Talk:Lunar phase

From Wikipedia, the free encyclopedia

	This article is supported by the Moon WikiProject. This project provides a central approach to Moon-related subjects on Wikipedia. Please participate by editing the article, and help us improve articles to good and 1.0 standards, or visit the wikiproject page for more details.
B	This article has been rated as B-Class on the quality scale.
Top	This article has been rated as Top-priority on the priority scale.
Comments: edit – history – watch – refresh Might contain original research (uncited formula)

	To-do list for Lunar phase:	edit  · history  · watch  · refresh
Add section describing the orientation of the Moon as viewed by an observer on Earth.

Contents 1 Irregular time between moon phases? 2 Lunar Fortnight? 3 Southern Hemisphere difference in moon phases an error? 4 Accuracy of phase calculation? 5 Contradiction with Eclipse 6 Daylight Saving Time for table of Lunar phases' 2005-2020? 7 Russian mnemonic 8 Expansion of Earth-shadow misunderstanding common to society 9 Vandalism 10 Merge comments 11 Proposed deletion of "Patent" section 12 Capitalization of New and Full Moon 13 Orientation of the Moon 14 Question 15 Another question 16 The First Picture 17 Proposed deletion of Lunar phase equation

[edit] Irregular time between moon phases?

The article states that a complete moon phase cycle takes 29.53 days. But, according to the moon phase tables (for example): Between Full on Jan 14 09:48 UTC and Last Quarter on Jan 22 15:14 UTC there are 8 days, 5 hours and 26 minutes. However, between Full on June 11 18:03 UTC and Last quarter on June 18 14:08 UTC there are 6 days, 20 hours and 5 minutes. (calculated with [1]) Where does that difference come from? Is this due to libration? Is it consistent with a stated regular moon cycle of 29.53 days? -Tom-b 00:38, 15 May 2006 (UTC)

UPDATE: I just bought this answer at Google, feel free to add it to the article [2]

[edit] Lunar Fortnight?

Would be nice if somebody could define lunar fortnight such as used on the fasting page Togo 05:17, 20 May 2004 (UTC)

what lunar really means?

[edit] Southern Hemisphere difference in moon phases an error?

Is it true that "In the southern hemisphere, the above is reversed" regarding left and right sides of the moon? How could that possibly make a difference? -Unknown

I believe it is correct, because they are looking at it from a different viewpoint - they are "upside-down". While the moon seems to orbit in a counter-clockwise fashion when looked on from a point above the North Pole, it will appear to rotate in a clock-wise fashion when looked on from a point above (or below, however you want to think about it) the South pole. Also refer to: [3] -Unknown

[edit] Accuracy of phase calculation?

I'm having trouble getting the lunar phase calculation at the bottom working. I think partly because of the terms it is written in. They seem cryptic to me. I also question the accuracy of the formula, as I found an alternative online and it was many times more complex. Perhaps this one will only work between such and such years? This is unclear, again as I said, as is the way to use the formula. For these reasons I suggest the formula be removed until someone more mathematically able than I has the time to fix it, or add more explanation for it's use. Weatherlawyer 21:35, 19 January 2006 (UTC)

The calculation does not take account of short term variations of the moon phase, which can put the calculated moon phase up to 15 hours out. Karl Palmen 09:30, 20 January 2006 (UTC)

It is horribly vague. Unless fraction() is a special function, or unless there are some missing parenthesis, I don't see how ".2 + x * .03" could ever yield anything anywhere near a new moon. (which would be 0) Can anyone clarify, or can I just angrily rip this equation out and vehemently shred it into figurative bits? --tyroney 18:39, 13 August 2006 (UTC)

The phase calculation appears to assume a circular orbit since it is a linear function. The fraction function is: frac(x)=x-int(x), that is always between zero and one. I have better a formula from a book Astronomical formulaee for calculators, 4th edition, 1988, by Jean Meeus. (Chapter 32). However it is much more complicated. ANYWAY, I've not tested the formula given here, and there's no apparent source reference (except link to author's website?), so I have no problem with removal. Tom Ruen 21:11, 13 August 2006 (UTC)

[edit] Contradiction with Eclipse

In the eclipse article, it is stated:

• Lunar eclipses - the Earth obscures the Sun, from the Moon's point of view. The Moon moves through the shadow cast by the Earth. This can only happen at full moon.
• Solar eclipses - the Moon occults the Sun, from the Earth's point of view. The Moon casts a shadow that touches the surface of the Earth. This can only happen at new moon.

However, in this article, the opposite appears to have been said:

"Note that the plane of the Moon's orbit around the Earth is tilted by about 5 degrees with respect to the Earth's orbit around the Sun. Therefore, eclipses of the Moon during the full Moon and of the Earth by a new Moon are rare and usually newsworthy."

I believe what is being said is that it's rare to happen during a perfectly full or new moon, but I don't know enough about the subject to tell which is correct--someone who does should make them match up. Chris 01:24, 13 May 2005 (UTC)

Clarified this somewhat. Johantheghost 09:48, 19 September 2005 (UTC)

[edit] Daylight Saving Time for table of Lunar phases' 2005-2020?

Seems to me that if UT is to be used for this table, daylight saving time should be ignored. I'd like to fix this - any agreement? -Unknown

[edit] Russian mnemonic

On 7 July 2005, 67.138.153.217 added this statement to the Mnemonics section:

In Russian, the 'C' stands for "Стареющая" or "[a moon] becoming old", while a line is added to the waning crescent to form 'P', which stands for "Рождающаяся" or "[a moon] getting born"

That statement remained in the article until changed on 15 December 2005 by Flume to:

In Russia, the 'C' stands for "Стареющая" or "[a moon] becoming old", while a line is added to the waxing crescent to form 'P', which stands for "Рождающаяся" or "[a moon] getting born"

That edit changed waning to waxing. Can anyone cite an independent source supporting either the original statement or the change? --DannyZ 05:48, 16 December 2005 (UTC)

If I read the paragraph correctly, "[a moon] becoming old" means a waning moon, and "[a moon] getting born" would be the waxing moon, just coming out of the new moon state, therefore the phrase "while a line is added to the waning crescent to form 'P'" is wrong and correctly changed to waxing.

In my little russian knowledge, "Стареющая" means aging and "Рождающаяся" means emerging --RobiBuecheler 06:08, 10 February 2007 (UTC)

[edit] Expansion of Earth-shadow misunderstanding common to society

As illustrated in this comic: a lot of people are probably telling their kids the wrong thing, and believe it themselves. I actually realized that I had misunderstood the cause too! Tyciol 07:12, 26 May 2006 (UTC)

[edit] Vandalism

I wonder why this page is a semi-regular target for vandalism? AnonMoos 00:53, 2 November 2006 (UTC)

[edit] Proposed deletion of "Patent" section

I propose to delete the "Patent section." Perhaps someone can explain why this should be here? Lunokhod 11:07, 23 November 2006 (UTC)

TP: it is ... peculiar. I think it is potentially useful information; I suppose it could find a place under External Links. Tom Peters 11:31, 23 November 2006 (UTC)

[edit] Capitalization of New and Full Moon

Does anyone here know if "new moon" and "full moon" should be capitalized? Obviously "the Moon" is capitalized, and a generic "moon" is not. Perhaps both are acceptable? And if it is capitalized, is it "full Moon" or "Full Moon"? My guess is since these do not need to be preceded by "the" that they are not capitalized.Lunokhod 20:34, 28 November 2006 (UTC)

I don't know if they "should" be capitalized or not. I would be dismayed if there are rules that dictate such details. I always capitalize out of reverence for all the things upthere. Tom Peters 00:28, 29 November 2006 (UTC)

The Journal for the History of Astronomy specifically requires that when they are discussed as astronomical bodies they must be capitalized (Moon, Sun), but when their appearance in the sky is discussed they must not be capitalized (new moon, full moon, etc. JHA style sheet quote: "Sun, Moon, Earth if treated as astronomical bodies; sun, moon, earth if essentially appearances in the sky." — Joe Kress 07:03, 29 November 2006 (UTC)

[edit] Orientation of the Moon

I think that this page needs a section describing the "orientation" of the Moon, and how it depends on the day of the month and observation latitude. Does anyone want to take a stab at writing this? Lunokhod 09:57, 29 November 2006 (UTC)

I agree: apparently, the closer to the equator, the more the crescent resembles a smile (or, the U in the classic OUO "Diana" symbol) while the farther from the equator the more the crescent resembles a C or D. I have never quite been able to understand the geometry involved - naively, the Earth-Moon distance should totally overwhelm any observer-equator distance, and the angle of the crescent should depend only on the inclination of the lunar orbit. Apparently, this is not the case, and I'd like to understand why. (Yes, that's what brought me to this page.) --Jon Shemitz 01:53, 22 January 2007 (UTC)

[edit] Question

Could someone explain to me, or clarify in the text, what this means?

"A first-quarter moon follows a daily path in the sky corresponding to that of the Sun after three months"

The rest of the paragraph is confusing as well. Lunokhod 23:31, 30 November 2006 (UTC)

If we neglect the 5° inclination of the Moon's orbit to the ecliptic and consider that there are three months in a quarter year, then the first quarter moon is at a position on the ecliptic that the Sun will be at three months later. At some point during each nychthemeron (night or day) the moon will cross the observer's meridian, when it reaches its greatest altitude above the horizon for that day. This will be above the southern horizon for northern hemisphere observers, and will be above the northern horizon for southern hemisphere observers. The ecliptic is inclined to the equator by about 23.5°, so regardless of the Sun's position on it, that daily altitude will cycle once each lunation from a high angle ("moon rides high") to a low angle ("moon runs low") and back again. The lunar phases when these angles occur depend on the month of the year, or, more specifically, they depend on the lunation in the tropical year. The paragraph only discusses the phases of the moon at the equinoxes and solstices. The terms "moon rides high" and "moon runs low" are each noted once a month on some day in the Old Farmer's Almanac. A Google search indicates that "moon rides high" is far more popular than "moons runs high", "moon rides low", or "moon runs low". — Joe Kress 02:55, 1 December 2006 (UTC)

[edit] Another question

Could someone explain the rational for doing this?

"If you photographed the Moon's phase every day for a month, starting in the evening at sunset, and repeating approximately 50 minutes later each successive day, you could create a composite image like the example calendar below from May 8, 2005 to June 6, 2005."

I am a bit confused as to how long the "month" is supposed to be here. I think the idea is to take a picture of the Moon every 1/30.5 days, but this is not too clear (to me). Lunokhod 23:42, 30 November 2006 (UTC)

A daily lunar photo will show the moon cycling through all lunar phases during each lunation. A synodic month lasts about 29.5 days, so each successive day the photo will have to be taken 1/29.5 days or 49 minutes later to equally distribute the phases. Each photo is 1.034 days later than the last. Although multiplying 1.034 × 29.5 does produce 30.5, that is the number of circuits the moon takes each lunation from the perspective of an observer on a rotating earth, one for each day plus one for the lunation. Remember the embarrasing SAT question (from the College Board's point of view) which asked how many times does a small circle rotate as it rolls once around a large circle (ratio of radii 1:3). The correct answer of 4 was not among the multiple choices! [4] — Joe Kress 05:02, 1 December 2006 (UTC)

Joe, I think you are confused here. You can take a picture of the Moon as frequently as you want, and get a nice film as long as you do it regularly and cover the full synodic month. But every solar day the Moon has moved ahead, i.e. Eastward (counter-clockwise as seen on Northern hemisphere) w.r.t. the Sun. The (mean) Sun passes the meridian every 24 hours by definition. The Moon will pass the meridian 49 minutes later each day on average. So if you photograph the Moon repeatedly 1 day + 49 min apart during a lunation, you will capture it at about the same hour angle. However, as its declination varies by the full range during a month, the Moon will be at different positions in the sky and at different altitudes above the horizon. Also 1.034 days does not divide 29.5 so you would not take an integer number of pictures in a synodic month. In any case, the Moon does not make 30.5 circuits w.r.t. an observer on the surface of the Earth in 1 synodic month, and not even 29.5, but rather 28.5 . Tom Peters 12:12, 1 December 2006 (UTC)

The image caption for the phase calendar says that the images are taken 25 minutes later each day. Is this a mistake? Also could you explain where the 28.5 days comes from (and why it is not 29.5) and update the main article accordingly? This subject is obviously causing some confusion and needs to be clarified. Perhaps it would be worthwhile describing this in terms of lunar zenith to lunar zenith times (assuming the obliquity of the Earth and lunar orbit inclination were zero). A table of such times might prove useful. Lunokhod 14:14, 2 December 2006 (UTC)

Missed your comment, so late response. I suppose that the lunar phase calendar was generated as specified, i.e. 1d25m apart as stated. That makes sense. If you try to photograph the Moon every 24h it will be under the horizon in half of the cases. If you photograph it every 24h50m it will (approximately) in the meridian (or other specific hour angle) each time, but in half of the cases during day light so often invisible. If you take them 24h25m apart, starting from a new crescent Moon over the Western horizon just after sunset, then you'll catch a Full Moon around midnight in the meridian, and an old crescent Moon over the eastern horizon just before sunrise: i.e. you photograph at the most appropriate moments during the night only. On the 28.5: as stated, that is approximately the number of CIRCUITS that the Moon makes w.r.t. an observer on the surface of the Earth during the 29.5 DAYS that a lunation takes. In any case it is not 30.5 as Joe Kress stated. The Moon takes longer than a day between merdian passages, not shorter. Tom Peters 12:34, 14 December 2006 (UTC)

[edit] The First Picture

What exactly is the first picture supposed to be? It shows the moon with the words '2nd quarter' and the value '72%' below it. According to the lunar phase chart on here, we're between the first quarter and full moon, so the picture doesn't represent the current moon phase. What is its purpose?--Jcvamp 16:29, 30 December 2006 (UTC)

Actually, the Moon did reach the 72% illumination mark yesterday, so it's not that far off. What I don't like is that it uses "2nd quarter" to mean something different than it does in the article, apparently meaning between 50 and 100% lit. Saros136 16:59, 30 December 2006 (UTC)

The thing that threw me off was that it says '2nd quarter' when it fact it's between the first quarter and the full moon. It's the text describing that moon as '2nd quarter' that needs to be changed.--Jcvamp 08:48, 31 December 2006 (UTC)

That's called a "gibbous Moon" (which may be waxing or waning). There is no such thing as a "2nd quarter". There are first and last quarters, and a Full Moon in between. Tom Peters 14:39, 31 December 2006 (UTC)

I know it's called a gibbous, (in fact it's a waxing gibbous at the moment) but I was discussing the phase in terms of quarters as it mentions the phase as the '2nd quarter' on the page. I also know that it isn't usually called the second quarter, but that's what it says erroneously (which is the point I'm making here) on the page.--Jcvamp 14:20, 1 January 2007 (UTC)

So complain at User:Dbachmann who provides the Wikiprogram that draws this.

Is the picture the same whereever you are in the world, surely the moon looks difrent depending on your latitude ? —The preceding unsigned comment was added by 130.36.87.102 (talk • contribs).

No, it is affected by horizontal parallax (also, besides your latitude, depends on your longitude and the time): it is about 1 deg., so not noticeable at the scale of the picture. Tom Peters 16:27, 24 January 2007 (UTC)

[edit] Proposed deletion of Lunar phase equation

I do not find the equation that predicts the lunar phase as a function of time useful, and propose it for deletion. In particular, it would be much easier to go to any of the numerous online lunar phase calculators to determine this, than to determine how many days have passed since 12AM, January 1, 2001 (Greenwich time). Perhaps we should just give a reference on where to find this equation for those of us who might want to calculate this quantity. Lunokhod 20:53, 30 December 2006 (UTC)

You seem to really hate numbers (considering your similar proposals in New moon, Full moon, and Full moon cycle. You are correct that there are many places on-line where one can find the current phase of the Moon. There are several links in the References section. However, I don't see why such information should not also be included in the Wikipedia, which as an encyclopedia is a collection of knowledge taken from elsewhere. However I don't know of on-line resources that explain HOW to computate such results. This may be useful to check phases in the past or future. This particular formula appears to be taken from an almanac. Tom Peters 14:47, 31 December 2006 (UTC)

For the record, I'm a theoretical geophysicist, so I have nothing against equations! My concern with this one is the following: Only a very small percentage of people who read this page will ever want to calculate the lunar phase as a function of time. Of those specialists that would like to do this calculation, they will probably find this equation to be inadequate: I doubt that that the number of decimal places quoted is justified, and it is clear that this equation neglects the eccentricity of the lunar orbit. How one deals with changing time zones is not described, nor is the range of dates for which this equation is accurate. Furthermore, though not a criticism of the equation itself, a non-expert or expert could easily make an error concerning leap years, or mistakenly use year 2000 (which is the standard epoch for orbital calculations) instead of 2001, which is used in this formula. Lunokhod 16:31, 31 December 2006 (UTC)

OK, I would agree that the expression should use a standard epoch, be referenced, and have some caveats. But I dispute the notion of "little interest"; such simplified expressions are routinely given in almanacs. Wikipedia does not have the limits of a paper encyclopedia so I see no reason to not serve all audiences. Also such computational details seem not to be readily available elsewhere on-line, and I think it is completely within the purpose of Wikipedia to make such things available on-line. Tom Peters 09:57, 1 January 2007 (UTC)

It gives mean lunar phase (i.e. using a constant angular velocity) as a function of GMT or UTC or whatever the latest acronym is. An accurate equation for the moon's non-averaged position would take up multiple pages (it did take up multiple pages in the book that Laplace (I think it was) devoted to deriving it). AnonMoos 00:10, 1 January 2007 (UTC)

Yup, and an (obsolete) reference to Meeus to compute more precise phases follows. Do you have anything more specific on who derived which equations? The earliest I know are from Hansen (Leipzig 1858, 1863). Tom Peters 09:57, 1 January 2007 (UTC)

Could someone please at least add a reference to the equation in question so that I can verify it (the equation just changed by 12 hours if I understand the last edit correctly!)? Otherwise, in accordance with Wikipedia policy, I will delete it irregardless of the community's concensus on whether it is useful or not. Lunokhod 11:02, 1 January 2007 (UTC)

Ooops, sorry for not discussing it. 12 AM seemed ambiguous; I've seen it refer to midnight and noon. I thouhgt 0 ut was intended because it gave a better match to some dates of phases I tried. I'm not sure I was right, though-it might depend on the years. For 2001, 0 UT would have given a closer match at the times of full moons, 10 of 12 times, 9 of 12 1st quarters, 6 of 11 3rd qtrs, and it's a tie on new moons. I haven't done all of 2007, but the 0 UT base is better for new and full moons.

Similar equations were in all the annual Naval Observatory almanacs (though I haven't seen one in a while), and I assume somone changed it from a noon to midnight epoch (or vice versa) -- it doesn't really matter which one is used, as long as it's made clear. AnonMoos 12:41, 1 January 2007 (UTC)

The equation in question was added on 11 March 2005 by Monsieur le docteur Ralph without any attribution as follows:

phase = fraction(.20439731 + t * .03386319269),

where t=[UT]-[12AM, January 1, 2001], days

such that new moon=.0, first quarter=.25, full moon=.5, last quarter=.75

The equation is approximately correct for midnight. An equation close to that by Ralph can be obtained as follows:

JDE = 2451550.09765 + 29.530588853k + 0.000 1337T² − 0.000 000 150T³ + 0.000 000 000 73T⁴

where JDE is the Julian Ephemeris Date of the mean phase of the moon, k is the number of mean synodic months (including any fraction thereof) since the first mean new moon after J2000.0 (2000 January 6), and T is the number of Julian centuries since J2000.0. From Jean Meeus, Astronomical Algorithms, p.319. Meeus' warning that any value of k other than 0, 0.25, 0.50, or 0.75 is meaningless only applies if the equation is used to calculate the true phase of the moon.

1/29.530588853 = 0.03386319199

Julian Date of 2001 January 1 00:00 = 2451910.5

k = (2451910.5 − 2451550.09765)/29.530588853 = 12.204374

0.20439731 − 0.204374 = 0.00002331 ≈ 2 seconds

Because ΔT = TT − UT = 64.09s at 2001.0, a 2 second error implies that the time should be TT, not UT. The phase calculated via the following equation (with the given precision) is accurate to ±0.004 after three millennia when the small terms in T are ignored.

phase = (0.2044 + t×0.033 863 182) mod 1

where t is the number of days since 2001 January 1 00:00 TT, 0 ≤ phase < 1

— Joe Kress 06:23, 5 January 2007 (UTC)