing to see what was deleted in your edit could do so via this diff.- Ac44ck (talk) 17:11, 7 January 2009 (UTC)

Edited "2 Length", due to errors.

Since there are blatent errors, I have edited the "2 Length" part of this article. My reasons are above in 18, Twilight All Night: Astronomical, Nautical, Civil. Sulasgeir (talk) 03:16, 16 January 2009 (UTC)

Civil Twilight - redirection?

How about the South African band "Civil Twilight"? It doesn't seem that there's an entry about the band in wikipedia, and it seems a bit unrelated to put it in the disambiguation page of "Twilight". How about adding a new disambig page for "CIvil Twilight" and to reconsider redirecting all links of "Civil Twilight" to "Twilight#Civil Twilight". —Preceding unsigned comment added by 142.151.157.70 (talk) 22:26, 28 February 2009 (UTC)

Rotational speed

This is because at low latitudes the sun's apparent movement is perpendicular to the observer's horizon, in addition to the fact that the rotational speed of a specific location is highest at the Equator and slower as latitude increases. This doesn't make sense / isn't clear to me ... the rotational speed of the earth should be the same everywhere. Unless it's referring to rotational speed on an axis perpendicular to the sun? Anyway, it would be cool if someone were to reword. Gerardw (talk) 21:38, 12 May 2009 (UTC)

This effect is completely due to the steepness of the (apparent) path that the sun takes after setting, and due to the fact that the set angle precipitously bends up in alt-az even at moderate latitudes (ESPECIALLY in summer), because small circle(s) are involved. Angular velocity (°/hr) is constant. We need more miles per hour at lower latitudes simply because the Earth is so much wider there and still we need to get there, like today.. The Sun still sets 15 degrees an hour, and speed of either kind is irrelevant to different lengths of twilight.Sagittarian Milky Way (talk) 06:54, 26 May 2009 (UTC)

In response to Gerardw. http://www.ncgia.ucsb.edu/education/curricula/giscc/units/u014/tables/table02.html The Earth's Equatorial circumference is 24901.5 miles therefore 1 degree longitude equates to 69.17 miles.This 69.17 miles also represents the speed at which the Earth turns at the Equator every 4 minutes and makes an entire 24901.5 mile rotation in 24 hours.At 60 degree latitude,the distance for 1 degree and speed for 4 minutes is 34.67 miles.It follows that the quicker speed at the Equator gives the effect of a rapid transition from daylight to darkness while at 60 degrees latitude,a person experiences a slower transition from daylight to darkness.

The huge problem is that the widely accepted value for the Earth's 360 degree rotation is not 24 hours but a false late 17th century contrivance otherwise known as 'sidereal time'.That value cannot express the rotational speeds needed to explain the twilight effect as it contains no information about planetary shape or rotational characteristics.If you try to use the 23 hour 56 minute 04 second value along with planetary geometry to explain the difference between Equatorial and more polar latitudes in terms of twilight,you will quickly develop a distinct antipathy towards 'sidereal time' and the reasoning behind it.Oriel36 (talk) 22:24, 12 July 2009 (UTC)

That two phenomena coincide does not imply that one follows from the other. Nor does the alleged inaccuracy of the sidereal day have anything to do with the varying duration of twilight. —Tamfang (talk) 18:44, 15 July 2009 (UTC)

I agree that the discussion about rotational speed was confusing; so I removed it. The surface speed seems to have nothing to do with the length of twilight according to the formulae here:

http://herbert.gandraxa.com/herbert/lod.asp

where the mention of "radius" is an angular radius from an apparent center, not the linear distance from the 3D center. The latter is related to surface speed as measured by an observer in the "fixed" frame of reference for sidereal time.

There is nothing in the formulae in the link above to suggest that the length of twilight would be affected by doubling the radius of the earth. If surface speed were an issue, then the length of twilight should be shorter at the same latitude after doubling the radius of the earth.

I also modified the references to twilight lasting "all night". The term "night" is given a specific meaning in the "definitions" section. It may be confusing to read that "twilight" can last all "night".

The mention of white nights seemed out of place where the discussion about 24-hour twilight applied to all three kinds. I deleted it and moved the wikilink. - Ac44ck (talk) 04:10, 2 August 2009 (UTC)

Latitudinal speeds are crucial to understanding why there is a rapid transition to darkness at the Equator and less so towards the geographical poles as rotational speeds diminish with latitude. Removing daily rotation and subsequently, different latitudinal speeds as the main cause of latitudinal variations in twilight leaves nothing but the useless explanation based on the motion of the Sun as a cause.Before you remove planetary dynamics from the explanation and focus on the motion of the Sun,I strongly suggest you come to terms with astronomical scale - http://images.funadvice.com/photo/image/old/39032/Sun__Earth_size_comparison_labeled.jpg Restore planetary dynamics to the explanation and clarity returns to the twilight explanation while omiting dynamics leaves nothing but a geostatic mess based on the motions of the Sun. Oriel36 (talk) 10:50, 10 August 2009 (UTC)

What verifiable mathematical formula shows that the length of twilight depends on surface speed? Please demonstrate mathematically that the length of twilight at a given latitude will change if we double the surface speed by doubling the planetary radius while the rate of rotation remains the same. - Ac44ck (talk) 03:25, 11 August 2009 (UTC)

Oriel, are you saying that the apparent motion of the Sun is not a cause of the day-night cycle?! —Tamfang (talk) 19:11, 11 August 2009 (UTC)

Why are you begging an 'If' question from me ?, the Equatorial speed of the Earth is 1669.8 km per hour and 837 km per hour at 60 degree latitude causing the twilight effect to be longer as the rotational speed diminishes towards the geographical poles.If you can't comprehend that as a location rotates into the orbital shadow at a slower speed you get longer twilights you belong nowhere near this topic.Oriel36 (talk) 17:06, 11 August 2009 (UTC)

A thought-experiment, varying that which is fixed, is a standard way for scientists to understand what's really going on.

Don't make cracks about other people's comprehension until you can address the criticisms of your position, in a way that shows you understand those criticisms, rather than merely repeating yourself. —Tamfang (talk) 19:17, 11 August 2009 (UTC)

The imperative I have to give to you is to discover that the Earth is round and rotating and the values for rotation at different latitudes are known and have definite effects such as the faster the rotation the quicker the transition from daylight to darkness,anyone who can't respond to this cause and effect with the appropriate affirmation has no business near astronomical timekeeping,planetary dynamics and the effects of these daily rotational and orbital dynamics.Oriel36 (talk) 17:16, 12 August 2009 (UTC)

Oriel36, you replied twice without providing a single formula. You haven't demonstrated "cause and effect". If surface speed determined how fast day transitions to night, why does the transition happen faster at the same location in winter than it does in summer? The surface speed did not change between seasons. Where is this "cause and effect" (slower surface speed causes longer transition from day to night) in winter? -Ac44ck (talk) 01:20, 13 August 2009 (UTC)

Daily rotation causes the day and night cycle,have you got that straight?,now,the next thing you can say is because the Earth is a rotating sphere, the maximum rotational speed is at the Equator and diminishes towards the geographical poles.Here are those values representing 4 minutes/1 degree of rotation- http://www.ncgia.ucsb.edu/education/curricula/giscc/units/u014/tables/table02.html .Now,common sense,and it is common sense,determines that at any given time,a location at the Equator is transiting through the boundary between solar radiation (daylight) and the orbital shadow (darkness) at 1669.8 km per hour while a location at 60 degrees Latitude either side of the Equator is transiting through the same boundary at 837 km per hour irrespective of seasonal differences.The effect at the Equator compared to that at 60 degree latitudes is that twilight is longer compared to the rapid transition at the Equator at all times.Of course,the same people who cannot describe the twilight effect properly based on rotational dynamics refuse to accept the rotational speeds in that table above and this means that you have a problem with the basic day and night cycle let alone latitudinal twilight comparisons and their causes.Let me see what value you give for daily rotation per hour at the Equator and 60 degrees and then we can talk.Oriel36 (talk) 09:16, 13 August 2009 (UTC)

So you agree that rotational speed, at a given latitude, does not vary with the seasons. Do you also acknowledge that the length of twilight, at a given latitude, does vary with the seasons? Can you see why some people might have a problem with the proposition that something which varies is wholly determined by something which does not vary?

No one is denying that rotational speed depends on latitude. That's a red herring. (Gerardw didn't get it at first, but he's not now participating.) If I remember right, someone did quibble (in Talk:Earth's rotation) with your exact value for the speed; but that's irrelevant, since the variation is proportional. If we stipulate that your favorite webpage is accurate, will you then answer Ac44ck's challenge for a formula?

What you're not getting (it seems) is that our movement through the twilight zone varies not only in linear speed but also in angle.

Be careful with language like "people who cannot describe the twilight effect properly based on rotational dynamics"; to the rest of us, that's you. —Tamfang (talk) 20:53, 13 August 2009 (UTC)

Oriel36, you're spouting nonsense; for example:

darkness descends quickly at the Equator and slower towards either poles,irrespective of the seasons

How quickly does darkness descend at 89 degrees north latitude on 21 June? There is no darkness at 89 degrees north latitude on 21 June.

Please tidy up your language:

Let me see what value you give for daily rotation per hour at the Equator and 60 degrees and then we can talk.

"Daily rotation per hour" is word soup which seems to suggest angular velocity; it doesn't vary with latitude. Linear velocity at the surface – relative to some 'fixed', off-world observer – does vary with latitude. Clarity of expression may promote clarity in understanding.

Watch your step off that high horse. - Ac44ck (talk) 03:57, 14 August 2009 (UTC)

At summer solstice, near the pole, darkness descends very very very slowly indeed. —Tamfang (talk) 05:13, 14 August 2009 (UTC)

It is not at all difficult to understand the cause (latitudinal rotation speeds) and effect (twilight variations) and the dynamical explanation should present no comprehension problem for any student or reasonable person.The previous explanation is good enough for anyone and if you cannot understand the link between rotational dynamics and why darkness descends quickly at the Equator and slower towards either poles,irrespective of the seasons,I have no intention of listening to any other cause than rotational dynamics for the simple reason is that there is no other explanation.Oriel36 (talk) 21:14, 13 August 2009 (UTC)

Then it should be no trouble at all for you, with your superior understanding, to provide a formula that gives the length of twilight, at each season, in terms of rotational speed (and no other variables). —Tamfang (talk) 22:44, 13 August 2009 (UTC)

There is no superiority,the cause and effect is so simple to grasp ,latitudinal variations in rotational speed correlate with longer twilights with the most rapid transition to darkness existing at the Equator that any other imagined cause would be beneath me to consider.The latitudinal twilight variations are an intrinsic part of daily rotation and after the cause of the daylight/darkness cycle due to daily rotation ,the next insight is straightforward enough as the Earth is a rotating sphere with rotational characteristics which see latitudes towards the geographical axis produce longer and longer twilights as part of the daily cycle.If you pair cannot understand it this way,how would you expect the wider population to understand it.Oriel36 (talk) 13:07, 14 August 2009 (UTC)

Certainly it's simple: too simple to account fully for the phenomenon, no matter how many times you say it. Rotational speed cannot account for the seasonal variation. You will never be taken seriously unless you address this point. Are you unaware (or do you deny) that the seasonal variation exists? —Tamfang (talk) 05:13, 15 August 2009 (UTC)

TMI about length

ISTM that the section on length is beginning to read more like an almanac than an encyclopedia. All the details about when twilight happens, how long it lasts in a multitude of particular spots, etc., seems a bit overmuch to me. -Ac44ck (talk) 03:44, 9 July 2009 (UTC)

Nor does a footnote with the wording "Therefore it seems to me ..." read like an encyclopedia. - Ac44ck (talk) 03:06, 17 July 2009 (UTC)

Material about the change in obliquity over centuries belongs in another article. —Tamfang (talk) 22:49, 17 July 2009 (UTC)

Perhaps here, which is where obliquity redirects to, would be a good place for a detailed discussion of that. The topic has some relevance in a discussion about twilight, be maybe not in such depth. -Ac44ck (talk) 20:48, 18 July 2009 (UTC)

Relevance to twilight can be disposed of in one sentence: "These limits change as the axial tilt varies over the centuries." —Tamfang (talk) 20:39, 19 July 2009 (UTC)

I reverted the restoration of info about latitudes. Note the text at the bottom of an edit page: "Content that violates any copyrights will be deleted. Encyclopedic content must be verifiable".

Was the referenced email private? Is it verifiable? Was their permission granted to publish it in Wikipedia?
The meaning of "at present" is not defined
"Twilight lasts all night" sounds like a contradiction. If it is "night", how is it "twilight"? Maybe twilight lasts for 24 hours?

-Ac44ck (talk) 00:34, 21 July 2009 (UTC)

As a matter of style I'd prefer to avoid the phrase "24 hours" (which means "24 times 1/24 of a day-cycle", which is a bit silly); how about "full night never falls"? —Tamfang (talk) 22:49, 13 August 2009 (UTC)

Actually, "24 hours" is problematic because one rotation doesn't happen in exactly 24 hours. "Full night never falls" seems problematic because the article essentially defines "night" as not-during-astronomical-twilight. I didn't like it saying "twilight lasts all night" so I "fixed it" with something that is problematic in a different way. Feel free to fix the fix. It may be technically correct to say that "twilight lasts throughout a synodic day", but it makes more work for a general reader to decipher "synodic day". Punt. - Ac44ck (talk) 02:53, 14 August 2009 (UTC)

I don't get how "night" is a problem, but okay, it's not important that *I* get it. ;) —Tamfang (talk) 23:34, 23 August 2009 (UTC)

Maybe I am slicing things too thin. "Night" is defined in the article:

Definition - Position of sun

Night - more than 18°

Thus, "night" and any of the three kinds of twilight seem to be mutually exclusive. To say that "twilight" lasts all "night" seems comparable to saying, "The traffic light was green for the entire time that it was red." YMMV. - Ac44ck (talk) 23:53, 23 August 2009 (UTC)

Yes, I got that. That's why I suggested a paraphrase of there is no night, but you didn't like that either, and that's what I don't get. —Tamfang (talk) 05:15, 24 August 2009 (UTC)

The suggestion was "full night never falls". So "partial night" does fall? What is "full night"? Besides, "night" is not-astronomical-twilight. I suspect that most readers are interested in civil twilight lasting "all night", which would seem to make their period of interest "nautical twilight never occurs during a synodic day." Or something to that effect. - Ac44ck (talk) 02:56, 26 August 2009 (UTC)

the line break after the ammonite picture

Okay, let's discuss it. What's the spacing issue here? To me it looks about the same either way; is there a problem in narrow windows or something? —Tamfang (talk) 03:05, 21 August 2009 (UTC)

In some browsers, in particular Firefox 3, the text overlaps onto the picture making the text hard to read. The extra line break solves this problem for those who have this browser and causes no disadvantage to those who don't, so it makes sense to keep the line break. Hope this helps. Set Sail For The Seven Seas 174° 7' 0" NET 11:36, 21 August 2009 (UTC)

It looks like this is a generic MediaWiki problem when a larger image is directly below a smaller one. Adding a <br> puts extra spacing after the section header, though, so this isn't an ideal solution.

Since the duration graph wouldn't be as useful as a smaller thumbnail, perhaps we could just drop the "Twilight at Lyme Regis, UK." image, which doesn't seem to be related to the section? --McGeddon (talk) 11:57, 21 August 2009 (UTC)

After a quick look, that does seem to be the case. Alternate options would be to either resize the pictures so the the top one is the same size or bigger than the small one or switch the two pictures round; that way both pictures can be kept. Set Sail For The Seven Seas 222° 11' 15" NET 14:48, 21 August 2009 (UTC)

Swapping seems easiest, although I don't really see what a picture of a Lyme Regis sky adds to the section. (If it was a city named in the section, fair enough, but it's just a random British town.) --McGeddon (talk) 21:01, 28 August 2009 (UTC)

That seems like a conclusion. Swapping it is, though the picture could be deleted at a later stage or replaced with a more relevant one. Set Sail For The Seven Seas 322° 12' 15" NET 21:28, 28 August 2009 (UTC)

Done Set Sail For The Seven Seas 323° 16' 45" NET 21:33, 28 August 2009 (UTC)

Upon looking into this, it turns out that it is the new vector skin that is causing this and not the Firefox 3 browser as previously stated. Set Sail For The Seven Seas 188° 36' 0" NET 12:34, 20 September 2009 (UTC)

Terminology for extended twilight

I believe that the point of taking note of twilight lasting "all night" is that it lasts for at least 24 hours; or one rotation; or some such.

While it may be incorrect to say "twilight lasts for exactly 24 hours" for days on end, the intro which said "These are the largest cities, of their respective countries, where 24-hour twilight can occur" may have been accurate.

Dusk and dawn are not defined verbally in this twilight article. The article on dusk isn't very precise; but the article on dawn is: "civil dawn" is different from "nautical dawn". Can there be a transition from dusk to dawn if it doesn't stop being twilight? - Ac44ck (talk) 01:53, 9 September 2009 (UTC)

Well firstly, the order should be: night, astronomical dawn, astronomical twilight, nautical dawn, nautical twilight, civil dawn, civil twilight, sunrise, daytime, sunset, civil twilight, civil dusk, nautical twilight, nautical dusk, astronomical twilight, astronomical dusk, then back to night; and if terms are used more generally then the order should be: night, dawn, twilight, sunrise, daytime, sunset, twilight, dusk, then back to night. Also, each of these should be mutually independent from the other, so that when one starts, another ends. So when I said from dusk to dawn, what I actually meant was the dusk and dawn of the twilight above the one that lasts "all night". So, for astronomical twilight lasting "all night", dusk and dawn represented nautical dusk and nautical dawn; for nautical twilight lasting "all night", dusk and dawn represented civil dusk and civil dawn and for civil twilight lasting "all night", dusk and dawn represented sunset and sunrise. I have improved the definitions to include the more specific dusk and dawn. Finally, this idea of twilight "all night"; if the definition of night in the night article is used, then it is perfectly fine to say twilight "all night" as it defines night to be when the sun is below the horizon as opposed to 18° below the horizon. I have started a discussion about this double definition of night at the night talk page. You may want to join in as it affects the definitions here. Hope this helps. Set Sail For The Seven Seas 196° 18' 0" NET 13:05, 9 September 2009 (UTC)

Thanks for the invite to the discussion elsewhere. It might be some time before I can contribute there.

I am not grasping how twilight can last from dusk to dawn. It seems to me that once it becomes dusk, it stops being twilight. The following image may be a helpful reference.

A larger view is available by clicking the image. The red line is solar altitude. It is twilight when the solar altitude is above the horizontal line. The numbers beneath the vertical lines are 24-hour clock times. My interpretation is that twilight lasts for 60 hours: from 6 am on Day 1 until 6 pm on Day 3; from dawn on Day 1 until dusk on Day 3. Is this not a correct interpretation?

As for describing the duration of twilights which last "all night" how about interspersing "round-the-clock" with other descriptions? It doesn't say how many times "round", so it may be compatible with twilight lasting for weeks. - Ac44ck (talk) 04:25, 10 September 2009 (UTC)

The picture seems a bit misleading, at it implies at least 6 hours of night on day 1 and day 3 and no night on day 2. Generally, before and after the period with twilight "all night (sun below the horizon)", the night (sun 18° below the horizon) will be short and its length will increase or decrease gradually as one leaves or approaches the period with twilight all nighttime respectively.

Now to explain twilight lasting from dusk to dawn. If we think of the order above in a different perspective, it becomes: daytime, sunset, civil twilight, civil dusk, nautical twilight, nautical dusk, astronomical twilight, astronomical dusk, night, astronomical dawn, astronomical twilight, nautical dawn, nautical twilight, civil dawn, civil twilight, sunrise, then back to daytime.

During astronomical twilight "all night", the order becomes: daytime, sunset, civil twilight, civil dusk, nautical twilight, nautical dusk, astronomical twilight, nautical dawn, nautical twilight, civil dawn, civil twilight, sunrise, then back to daytime; here, astronomical twilight lasts from nautical dusk to nautical dawn.

During nautical twilight "all night", the order becomes: daytime, sunset, civil twilight, civil dusk, nautical twilight, civil dawn, civil twilight, sunrise, then back to daytime; here, nautical twilight lasts from civil dusk to civil dawn.

During civil twilight "all night", the order becomes: daytime, sunset, civil twilight, sunrise, then back to daytime; here, civil twilight lasts from sunset to sunrise.

Each time it is the dusk and dawn of the twilight above the one that lasts "all night" that is being referred to, not the one with the same first word in the name.

This way, the definitions are completely accurate and there is no disambiguity with the terminology. Hope this helps. Set Sail For The Seven Seas 235° 43' 45" NET 15:42, 10 September 2009 (UTC)

One other thing, "These are the largest cities, of their respective countries, where 24-hour twilight can occur", implies that there is no day, no night and that the sun remains within 0° and 18° below the horizon for the entire day. This can only occur at latitudes greater than 72°, which is well inside the polar circles. Hope this helps. Set Sail For The Seven Seas 338° 54' 15" NET 22:35, 11 September 2009 (UTC)

While the combination of terms is technically accurate in two out of the three cases:

Each time it is the dusk and dawn of the twilight above the one that lasts "all night" that is being referred to, not the one with the same first word in the name.

The introductory sentence doesn't make the caveat. And some kind of reference to twilight lasting for more than one day/rotation/whatnot seems to have more wow-power than a statement which relies on an ad hoc combination of technical terms with different reference points.

The section now says that it is:

Civil twilight from sunset to sunrise

That doesn't seem to be as advertised in the introductory sentence which would lead one to expect examples of twilight lasting "from dusk to dawn"; as there is no dusk or dawn during the period when twilight lasts from sunset to sunrise.

As for the one other thing, I don't infer that an absence of sunshine during twilight is required by the context. Why can't "24-hour/whole-day/insert_better_term_here" twilight include a "midnight sun"?-Ac44ck (talk) 19:21, 12 September 2009 (UTC)

I think I can see what you mean. I've rephrased the sentence accordingly, but the other definitions should be left as they are since they are completely accurate. Hope this helps. Set Sail For The Seven Seas 305° 59' 15" NET 20:23, 12 September 2009 (UTC)

"Why can't "24-hour/whole-day/insert_better_term_here" twilight include a "midnight sun"?" It could easily be confused with the midnight sun. Hope this helps. Set Sail For The Seven Seas 345° 2' 30" NET 23:00, 12 September 2009 (UTC)

These statements carry equal implication of the sun remaining below the horizon:

Twilight lasts for at least 24 hours
Twilight lasts from sunset to sunrise

It depends on how one interprets the word "twilight". Neither description above rules out a midnight sun during technical twilight. The colloquial meaning of twilight, "no sunshine but not dark yet", would keep the sun below the horizon for a full rotation in either case.

A main point in this section is that twilight can span more than one "day". I think it is preferable to find a simple yet accurate way to say that.

Using an invariable number of hours to describe a "day" can be problematic. A rotation happens in something less than 24 hours. Twilight might last for 23 hours and 59 minutes and span two "days" once per year just inside the arctic circle. For that rare exception, we might abandon a reference to 24 hours.

Extended nautical twilight, which is of limited interest, includes the period from civil dusk on one day until civil dawn another day. The description provides the technical start/stop limits for the unusual occurrence of twilight. Ditto for extended astronomical twilight, which is of even more limited interest.

Extended civil twilight, which is of most interest, includes the period from sunset on one day until sunrise another. Unlike the other descriptions of "all-night" twilights, the text doesn't mention the start/stop limits for the unusual occurrence of civil twilight. Some period of twilight after sunset is not unusual. One might say that "all-night" civil twilight lasts from "noon to noon" with equal correlation to the dawn/dusk limits.

I think it would be preferable to simply yet accurately convey the notion that twilight can span more than one "day". -Ac44ck (talk) 16:18, 13 September 2009 (UTC)

Technical twilight is generally used for computational purposes, hence the name. Here, twilight is not the same as technical twilight. Twilight can indeed span more than one day, but again as above, only at latitudes greater than 72° with the sun remaining within 0° and 18° below the horizon for the entire day. Also, don't forget that this is how twilight is defined in the definitions section of the article, so referring to technical twilight as twilight would not only be incorrect, but clash with the original and precise definitions as defined in the definitions section of the article. This has already been discussed here with the conclusion that technical twilight is not the same as twilight. The article is fine as it is now. Hope this helps. Set Sail For The Seven Seas 190° 1' 45" NET 12:40, 14 September 2009 (UTC)

I believe the term "technical twilight" is a neologism to encompass the three kinds of non-colloquial "twilight" succinctly.

A list with only three varieties is introduced by the sentence:

These are the largest cities, of their respective countries, where the various twilights can continue through local midnight

It doesn't have an entry for colloquial twilight where the sun stays below the horizon. I think the limit of 18° below the horizon wouldn't be appropriate in such an entry. I suspect that most people would call it "dark", not "twilight", before the sun is very far into nautical twilight; almost certainly so before the end of astronomical twilight.

If the last comment here is authoritative, then it stops being any kind of twilight at sunrise and the image in the article showing "continuous civil twilight" at the North Pole in summer would seem to be wrong.

Entering 89° north latitude after scrolling down to Form B here says "Sun continuously above twilight limit" for all of June. There would be no twilight of any variety in June at 89° north latitude if "the idea that twilight continues through the day is preposterous and incorrect." It suggests there are two limits for civil twilight: 0° below the horizon, and 6° below the horizon. But the Navy site doesn't say "Sun continuously outside twilight limits" (plural). How, then, to interpret the sun being both "above twilight limit" and above the horizon for an entire month? -Ac44ck (talk) 03:55, 16 September 2009 (UTC)

This is part of the description given for the summary for the image in the article.

"Approximate length of daylight, in hours, as a function of latitude and time of year. Here, the length of daylight is defined as the period from the beginning of civil twilight in the morning to the end of civil twilight in the evening. Morning civil twilight begins before sunrise when the sun is geometrically 6 degrees below the horizon and ends when the sun rises; evening civil twilight lasts from sunset until the sun is geometrically 6 degrees below the horizon. Civil twilight is roughly the period before sunrise and after sunset during which, under clear skies, outdoor activities can be carried out without artificial light."

It clearly states the definition of civil twilight and therefore the continuous civil twilight in the picture refers to technical twilight. The text in the picture has been badly worded, it should say continuous daylight. The sun being both above twilight limit and above the horizon for an entire month can be interpreted by simply having the horizon be the twilight limit, therefore above twilight limit and above the horizon mean the same thing. Better still would be to call 6 degrees below the horizon the lower twilight limit so that the horizon can be called the upper twilight limit. Hope this helps. Set Sail For The Seven Seas 331° 29' 0" NET 22:05, 16 September 2009 (UTC)

This may be of interest (Wikification added):

http://www.staleyhome.net/intel/IT0565.pdf

(2) Beginning morning civil twilight (BMCT) begins when the sun reaches an attitude of 6° below the horizon. ... BMCT ends when the sun reaches the horizon and sunrise occurs.

It has similar descriptions for the beginnings and endings of other twilights. The same info is found in a military field manual: FM 34-130. But the versions I found on the web were PDF files with the pages scanned as images rather than documents containing text.

Perhaps the current diagram in this article which shows "technical twilight" at noon is incorrect. Maybe it is either one (and only one) form twilight or it isn't twilight at all - technical or otherwise.

There is the wrinkle of defining "sunrise": with or without refraction; how much refraction; which part of the sun at the horizon.

http://aa.usno.navy.mil/faq/docs/RST_defs.php

For computational purposes, sunrise or sunset is defined to occur when the geometric zenith distance of center of the Sun is 90.8333 degrees. That is, the center of the Sun is geometrically 50 arcminutes below a horizontal plane.

Accuracy of rise/set computations: even under ideal conditions (e.g., a clear sky at sea) the times computed for rise or set may be in error by a minute or more.

Day length

More conveniently, atmospheric refraction is ignored and the center of the sun is often used in place of the upper limb for computing the day length.

Note the graph in the day length article shows "24-hour Daytime" rather than "continuous civil twilight". Does anyone want to go there and point out that a day isn't exactly 24 hours?

The beginning of any form of twilight can be calculated with great accuracy. The length of twilight, seemingly not so much. - Ac44ck (talk) 02:22, 17 September 2009 (UTC)

I think that that the diagram is correct, but that the technical twilight part has been put there to help understand it in comparison to twilight. Strictly speaking, technical twilight should solely be used in computation where it simplifies the creation of programs to calculate twilight times, but the output should display twilight with daytime and not technical twilight. Also, it shouldn't be too difficult to calculate the length of twilight once you have the beginning and the end of that particular twilight. The various links here link to online calculators for twilight. Also, I think we can agree now that the twilight article is accurate and that the terminology is correct. Hope this helps. Set Sail For The Seven Seas 332° 54' 0" NET 22:11, 18 September 2009 (UTC)

Where might one find a verifiable definition of "technical twilight"? It isn't defined in the article. It is shown in the image, but I believe it to be a neologism. The duration of so-called "technical twilight" answers the question, "How much time is available for an all-day activity that can't use artificial light?" It saves adding the durations of daytime and morning/evening twilights, but it seems to be an ad hoc term.

I think that:

The depiction of "technical twilight" should be deleted from the diagram in the Definitions section, and
The graph in the length section is not relevant in this article. The length of twilight should go to zero in both summer and winter at both poles. - Ac44ck (talk) 05:29, 19 September 2009 (UTC)

I disagree, I think that a new section on technical twilight should be created instead of removing all traces of technical twilight from the article, after all, technical twilight is still a valid definition and should be recognised accordingly. I also doubt that a separate article will be created on technical twilight and this is definitely the best place for it to be. This section should mention the difference between twilight and technical twilight alongside the fact that technical twilight is generally only used for computation purposes to make the programming for twilight calculation easier. Also, graph in the length section is definitely relevant to this article, it has just been badly worded. The original description that goes with the picture explains it much clearer. It should say continuous technical civil twilight or continuous daylight rather than continuous civil twilight. Hope this helps. Set Sail For The Seven Seas 324° 37' 15" NET 21:38, 19 September 2009 (UTC)

Where is "technical twilight" a valid definition? If it isn't a professionally recognized term, it doesn't need its own section. The phrase appears fewer than 2500 times here:

http://www.google.com/search?q=%22technical+twilight%22

In looking through the first few pages of hits, I didn't find it used in the sense proposed here.

Wikipedia is to contain verifiable information. I doubt that a reference untainted by this article defines "technical twilight" as starting before sunrise, continuing through the day, and finally ending after sunset.

The notion of "technical twilight" came from a run-on reading of

http://aa.usno.navy.mil/faq/docs/RST_defs.php

Civil twilight is defined to begin in the morning, and to end in the evening

A cursory reading (not helped by the construction of the sentence, previous discussion on this talk page, nor the graph in the article) may give the impression that once twilight starts, it continues through the portion of the day when there is direct sunlight. But the definition doesn't say that.

As for ease of calculation, a previous comment said both:

it simplifies the creation of programs to calculate twilight times, and
it shouldn't be too difficult to calculate the length of twilight once you have the beginning and the end of that particular twilight

I found it easier to write a program to calculate the duration from BMCT (beginning of morning civil twilight) to EECT. It doesn't require comparison with sunrise/sunset times, nor taking any stand on what "sunrise" means. But the result of the simpler program doesn't answer the question "How long is twilight?" Nor does the graph in the article. - Ac44ck (talk) 04:19, 20 September 2009 (UTC)

The reason why the term technical twilight doesn't appear very much is because it is sometime wrongly referred to as twilight as has been done above. The point here is to disambiguate between twilight and technical twilight, then use them accordingly in order to prevent confusion between twilight and technical twilight. The graph in the length section definitely has relevance here. It may not answer the question, "How long is twilight?", but that is not the reason it is here. It is here because it shows an important effect of twilight, which is the fact that those in between the latitudes 60° 33’ 43” and 72° 33’ 43” can experience 24 hour daylight or continuous technical civil twilight at the summer solstice, which is equivalent to the presence of sunlight all day, and not experience 24 hour darkness or civil polar night at the winter solstice. However, it would be interesting to see an equivalent graph showing twilight as opposed to technical twilight. The only thing is that since twilight can have two parts, morning and evening, they would have to be added together. Such a graph would probably show the length of twilight at the summer solstice increasing slowly with latitude, until the maximum extent of civil twilight continuing through local midnight, where it would decrease quickly and reach zero at the polar circle. It would also show the length of twilight at the winter solstice increasing slowly with latitude, until the polar circle, where it would decrease quickly and reach zero at the maximum extent of civil polar night. So, while adding a graph showing twilight would be of interest, the current graph should remain. Hope this helps. Set Sail For The Seven Seas 202° 54' 30" NET 13:31, 20 September 2009 (UTC)

Where is the term "technical twilight" used at all, in the sense proposed here, apart from this article or documents influenced by it?

The graph is located in the "Length" section of the "Twilight" article. That suggests it was included in the article to answer the question, "How long is twilight?" It doesn't answer that question. It is irrelevant in, and brings confusion to, the "Length" section. I see no other place in the article where it would be more relevant. It is a nicely done image; it just doesn't seem to be helpful in this article. The graph at day length is comparable and is labelled more correctly.

I think that the existing graph should be replaced with one showing the "length of evening civil twilight". The greatest number of people are awake in the evening. Some are not "morning people" and may take note of morning twilight less frequently. The length of morning twilight would be nearly the same, so a separate graph shouldn't be needed for it. The new graph would show 24-hour civil twilight in the appropriate regions. I don't think that a graph showing the lengths of morning and evening twilights added together would be of widespread interest. -Ac44ck (talk) 15:23, 20 September 2009 (UTC)

If there is to be a showing the length twilight then the morning and evening twilights must be added together, otherwise there will be a discontinuation or a "jump" at the maximum extent of civil twilight continuing through local midnight for the summer solstice or the polar circle for the winter solstice as the length of twilight suddenly doubles instantaneously as the morning and evening twilights merge into one. This effect can be seen in the comment above while explaining the idea of "twilight lasting from dusk to dawn". The fact that less people take note of morning twilight is irrelevant here and there is the situation where the sun rises so late that most people are awake when morning twilight occurs. Also, the graph in the length section should not be removed simply because it is badly worded. It should be kept, it just needs rewording and moving to another section, possibly a new section for technical twilight, which should be created to disambiguate between twilight and technical twilight. The fact that the term exists is enough to support a new section on it. Hope this helps. Set Sail For The Seven Seas 248° 57' 15" NET 16:35, 20 September 2009 (UTC)

There will actually be four discontinuities in the graph; two at both the top and bottom:

When morning and evening twilights merge into one, and
When the length of continuous twilight goes to zero as continuous sunlight reigns.

The jumps would illustrate reality and highlight the freakishness of the situations. The current graph obscures discontinuities.

I think the number of people who notice evening twilight is relevant when envisioning the audience of general readers.

Again, where does the term "technical twilight" exist? Before a section is created for "technical twilight", I think we need an independent reference illustrating that the term is recognized by a significant number of professional astronomers. I have said similar a few times. I haven't seen a link demonstrating such. If the term isn't generally recognized, I wouldn't favor making a section in the article to explain it or keeping a graph to illustrate it. - Ac44ck (talk) 17:49, 20 September 2009 (UTC)

There would not be a discontinuity when the length of continuous twilight goes to zero as continuous sunlight reigns as the length of continuous twilight would gradually go to zero. Plus, the "jumps" would be misleading as it is completely exaggerated. There should be no discontinuities since this is only the length nautical twilight gradually increasing from or decreasing to zero and is therefore illustrating no freakishness whatsoever. The whole process is very smooth and had absolutely no discontinuities at all. This is why the morning and evening twilights would need to be added together. Also, there cannot be a bias in favour of evening twilight over morning twilight just because more people observe it, both are as relevant as each other. The point of creating a section is to disambiguate between twilight and technical twilight and reduce confusion. If there is no disambiguation, there will be confusion as to which twilight is being referred to. It is mainly to simplify understanding of twilight and technical twilight, which can all too easily be confused with one another as has been the case here. Hope this helps. Set Sail For The Seven Seas 277° 55' 45" NET 18:31, 20 September 2009 (UTC)

This is a curious exchange. I started with the misconception that the so-called "technical twilight" was defined by astronomers and that the graph in the Length section was valid. You convinced me that twilight happens only while the sun is below the horizon, as I thought before exposure to the errant notion in this article. Now I'm calling for deletion of the graph which shows twilight continuing past sunrise, and you want to keep it.

I have asked at least four times for a reference defining "technical twilight". I have looked for one and posted a link to Google. I haven't found an independent one. Have you? Do we need more than one before making a section to describe it?

The discussion here, referenced in the reply above, does not contain the term "technical twilight".

There will be discontinuities in a correct graph showing the length of twilight:

The transition from continuous twilight to continuous sunshine will be abrupt when the sun emerges from below the horizon. There is nothing smooth about the transition in "There was a beginning of evening civil twilight yesterday, but there won't be today."
The transition from long evening twilight to continuous civil twilight will be abrupt when the sun fails to go more than 6 degrees below the horizon. There is nothing smooth about the transition in "There was an end to evening civil twilight yesterday, but there won't be today."

I don't perceive that adding the lengths of morning and evening twilights together will be of interest to a general reader. Who would use such information and how?

Favoring "evening twilight over morning twilight just because more people observe it" is exactly what we should do unless we're going to show two practically identical graphs, or try to force people into more recognition of morning twilight. Is there some reason to have a "fair and balanced", politically correct graph showing the length of twilight? Are the NPOV police going to take us to task for not giving morning twilight its due? - Ac44ck (talk) 20:03, 20 September 2009 (UTC)

You're right, this is getting a little muddled up, so I'll simply state the points I'm making here:

1. Twilight is limited to before sunrise and after sunset.

2. Daytime is the time between sunrise and sunset.

3. Technical Twilight is Twilight plus Daytime.

4. Twilight and Technical Twilight are not the same and this needs to be made clear to the readers.

5. During astronomical twilight "all night", astronomical twilight lasts from nautical dusk to nautical dawn.

6. During nautical twilight "all night", nautical twilight lasts from civil dusk to civil dawn.

7. During civil twilight "all night", civil twilight lasts from sunset to sunrise.

"Also, there cannot be a bias in favour of evening twilight over morning twilight just because more people observe it, both are as relevant as each other."

OK, ignore this bit, it's irrelevant to my point and not a good arguement.

"The point of creating a section is to disambiguate between twilight and technical twilight and reduce confusion. If there is no disambiguation, there will be confusion as to which twilight is being referred to. It is mainly to simplify understanding of twilight and technical twilight, which can all too easily be confused with one another as has been the case here"

The situation here is that there are two terms, twilight and technical twilight, two definitions, without daytime and with daytime, and a huge clash. By stating that twilight is without daytime and technical twilight is with daytime, which is the definition given in the picture, it makes understanding both twilight and technical twilight so much easier.

I get the idea that you are inclined towards wiping out all traces of technical twilight and sticking to twilight. I would agree with it, but my only concern with this is that some readers will still think of twilight being with daytime and then we have this scenario all over again and then we would be going round in circles and not getting anywhere. If you can sort out this concern then I will agree and remove the technical twilight arrow from the picture in the definitions section.

"If there is to be a showing the length twilight then the morning and evening twilights must be added together, otherwise there will be a discontinuation or a "jump" at the maximum extent of civil twilight continuing through local midnight for the summer solstice or the polar circle for the winter solstice as the length of twilight suddenly doubles instantaneously as the morning and evening twilights merge into one. This effect can be seen in the comment above while explaining the idea of "twilight lasting from dusk to dawn"."

The reason I mentioned this is because at the point that the morning and evening twilights merge into one, morning twilight is suddenly included in an instant. My point here is, Why the sudden inclusion of morning twilight? It brings an abrupt and misleading jolt to the length of twilight. I completely agree that the in the strict boundaries of the definitions, the passing between daytime, night and the various twilights are completely abrupt, but the length of twilight will never increase or decrease abruptly. Google Earth's sun feature shows this very well.

We can agree that the occurring of the various twilights passing through local midnight is based on the lowest point of the sun or the lower culmination of the sun. This lower culmination will gradually sink upon moving away from the summer solstice and gradually rise upon moving towards from the summer solstice, with the lower culmination of the sun being at its highest at the summer solstice. Google Earth's sun feature will confirm this.

If the latitude chosen is high enough, then upon approaching the summer solstice, the night (sun 18° below the horizon) will slowly decrease in length and reach zero as the lower culmination of the sun rises to and reaches 18° below the horizon then astronomical twilight "all night" occurs.

Astronomical twilight will then slowly decrease in length and reach zero as the lower culmination of the sun rises to and reaches 12° below the horizon, then nautical twilight "all night" occurs.

Nautical twilight will then slowly decrease in length and reach zero as the lower culmination of the sun rises to and reaches 6° below the horizon, then civil twilight "all night" occurs.

Civil twilight will then slowly decrease in length and reach zero as the lower culmination of the sun rises to and reaches the horizon, then the midnight sun occurs.

Each twilight will only start to decrease once the lower culmination of the sun has reached its lower limit. The reverse happens upon leaving the summer solstice.

"The graph in the length section is definitely relevant to this article, it has just been badly worded. The original description that goes with the picture explains it much clearer. It should say continuous technical civil twilight or continuous daylight rather than continuous civil twilight"

The only problem that occurs to me here is that the text is wrong, which is a minor problem. Suppose the text in the graph said continuous daylight instead of continuous civil twilight and no daylight instead of no civil twilight. Would it then be worthy of remaining in the article? I would think so because the graph points out an interesting phenomena resulting from civil twilight, which is that the latitudes 60° 33’ 43” and 72° 33’ 43” can experience 24 hour daylight or continuous technical civil twilight at the summer solstice, which is equivalent to the presence of sunlight all day, and not experience 24 hour darkness or civil polar night at the winter solstice.

I think I've explained everything as clearly as possible. Hope this helps. Set Sail For The Seven Seas 340° 18' 0" NET 22:41, 20 September 2009 (UTC)

I was with you until here:

3. Technical Twilight is Twilight plus Daytime.

How do we know that? I haven't found an independent source on the web which says so. Have you? This is now at least the fifth time I have asked for a reference.

Who says this?

The situation here is that there are two terms, twilight and technical twilight, two definitions, without daytime and with daytime

There is this:

the definition given in the picture

But the picture was made by one who was under the impression that astronomers defined twilight such that it continued beyond sunrise. Where do we find evidence that a significant number of astronomers use that definition?

As for this:

at the point that the morning and evening twilights merge into one, morning twilight is suddenly included in an instant.

There ceases to be a distinction between morning and evening twilights in an instant. It remains twilight because the solar altitude remains above the twilight limit. The graph would need to be labeled carefully. Explaining the sudden jump to 24-hour twilight may be easier than convincing the general reader that it is their job to divide by two if they want to know the length of evening civil twilight at a moderate latitude.

The jump to 24-hour twilight would not be alone. There would be another where it goes from 24-hour twilight and 0 hours of day to 24-hour day and 0 hours of twilight because the sun emerged from below the horizon and will stay above the horizon for weeks.

As for this:

I get the idea that you are inclined towards wiping out all traces of technical twilight and sticking to twilight.

That's exactly right. I thought I understood what "twilight" meant before I adopted the erroneous notion of "technical twilight". It seems to me that the confusion was entirely of Wikipedia's making. If so, the article should not continue to promote that confusion. The length of this present discussion will probably drive the current version of this talk page into an archive before long, so the casual reader won't be as likely to stumble on the confusing discussion here.

As for this:

I would agree with it, but my only concern with this is that some readers will still think of twilight being with daytime

How so? Other than the two misleading images in the article and the previous discussion on this talk page, what would lead a reader to conclude that?

I think we need more than this:

the occurring of the various twilights passing through local midnight is based on the lowest point of the sun or the lower culmination of the sun.

If it is going to be a particular kind of twilight "all night", the sun needs to stay within a certain range of solar altitudes "all night".

I'll try to reply to this:

Suppose the text in the graph said continuous daylight instead of continuous civil twilight and no daylight instead of no civil twilight. Would it then be worthy of remaining in the article? I would think so because the graph points out an interesting phenomena resulting from civil twilight, which is that the latitudes 60° 33’ 43” and 72° 33’ 43” can experience 24 hour daylight or continuous technical civil twilight at the summer solstice, which is equivalent to the presence of sunlight all day, and not experience 24 hour darkness or civil polar night at the winter solstice.

The graph at day length shows "24 hour daylight ... at the summer solstice". The difference between that image and the one here is that the image here includes the lengths of morning and evening twilight. I don't believe that the sum is of general interest. To use a Wiki-word: How is the sum of twilight and day durations "notable"?

I don't disparage the graph as being "unworthy". It is a nice image. I just don't think it illustrates something that brings clarity to this article. - Ac44ck (talk) 01:12, 21 September 2009 (UTC)

I think of technical twilight as just a computerised version of the term twilight, a term used to help disambiguate between the two. There wouldn't be any references to it as it is pretty much a dummy term; that said, it does help with the understanding of twilight quite significantly.

"It seems to me that the confusion was entirely of Wikipedia's making"

I was thinking that the confusion was started due badly written external sources, hence the reasoning of keeping technical twilight, since AdeBarkah quoted external sources when the arguement was made. So it stands to reason that another reader could easily follow in AdeBarkah's footsteps saying that the definitions are wrong. Now, I was thinking that we could explain to this reader that their definitions were actually technical twilight, tell them that this is a computerised definition and then explain to them the actual definition of twilight. On the other hand, AdeBarkah misread the source article, so it stands to reason that it could be an error on the reader's behalf. If the confusion is indeed solely of Wikipedia's making, then I will agree with the wiping out of technical twilight, but there should be a mention of disambiguation. OK, I agree that an entire section on technical twilight is complete overkill, so I'm thinking a small sentence at the top along the lines of, "not to be confused with technical twilight, which is ..." or something similar, that way it makes the definition of twilight completely crystal clear with no disambiguation.

"There ceases to be a distinction between morning and evening twilights in an instant. It remains twilight because the solar altitude remains above the twilight limit. The graph would need to be labeled carefully. Explaining the sudden jump to 24-hour twilight may be easier than convincing the general reader that it is their job to divide by two if they want to know the length of evening civil twilight at a moderate latitude. The jump to 24-hour twilight would not be alone. There would be another where it goes from 24-hour twilight and 0 hours of day to 24-hour day and 0 hours of twilight because the sun emerged from below the horizon and will stay above the horizon for weeks."

I think I can see what is going on here. You're referring to the extremely high latitudes, specifically the poles, while I'm thinking of latitudes around those mentioned in the length section. At the poles, I will completely agree, it is true that there is this extremity at the various transition points, but I believe the only living at and around the vicinity of the poles are those at the Amundsen-Scott South Pole Station, and I doubt they would be reading this Wikipedia article! Mind you, it is a point of interest and so should be shown. Another thing, the scenario where the morning and evening twilights merge into one doesn't occur at the poles and so is irrelevant for that latitude.

"The occurring of the various twilights passing through local midnight is based on the lowest point of the sun or the lower culmination of the sun with the highest point of the sun or the upper culmination of the sun being above the horizon."

This is what it should have been considering the latitudes I was referring to. Of course, at the higher latitudes where polar night exists, then there will be no twilight, then a quick but smooth increase, then a slow decrease, then a slow increase, then a quick but smooth decrease until there is no twilight again. As the latitude increases, the changes in twilight duration will get faster and faster until they become instantaneous at the poles.

"Such a graph would probably show the length of twilight at the summer solstice increasing slowly with latitude, until the maximum extent of civil twilight continuing through local midnight, where it would decrease quickly and reach zero at the polar circle. It would also show the length of twilight at the winter solstice increasing slowly with latitude, until the polar circle, where it would decrease quickly and reach zero at the maximum extent of civil polar night."

This bit explains what happens at the solstices as the latitude varies.

"Suppose the text in the graph said continuous daylight instead of continuous civil twilight and no daylight instead of no civil twilight. Would it then be worthy of remaining in the article? I would think so because the graph points out an interesting phenomena resulting from civil twilight, which is that the latitudes 60° 33’ 43” and 72° 33’ 43” can experience 24 hour daylight or continuous technical civil twilight at the summer solstice, which is equivalent to the presence of sunlight all day, and not experience 24 hour darkness or civil polar night at the winter solstice."

I think the actual point here is not the inclusion of the graph, but about the inclusion of the interesting phenomena resulting from civil twilight that is shown in the graph. So maybe if a section on this phenomena is added to the article, the graph can be added there; if not though, the article should still mention the phenomena, as it is solely based on twilight. Hope this helps. Set Sail For The Seven Seas 264° 4' 15" NET 17:36, 21 September 2009 (UTC)

The term "technical twilight" was coined by a Wikipedia editor as inclusive shorthand for the various kinds of twilight, based on a misunderstanding of their definitions. It was not meant to take on its own life as a formal definition. It is not a verifiable term and, rather than be elevated to a formal definition, should be removed from the article wherever it occurs or is implied, including the graph which shows the variation in its length.

Enabling an inaccurate reading of an easy-to-misinterpret external source would seem to be a disservice. Let's be crystal clear about the definitions. The image which contains the label "technical twilight" doesn't go far enough. This Twilight article is in the category "Parts of a day". So let's show all the mutually exclusive parts of a day in the diagram: Day, Night, and six periods of twilight without overlap.

There isn't a need to economize on words. The wording in the external source is too terse. We don't need to make the same mistake here. Einstein: "Things should be as simple as possible, but not simpler."

Perhaps elements of the following could be built upon for wording that is intended to be purposely unambiguous:

Civil Twilight: A period during which the sun is below the horizon but not more than 6° below the horizon. Generally, there are two periods of civil twilight every diurnal period. Exceptions occur near the polar regions where there may be one or no period of twilight during an equatorial diurnal period at the same longitude. Where context does not make obvious the occurrence of twilight in mind, one might refer to "morning civil twilight" or "evening civil twilight". For conversational purposes, one may usually omit the distinction "civil" twilight, as unspecified references to twilight generally have civil twilight in mind.

Nautical Twilight: A period during which the sun is more than 6°F below the horizon but not more than 12° below the horizon. Generally, there are two periods of nautical twilight every diurnal period. ...

The limits for the various forms of twilight, in terms of solar altitude measured to the center of the sun, are as follows:

Type of twilight         Upper twilight limit          Lower twilight limit
Civil twilight           < -0°50' (sunrise/sunset)     <= -6°
Nautical twilight        < -6                          <= -12
Astronomical twilight    < -12                         <= -18

It occurs to me that the graph now shown in the article is pretty tame compared to what a graph showing the length of twilight would look like. I haven't fully developed a mental picture of the new graph, but I suspect that a region of 24-hour twilight would appear as a sliver-shaped arc of varying thickness. I think the new graph would display a greater number of interesting features than are found in the current graph. - Ac44ck (talk) 04:25, 22 September 2009 (UTC)

Has anyone considered that "X technical twilight, xx.x hrs" is shorthand for "X twilight or worse for xx.x hours"? A length of evening twilight graph should max out at 12 hours (with a note to explain this) to keep the (nonexistant?) morning graph a mirror image of the evening. You can't start assigning all eternal twilight hours to evening just cause you want it to show 24, otherwise you can get the ridiculous notion that evening technical twilight occurs simulaneously with sunrise. Now what should be the cutoffs, mean local midnight and noon or actual? Sagittarian Milky Way (talk) 07:29, 22 September 2009 (UTC)

I agree with the limits for the various forms of twilight, though I'd probably think of it more like this.

Definition	Position of sun

Night	more than 18°
Astronomical twilight	12°<AT<18°
Nautical twilight	6°<NT<12°
Civil twilight	-0°50'<CT<6°
Day	sun above the horizon

Also, the boundaries are marked by the various dawns and dusks or sunrise and sunset.

Those are exactly correct. When I see 18 hours, I know what it means. It doesn't bother me at all. What can I say, I have a very flexible brain.

But for people without an intuitive knowledge of twilight this would be very misleading. It just needs a retitiling. "Hours of usable natural light". This would be useful in gauging how bright a night is at summer. Even so, reversing the hours might be slightly more useful for showing number of observable hours -- it would tell me that at Anchorage where many people live (for example), the sky would be bright enough to play baseball for pretty much the entire night, which is, needless to say, annoying. Sagittarian Milky Way (talk) 01:57, 24 September 2009 (UTC)

This shows I've been brought up on hourglass diagrams too much – it says there's 15 hours of twilight a day in Middle America and I don't bat an eye.. :) Sagittarian Milky Way (talk) 01:57, 24 September 2009 (UTC)

I think at this point I can agree with removing technical twilight, as things should indeed be a simple as possible, but not simpler, otherwise we end up with a spherical cow!

One other thing about the graph, it should show the total amount of twilight per day, that's why morning twilight and evening twilight should be added up, as has been done with the existing graph. Also, I think Sagittarian Milky Way has a good point about this. Both twilights should be included otherwise it is not the total amount of twilight. Hope this helps. Set Sail For The Seven Seas 243° 43' 15" NET 16:14, 22 September 2009 (UTC)

Many people would like to know how long it takes to get dark without having to divide by 2. I have an idea, make it like a population pyramid. Morning twilight on the left, evening on right. Upper scale on the bottom counts 0 to almost 12 from the center in both directions. What would be 12 instead says all day. Lower bottom scale is the same thing but double speed so it shows total twilight. Possibly put that scale under only one of the halves to avoid confusion. Double-arrowed line covering the top says 24 hours. I suggest we use the dark edge of nautical twilight if we can only have one. It is close to when it gets dark towards the sun in cities. You'll just have to interpolate if you want to know how long you can read a newspaper after dusk. No need for legal fictions here ("evenly dividing twilight hours and assigning them to evening and morning when those terms have become meaningless") Sagittarian Milky Way (talk) 01:57, 24 September 2009 (UTC)

So the length of evening twilight should max out at 12 hours, "otherwise we end up with a spherical cow." But we don't have to find the cutoff because there is no cutoff; that's why it's 24-hour twilight.

Nature has other discontinuities:

And last but not least, critical point; which may be analogous to merging twilights:

a critical point ... specifies the conditions .. at which a ... boundary ceases to exist.

Why must we obscure a discontinuity related to twilight? Ac44ck (talk) 03:47, 23 September 2009 (UTC)

I don't think it's about obscuring a discontinuity, but about adding a discontinuity where it doesn't exist. At the latitudes mentioned in the length section, the points where the twilights merge is just a local maximum not a discontinuity, but at the poles there is definitely a discontinuity and it should be recognised. The locations of discontinuities should be in the same place as those on the current graph. A new graph of twilight should really have all twilight accounted for somewhere, after all it's best to show the total twilight on a daily basis at varying latitudes. If someone wanted to find the length of twilight, looking at a graph would not be of best interest and greatest accuracy. They would only need to find the start and end time of the twilight in question and find the difference. Also, they'd only be interested in the twilight length at a particular time and place. In this situation they can be directed to the many external links at the bottom of the page, which have calculators that give start and end times of twilight. One other thing, when you say 24-hour twilight, you are referring to the phenomena that happens at latitudes greater than 72°, during the polar night, right? Also, as the latitude increases, the changes in twilight duration will get faster and faster until they become instantaneous at the poles, so we have a build up to the discontinuity. Hope this helps. Set Sail For The Seven Seas 216° 6' 0" NET 14:24, 23 September 2009 (UTC)

At the poles twilight length discontinuities are instant. The sun circles around and around and around and finally it pops under the horizon and goes from 0 twilight to 24 hours a day in an instant. If you follow the latitudes in the chart polewards the maximal rate of change gets faster and faster, I think this is what you were talking about when you were describing the asymptote. Although if one wanted to be very picky, on whatever year you choose there should be some odd and pretty much randomly useless number of hours of twilight (like 6.2) at each pole on a few key days. We don't need to needlessly obscure the idea of sudden turning off and turning on of twilight at the poles like a lightbulb. So a smooth average of contemporary years should be used, without horizontal lines every day to hide the fact that this is not quantitized. Since the 90° line is so different from the others, should an inner halo be given to all "isocrepuscules" for the advancement of making it more intuitive which side is which on the head-scratching ones? The contour intervals should get smaller where not inconvenient, due to the very rapidly changing spacing in that area. I've seen one, on sky and telescope's website (it only shows one twilight) and spacing changes so quick as low as the 40-50 gap I wish they had a 45° line to help interpolation. Sagittarian Milky Way (talk) 03:24, 24 September 2009 (UTC)

@First Comment - I agree as per previous comments that a graph showing number of observable hours would be slightly more useful to a reader, in fact the current graph in the length section does just that and also it points out an interesting phenomena resulting from civil twilight, which is that the latitudes 60° 33’ 43” and 72° 33’ 43” can experience 24 hour daylight or continuous technical civil twilight at the summer solstice, which is equivalent to the presence of sunlight all day, and not experience 24 hour darkness or civil polar night at the winter solstice, but it needs renaming as it is not continuous civil twilight as it states but continuous technical civil twilight or continuous daylight, then it would make more sense.

I was meaning astronomically observable hours, but for the public vs very slight convenience for astronomers, I agree we should just change all chart instances of civil twilight to something correct. Sagittarian Milky Way (talk) 01:30, 25 September 2009 (UTC)

@Second Comment - A population pyramid is actually quite a good idea, originally the discussion about creating a graph that would show only evening twilight but would suddenly include morning twilight when they merge at the point where the twilight continued through local midnight causing a jump as the length of twilight suddenly doubles intantaneously at the latitudes mentioned in the length section, which I disagreed with, after all, if morning twilight is to be included, it must be completely included not suddenly included.

@Third Comment - It is definitely true that twilight at the poles turns on and off like a lightbulb, going from 24 hours twilight to 0 hours twilight and from 0 hours twilight to 24 hours twilight instantaneously. However, this is not true for the latitudes being referred to in the previous comment, these being the latitudes mentioned in the length section. Here is how I explained it in a previous comment.

The occurring of the various twilights passing through local midnight is based on the lowest point of the sun or the lower culmination of the sun. This lower culmination will gradually sink upon moving away from the summer solstice and gradually rise upon moving towards from the summer solstice, with the lower culmination of the sun being at its highest at the summer solstice.

If the latitude chosen is high enough, then upon approaching the summer solstice, the night (sun 18° below the horizon) will slowly decrease in length and reach zero as the lower culmination of the sun rises to and reaches 18° below the horizon then astronomical twilight "all night" occurs.

Astronomical twilight will then slowly decrease in length and reach zero as the lower culmination of the sun rises to and reaches 12° below the horizon, then nautical twilight "all night" occurs.

Nautical twilight will then slowly decrease in length and reach zero as the lower culmination of the sun rises to and reaches 6° below the horizon, then civil twilight "all night" occurs.

Civil twilight will then slowly decrease in length and reach zero as the lower culmination of the sun rises to and reaches the horizon, then the midnight sun occurs.

Each twilight will only start to decrease once the lower culmination of the sun has reached its lower limit, before this it will increas at a rate slower than the upcoming decrease. The reverse happens upon leaving the summer solstice.

Forgive me for skimming, not remembering or just plain skipping parts - I try to read it all but my eyes glaze over after so much nitpicking with 0, 24, and the same latitudes, repeated.. :) Sagittarian Milky Way (talk) 01:30, 25 September 2009 (UTC)

I think that pretty much covers it. Hope this helps. Set Sail For The Seven Seas 354° 41' 30" NET 23:38, 24 September 2009 (UTC)

"I was meaning astronomically observable hours, but for the public vs very slight convenience for astronomers, I agree we should just change all chart instances of civil twilight to something correct."

Well, I found an SVG version of the chart and I'm in the process of doing just that. Well, when I say doing just that ... :).

"Forgive me for skimming, not remembering or just plain skipping parts - I try to read it all but my eyes glaze over after so much nitpicking with 0, 24, and the same latitudes, repeated."

That's fine. Mind you it took a bit of searching just to actually find it! As long as you understand the concept after reading through it this time round. I admit it is quite wordy, but I believe it is extremely clear and should be easy to understand. Hope this helps. Set Sail For The Seven Seas 355° 37' 0" NET 23:42, 25 September 2009 (UTC)