User:ABTOP/draft3

ABTOP 11:09, 30 January 2007 (UTC)

New COMMENTS:  I want to delete the reasoning for the changes I want to make in WIKIPEDIA "EQUINOX". Really only the material listed at the end of this document is required to devise new statements for the Wikipedia Article.

''I have to state that the equivalent information on Russian Wikipedia is quite correct in the article called "Ravnodenstvia" ("literally "Equal Days" or the Equinox"). It states the effects of the refraction by the atmosphere - the latter is necessarily absent for the effects of the official equatorial Equinox!''

The current article in Wikipedia I have to describe reluctantly as a travesty, because both types of "Equinox" need to be separated.

The Russian Wikipedia article on "Equinox": (Cited):

@misc{ wiki:xxx, author = "Википедия",(Wikipedia) title = "Равноденствие --- Википедия{,} свободная энциклопедия",("The Equal Days Phenomenon") year = "2006", url = "http://ru.wikipedia.org/w/index.php?title=%D0%A0%D0%B0%D0%B2%D0%BD%D0%BE%D0%B4%D0%B5%D0%BD%D1%81%D1%82%D0%B2%D0%B8%D0%B5&oldid=2639248", note = "[Online; accessed 20-январь-2007]" (20 January, 2007) }                         == My Experience: ==

At 72 years of age, I have for many years since college days been very interested in the subject of Time and related longitude/latitude effects. Originally from London, I have been staying here continuously since June 2004.

Part of my education was in a Polytechnic specialising in Horology, the classes being divided into four competitive houses named after prominent experts such as John Harrison. He, of course, provided the World with the first chronometer to function accurately in rough seas, enabling longitude to be better determined. As the attached interview with the local newspaper “Viva Cadiz” mentions, longitude and latitude were heavily involved in my calculations of my own observations of earth satellite orbital periods – which I transmitted to the Royal Aircraft Establishment at Farnborough – and were acknowledged as being very accurate, although by means of very elementary facilities.

Also relevant for the current subject of the Solar Radiation Hazard in Spain, as you will see, is my Experimental High Postgraduate Degree on the subject of the intensities of spectral UV emissions of gases present in reentry experiments concerning earth satellites (Imperial College, University of London).

Relevant == Wikipedia Reference ==

''Page name: Equinox Author: Wikipedia contributors Publisher: Wikipedia, The Free Encyclopedia. Date of last revision: 30 December 2006 23:34 UTC Date retrieved: 4 January 2007 10:45 UTC Permanent link: http://en.wikipedia.org/w/index.php?title=Equinox&oldid=97425906 Page Version ID: 97425906'' OR

''@misc{ wiki:xxx,

author = "Wikipedia", title = "Equinox --- Wikipedia{,} The Free Encyclopedia", year = "2006", url = "http://en.wikipedia.org/w/index.php?title=Equinox&oldid=97425906", note = "[Online; accessed 4-January-2007]"''

== COMMENCE: ==  (Minimal Explanation not required in actual change of article)

(Article submitted to local newspaper predicting differing Equinox Date)

== 1.0 Problems with Interpretation of the Equinoxes ==

In comparison to the ideas which prevailed with the "Druids of Stonehenge, England" in the past, it does seem that modern astronomers live in a fantasy world. It would be a few such astronomers using observations of the Sun being visible exactly for 12 hours on the day of the Equinox, who would have to believe that there is no atmosphere around the Earth to produce the mirage phenomena that we often see at sunset (the Sun can be seen when it is actually below the horizon, because the atmosphere acts like a lens, and refracts that image just above the horizon for some time). To us, the Sun will appear to be longer than 12 hours above the horizon on the 23rd of September, so that we will experience the Equinox 1 or 2 days later - and that will show up over the Weather Maps in the newspapers for that time! (Equal clock positions for both sunrise and sunset will indicate the twelve hours exactly, or perhaps closely enough to be significant).

To be fair, astronomers no longer observe the actual length of the day at Equinox, they are only interested in the exact time that the Sun is located at zenith over the Equator, and that will be at a particular place, too! According to information from Madrid which appeared locally on June 22nd, 2006, that time will be at 06:03 am Spanish Time, or 04:03 Z (UTC, GMT) on the 23rd September. That puts the location under the Sun near Kasimbar, on the island of Celebes in the Pacific.

But, "Hold on!", the Earth-Sun clock is running slower than the atomic clocks used by the astronomers at that date, by about 7 minutes! Thus, on a local sun clock at Kasimbar, the time of the Equinox will be shown as about 11:53 am in the time of that locality and not noon, as required! So, to where should the astronomers actually take their boat in the sea, in order to observe the actual event? That is a puzzle for me that will take more than 7 minutes to investigate!

== Observations in CADIZ, Spain: ==

My own investigations of the past week, using sunrise/sunset times at Gibraltar in the local Spanish language newspaper for that Metropolitan District ("El Faro"), show consistently that the times above the Weather Map average out to 1 minute 20 seconds LATER than the times shown for the same dates in most of the Diario newspapers printed in the Province of Cadiz! An explanation for the Sun to arrive earlier at Cadiz puts the source of such information somewhere in the location of San Pedro on the Costa del Sol, by the Mediterranean. As I stated above, the times for sunrise/sunset will look the same on the clocks about the 25th of September, when I should be able to attach proof from the newspapers here for that !

END of Newspaper Article

1.2  Print out of a proof (in Spanish), accompanied by newspaper copy of their own proof (Sunrise/sunset times on 26th September, 2006.

== La  C A Z A   el   “S N A R K” ==

== Lewis Carroll == ( Hunting  in  a  rowing boat  for  the  elusive  mythical  creature! ) Snark =    equinoccio  de  2,006  -  encima del ecuador

== Equinoccio oficial  fue  el  23  / IX / 2,006 ==

== El sabado el 23 de septiembre : ==

¡Precisamente en el dia DOCE HORAS del sol con la condicion de la Tierra sin atmosfera! ( ¡ El primero dia del otoño por los astronomos ! )

==  El martes el 26 de septiembre : ==

¡ Precisamente en el dia DOCE HORAS del sol con las dos de Tierra y atmosfera ! ¡De acuerdo con la tradicion: el primero dia del otoño!

1.3      == This is the basic ACADEMIC STATEMENT that I have submitted to the Cavendish Laboratory, Cambridge, Prof S. W. Hawking, FRS ==

== Title for the Conclusion of Academic Submission, 18th October, 2006==

.......  The Magnitude of the Perturbations in the Historic Data for .......  the Precession of the Equinoxes in the Orbit of the Earth

My recent investigation of the official and traditional dates of the Spring and Autumn Equinoxes shows that a major discontinuity in the historic data for the precession of the Equinoxes occurred at a conjectural date sometime around the 17th Century.

My letter to the newspapers of the 20th September, 2006, and the proof copy, demonstrate that the Equinox defined by the equatorial determination in Autumn is three days in ADVANCE of the date for the traditional “twelve-hour Sun” criterion at the latitude of San Fernando, Pcia Cadiz, Spain (An important astronomical observatory in maintaining the UTC TIME SYSTEM).

By symmetry, it is to be found that the Equinox defined by the equatorial determination in Spring is three days in ARREARS of the date for the traditional “twelve-hour Sun” criterion at the same latitude. Thus, it can be said that both in the past and currently, that the official, equatorially-derived Equinox dates were about six days closer together in the year than those previously determined by the “twelve-hour Sun” criteria. (The new Spring Equinox was later by three days, and the new Autumn Equinox was earlier by three days). The change must certainly have been apparent around the time of the 17th Century.

It might be a little unexpected that there is a latitude dependency in the observations and data. Another factor here is that there is also a “discrepancy” at the Equator for the differing types of the Equinox date determination, because there too, the Sun will appear to be visible for exactly twelve hours on a different date to the official, equatorially defined Equinox! Some clarification is required here.

a) There is NO true discontinuity in the determinations of the official equatorially-derived Precession of the Equinoxes! What is observed in the past was a sharp difference in the dates because the previously defined “twelve-hour Sun” Equinox was being contrasted to the steady, slow change in the equatorially-derived Precession data. Because of latitude dependency, the extension of the visibility of the Sun would be less than the seven to eight minutes perceived for the whole-day two manifestations at the latitude of the Mediterranean. The reason is that the entry and the appearance of the Sun at the horizon at the Equator is perpendicular to the horizon for both. Therefore the Sun will be seen for a shorter time while actually below the horizon. The additional time the Sun is visible at the Equatorial regions on the official Equinox date I estimate to be just more than six minutes over the 12 hours.

b) At other latitudes well above or below the Equator, the similar horizon phenomena would have to be at increasingly acute angles to the horizon with increasing latitude! At extreme Polar latitudes there are no effective Equinox dates for a Sun which appears around those dates, whether above or slightly below the horizon, for the whole 24 hours!

Thus in the Baltic regions, the separation of only the “twelve-hour Sun” Equinox dates will be such that they will be even further apart in the year than perceived in the Mediterranean or the Tropics! If Tycho Brahe in Copenhagen could have been able to determine the twelve hour versions of the Equinoxes accurately (virtually impossible for any observer at that epoch), his dates for the First of Spring and Autumn would have been quite different to those apparent in the Mediterranean!

 == END of Conclusion to Equinox Proof ==

.....My own explanation of the situation concerning differing Equinox dates

== PREPARATORY INFORMATION to Explain my forthcoming changes - which are only stated in outline explanatory form beneath this set of explanations ==

2.0            == Description for Students==

The dates for the first days of Spring and Autumn can be more important than we might think. The ideas about when those days occurred in the distant past have changed in the more recent past, and, nowadays, those dates centuries ago may appear to have been quite wrong! One of the reasons concerning that appearance is the fact that modern scientists and astronomers have "moved the goalposts" ( - making it difficult for us to score a goal, or in other words achieve an objective!).

The definition of what constitutes such days of the "Equinox" has changed fundamentally from previous eras. It was necessary to change to a new definition when it became apparent that the Equator existed and consequently its "properties" had to be recognised.

The previous definition of the "First of Spring" and the "First of Autumn" is familiar to many as those days on which the times of sunrise and sunset are exactly twelve hours apart ( and that is seen to happen quite clearly on only two days in the year ). Unfortunately, that is just the definition that must be put into the rubbish bin, and which the astronomers have not used themselves for a few centuries!

2.1                == THE  NEW  DEFINITION. ==

However, the new definition results in something which appears as only slightly different at first glance. What the astronomers now establish is the exact time that the Sun is recorded to be directly over the Equator (at zenith), twice a year (it results from the appearance that the Sun firstly travels north and afterwards south across the Equator during the year).

Examination of those happenings in an elementary way would seem to verify the requirements for the twelve-hour definition above, since, surely on those particular days, the whole Earth, North and South, gets equal treatment from the Sun in that "position"? In fact, if the Earth did NOT have an atmosphere around it, that supposition would be true. (Using such a concept of "NO atmosphere" is favoured by the astronomers because the days so-defined provide a "perfect average solution" for both the North and South Hemispheres, whereas it is found from accurate observations that the "twelve-hours of Sun" version of the Equinoxes actually results in chaotic dates over the Earth!).

2.2                   ''== A  HISTORY  OF  CHAOS? ==''

Before explaining that situation, we can find out what happened in the past. In "present times", we have only a short experience of the few precious estimates of the newly-defined equatorial Equinoxes compared to the long history of the more obvious Sun-derived data. In the earlier history there are available many versions, observations and ideas based on the traditional appearances of the sunrise and sunset phenomena. Examining such observations for a particular location in detail, we find that, yes, an observer has detected the same dates in other years, but that other observers in other places around that time noted something quite different! It is all too easy in this "chaotic" result to assume that some observers were wrong, either because timing was difficult - as it was in the past - or because the horizon might not have been "true" as a result of high ground on the horizon.

We might have ideas from, say, the Sumerians, Egyptians, Greeks and Romans concerning the "First Day of Spring",etc. Naturally, we must also be aware of the changes made in the various calendars used for the relevant years throughout this history! Even so, careful examination of any such data results in puzzling "chaos", and that could lead investigators to dismiss the evidence.

Can it be explained for some of the historic cases? Yes, enlightenment is achievable because we know now why the "twelve-hour" definition does not appear to be correct. Wherever on Earth (even at the Equator!) you observe the sunrise and sunset times, you will see the Sun for some minutes longer (up to hours longer in the Polar Regions!) than would have occurred if there was in fact NO atmosphere. That is because when the Sun is near the horizon (even just below the horizon) the atmosphere we have behaves like a mirage-forming lens, apparently putting the position of the Sun slightly higher above the horizon (this effect is also well-known for the constellations at the horizon). Sometimes that image also appears distorted, just like a mirage). So that, even on the official Equinox date, the Sun everywhere appears for more than twelve hours, but not measurably near the Poles.

There is another, very important, factor appearing at such times. At the Equator, around the time of the Equinox dates, the Sun is observed to dive into the horizon "vertically" and reappear in the morning by rising also "vertically". Although as said, the Sun appears visible for more than twelve hours, this "diving" procedure means that the Sun moves quickly and the mirage image does not last very long on the Equator (if I make an estimate, it would be a day which is just more than six minutes longer than twelve hours). Conversely, an observer in the Mediterranean Region sees the Sun "tackling the horizon" at a gentler angle from the "vertical" (further north still, observers see the Sun "glide" in or out of the horizon, so the Sun appears above the horizon for a much longer time in such locations, the mirage being available for much longer there).

What does this achieve for the investigator of historic Equinox dates? It means that he can explain clearly why the ancient Equinox dates were increasingly distant with increased latitude of the observers from the dates now derived by the new definition.

Not only that! (is this not remarkable how much knowledge comes out of observing the Sun around the Equinox date?) - if the investigator examines the dates for the traditional Equinox for a given latitude-dependent, historic observer (even, perhaps just Roman ideas!) he is able to determine the time elapsed between the two Equinox dates for that year. For that same latitude (not necessarily accurately known, though that would be better), that elapsed time is virtually the same today, or that elapsed time should have been the same for that observer, and can be determined accurately from present records (current elapsed observation times of the "12-hour Sun" at a given latitude can be used with an error of about one day in 3000 years). That investigator will know the elapsed time accurately for the NEWLY-DEFINED Equinoxes in the year, which is remarkably near-constant over centuries (generating the above observational error as stated). That opens the way to to estimating the historically changed date for the observer of the Equatorially-derived Equinox dates, even though the observer was unaware of the existence of the Equator in that era!

2.3               ''== WHAT IS THE USE FOR SUCH RESULTS? ==''

From this account it would appear that previously rejected observations can be of assistence in redefining regression factors for the computed "Precession of the Equinoxes" in the future. That is a factor in the fine-detail of investigating the accurate path of the Earth´s Orbit. In fact, virtually all that is required from the point of view of the current explanation in that respect is to know how the dates of the Equatorially-defined Equinoxes change with time!

Actually, those dates change extremely slowly with elapsed time. The September equatorially-derived Equinox varies slightly around September 22nd and 23rd! The March one varied in the past century from March 21st to March 20th. Because that change is so slow, it is not easy to establish accurately when change dates occurred, especially in the distant past. Thus, any information which can confirm behaviour in the distant past would be extremely helpful for defining the future behaviour of the Precession! An investigator today merely has to email the relevant observatory near the latitude of one of the historic observations to determine the RECORDED current twelve-hour Equinox dates for the year and hence their elapsed separation time. That elapsed separation time is very much a constant over at least 3,000 years and can then be applied to the changed situation in the past at that latitude - no longer, perhaps, would the perceived dates be "wrong" or "chaotic"!

2.4      The TABLE on the cited WIKIPEDIA introduction to the EQUINOX

(TABLE Liable to be garbled)

Years     Equinox days  Hours (approx)   Delta Equinoxes Sept/Mar     Sept/Mar           In Year (The Addition is useful in analysis.)

2002........23 -20........ 5 -  19........186d 9h    39m

  3........23 -21........11 -   01........186d 9h    47m

  4........22 -20........16½ -   07........186d 9h   41

  5........22 -20........22 –  12½ ........186d 9h   50

  6........23 -20........ 4 -  18½ ........186d 9h   37

  7........23 -21........10 -   00........ 186d 9h   44

  8........22 -20........16 -   06........ 186d 9h   56

  9........22 -20........21 -   12........ 186d 9h   34

 10........23 -20........03 –  17½........ 186d 9h   37

 11........23 -20........ 09 -   23........186d 9h   43

 12........22 -20........15 -   05........ 186d 9h   35

 13........22 -20........21 -   11........ 186d 9h   42

2014........23 -20........2½ -  17........ 186d 9h   32m

Mean "Delta" Here is: ... 186d 09h 41.3m for 13 instances currently

I calculate that the External Reference in Wikipedia for Times of Equinoxes for 1992 to 2020 yields a mean value for "Delta" of ... 186d 09h 44.24m for the 29 instances!

Reference to the larger table of Equinox dates and times in the External References I perceive to yield an error of less than one day in 3000 years in extrapolation to earlier dates.

Details of the Wikipedia Table (used to derive the above Delta Equinoxes

Discarded - on Wikipedia page!

END of  Expalanatory  PREAMBLE to explain reasons for the following changes

2.5               == ERRORS  IN  Wikipedia  Articles: ==

''== ONLY THIS SECTION is the reason for changes to the Wikipedia Article cited. There would be NO change to the format of the original article! ==''

Comments by William Plumtree added to original Wikipedia article texts

SECTION - WIKIPEDIA PAGE (cited)

==Heliocentric View of the Seasons ==(Original Wikipedia section title)

The cause of the seasons is that the rotation axis of the Earth is not perpendicular to its orbital plane, but makes an angle of about 23.44°, the obliquity of the ecliptic, and that this axis keeps its orientation in inertial space. By consequence, for half a year (from around 20 March to 22 September) the northern hemisphere tips toward the Sun, with the maximum Insert "at the Solstice" around 21 June, while for the other half year the southern hemisphere has this honour, with the maximum Insert "at the Solstice" around 21 December. The two instances that the Sun is overhead on the equator are the equinoxes. O K

Also at that moment both the north pole and south pole of the Earth are just on the terminator, and therefore day and night are equally divided over the whole globe. '''NO! This is wholly misleading.' There is NO Terminator anywhere near either Pole. The Sun circles completely around and above the horizon at BOTH POLES because of the refraction at the horizon caused by the presence of the atmosphere. Day and Night are not equally divided ANYWHERE on the Globe on the official Equatorially-defined Equinox date, not even on the Equator!''

The table above gives the instances of equinoxes and solstices over several years. A few remarks can be made. O K The actual equinox is a single moment in time — it does not take the whole day. O K

But the crossing of the Sun over the equator is slow enough that the equinox day will have 12 hours of daylight and 12 hours of nighttime, and within an accuracy of a few minutes, the day before and after too. '' NO! For the same reasons. For the official equatorially-defined Equinox, the statement is completely wrong, the Sun at the Equator shows for about 7 minutes longer than twelve hours, and on the days just before or after (according to whether in March or September), the Sun shows for about 2 minutes less. Most probably on the third day away the Sun shows for either EXACTLY 12 hours, or nearest to that condition at the time!''

It is 94 days from the June solstice to the September equinox, but only 89 days from the December solstice to the March equinox. The seasons are not of equal length because of the variable speed the Earth has in its orbit around the Sun. O K

CAN ADD HERE: Something which is effectively constant over centuries is the time elapsed in each of the two cases of the official equatorially-derived Equinox dates in the given year and that of the "12-hour Sun" duration Equinox dates in the given year. The first "official" instance can be derived from a table of dates and times for 29 years of examples as shown in the "External Reference" Table of Times 1992-2020 www.aa.usno.navy.mil/data/docs/EarthSeasons.html, and I calculate amounts to 186 days 9 hours and 44.24 minutes, with a very small error margin. The change over time is probably of the order of one day in 3000 years! The "computed" factor change can be estimated from the extrapolated data in the literature for the relevant epoch, if required. (See analysis above).

The instances of the Equinoxes are not fixed but fall about six hours later every year, amounting to one full day in four years, but then they are reset by the occurrence of a leap year. The Gregorian calendar is designed to follow the seasons as accurately as possible. It is good, but not perfect. Also see: Gregorian calendar#calendar seasonal error. Smaller irregularities in the times are caused by perturbations of the Moon and the other planets. Currently the most common equinox and solstice dates are 20 March, 21 June, 22 September and 21 December, the four year average slowly shifting to earlier times in the years to come. This shift is a full day in about 70 years (largely to be compensated by the century leap year rules of the Gregorian calendar). But that also means that as many years ago the dates of 21 March, 22 June, 23 September and 22 December were much more common, as older books teach and older people still remember. Note that the times are given in UTC, the time at Greenwich (ignoring British Summer Time). People living farther to the east (Asia, Australia) whose local times are in advance, will see the seasons apparently start later, for example in Tonga (UTC+13) an equinox occurred on 24 September 1999; a date which will not happen again until 2103. On the other hand people living far to the west (America) have clocks running behind in time, and may experience an equinox occurring as early as 19 March. ADD Here (recommended) Nevertheless, the official, astronomically accepted, Equinox Times should be the ones quoted worldwide in UTC (Universal Coordinated Time - historically called Greenwich Mean Time).

2.6        == ERRORS on a further page in Wikipedia ==

An equinox in astronomy is the event when the Sun can be observed to be directly above the equator. O K The event occurs twice a year, around March 20 and September 23. O K More technically, the equinox happens when the Sun is at one of two opposite points on the celestial sphere where the celestial equator and ecliptic intersect.O K

In a wider sense, the equinoxes are the two days each year when the center of the Sun spends an equal amount of time above and below the horizon at every location on Earth. ''Utterly misleading!! This would only occur realistically with the atmosphere absent! The "12-hour Sun Equinoxes" are very latitude-dependent because of the refraction variation with angle of interaction of the Sun with the horizon and many days can elapse between the relevant days!!''

The word equinox derives from the Latin words aequus (equal) and nox (night). O K

The two instances that the Sun is overhead on the equator are the equinoxes. O K Also at that moment both the north pole and south pole of the Earth are just on the terminator, and therefore day and night are equally divided over the whole globe. ''Absolutely incorrect!! There is NO terminator at the Poles at Equinox!! (Only for NO atmosphere!) The Sun circles around the horizon above the horizon at BOTH Poles!''

The table above gives the instances of equinoxes and solstices over several years. A few remarks can be made. The actual equatorially-derived equinox is a single moment in time —it does not take the whole day.OK  But the crossing of the Sun over the equator is slow enough that the equinox day will have 12 hours of daylight and 12 hours of nighttime, and within an accuracy of a few minutes, the day before and after too. Very misleading! ''The "few minutes" are a descriptive device which succeeds in obscuring the true events on the Equator. The day before the official March Equinox has just over 2 minutes less extra sun over the 12 hours and the day after the September one also has about 2 minutes less. TWO days different show about 4 minutes less!! Three days away can elapse on many such occasions before the Sun appears for approximately the twelve hours! On the official Equinox day the Sun appears for about 7 minutes extra!! However, it is true that the occurrence of the official Equinox at night times can displace the pattern of hours and dates to a degree. (see also similar criticism on page above)''

22.7             == Geocentric view of the seasons ==

The explanation given in the previous section would be useful for an observer in outer space. Seen from Earth, the explanation remains the same but the orientation changes. Now the Sun revolves in one year around the Earth, moving along a circle in the sky named the ecliptic which is a reflection of the orbit of the Earth around the Sun. The daily motion of the Sun, (day and night), however, takes place parallel to the equator. The equinoxes are now the points where the equator intersects the ecliptic and the solstices the points on the ecliptic farthest away from the equator. Also note, in the drawing, when the Sun appears to be at the vernal equinox as seen from Earth, that seen from the Sun the Earth is 180° away from it, and thus at the autumnal equinox of its orbit. The perihelion of the Earth's orbit, currently located at 101° longitude, therefore occurs at the beginning of January. O K

As mentioned above, on*** equinox day the Sun passes through the zenith for observers on the equator and is on*** the horizon for those on the poles (but see also below). ***Prefer to see "Above the horizon at both Poles because of refraction".

The March equinox marks sunrise at the north pole and sunset at the south pole, while for the September equinox it is just the opposite. Sorry, But you cannot say that since "The Sun circles above the horizon at both Poles!!.

For all observers on Earth the altitude of the Sun above the southern horizon at local noon is equal to the complement of the latitude (90° - φ). Example: an observer on 60° northern latitude (φ = +60°) will see the Sun at 30° in the south. An observer on 20° southern latitude (φ = −20°) will see the Sun at 110° in the south. But by then one has overshot the zenith (90° altitude), so that this value corresponds to 70° above the northern horizon. Acceptable approximation.

On the equinox day, the Sun rises in the morning, for every place on Earth (except at the poles), exactly in the east and sets exactly in the west in the evening ADD HERE (recommended), but only does so perpendicular to the horizon at the Equator.

(At high latitudes this may be shifted due to atmospheric refraction.) In the half year centred around June it rises and sets more towards the north, which means longer days and shorter nights for the northern hemisphere and shorter days and longer nights for the southern hemisphere. In the half year centred around December the Sun rises and sets more towards the south, and the day and night durations are reversed. O K

Also on the visible for 12-hours-Sun "equilux day", the Sun rises, for every place on Earth (except at the poles), at 6:00 in the morning and sets at 18:00 in the evening. But these times are not exact for several reasons. 'NEEDS that addition' because it cannot possibly happen on the equatorially-defined Equinox day anywhere - even on the Equator, result of refraction

Most places on Earth use a "Time Zone" which is not equal to the local time, differing sometimes up to an hour, and even two hours if summer time is included. In that case, the Sun can rise for example at 8:00 and set at 20:00. NEEDS extra ''for parts of Europe (having truly Double Summertime as in France and Spain over most of 90 years, for example). Indeed, in the case of Vigo, Gallicia, Spain, 9 degrees West, as this is being written in Winter, Spanish "Double Summertime" is effectively in force, with the local true midday time at about 13:52, when the location is effectively 28 degrees West of the set time meridian at 15 degrees East of Greenwich (because the "Equation of Time" makes effectively a further 4 degrees West at that time, having 16 minutes extra). A lower Equation of Time in Summer means that the area is then under "Double Summertime and a Half". The times quoted in this paragraph are far too short, needing to be 08:52 and 20:52 IN WINTER!''

Even those people fortunate enough to have their time zone just equal to the local time, they still will not see sunrise and sunset at 6:00 and 18:00, respectively. This is due to the variable speed of the Earth in its orbit, and is described as the equation of time. It has different values for the March and the September equinox (+8 and −8 minutes respectively). ''Glad to see this, however, I must point out for longevity here that the "Equation of Time" is constantly on the move. Iain Nicolson´s book on "The Sun" shows a plot for this valid for 1982, but in 2006-7 it is now FOUR MONTHS out of phase with reality. The April intersection with the axis has just occurred on December 25 approx. Surprisingly, the change is still advancing very rapidly (perhaps to maximum possible), for example, on the 30th January, 2007 it was advancing at about + 13 mins (I do not have access to an up-to-date plot).''

INSERT HERE:  In the following paragraphs, unless otherwise stated, the situation is precisely applicable to the Equator!

Sunrise and sunset are commonly defined for the upper limb of the solar disk, and not for its centre. The limb is already up for at least one minute before the centre appears, and likewise sets one minute before the last appearance of the limb sets too. O K, But refraction probably will alter that time depending on extreme conditions (such is likely on the hot plains of Africa).

Due to the atmospheric refraction the Sun, when near the horizon, appears a little more than its own diameter above the position than where it is in reality. This makes sunrise more than another ***two minutes earlier and sunset the equal amount later. The two effects add up to almost seven minutes, making the equinox day 12h 7m long and the night only 11h 53m. In addition to that, the night includes twilight. When dawn and dusk are added to the daytime instead, the day would be almost 13 hours.***''As your text infers, "3.5" seems to fit (also little different at middle latitudes). It should be recognised, that only at the Equator around the Equinox times, the Sun enters perpendicularly to the horizon (as distinct to other latitudes at that time, or to somewhere within the tropics at other times).''

The above numbers are only true for the tropics INSERT at other dates when the Sun achieves a trajectory to "zenith", as appropriate for the perpendicular motion. For moderate latitudes this discrepancy gets larger (London, for example: 12 minutes), and close to the poles it gets very large. There is more to this point here:- ''For all latitudes, especially away from the tropics, the angle at which the Sun enters or leaves the horizon necessarily increases the time that it takes the Sun to reach the critical depth below the horizon. At moderate latitudes the effect is not great, but nevertheless can be significant enough to alter the date/time of the "12-hour Sun Equinox".''

Up to about 100 km from both poles the Sun is up for a full 24 hours on equinox day. O K

Alteration IN ADDITION Here

''While the Sun enters and leaves the horizon at various angles appropriate to the latitude "at Equinox times" and since the Sun is about 1/2 degree in diameter, the 3.5 minutes is appropriate for perpendicular motion of the Sun at the horizon (which means just about 1.5 Sun diameters in appearance - for a rate of 2 minutes per half degree - somewhat affected by the distortion present). Approximate trigonometrical models can be drawn for a given latitude by representing the 3.5 minute depth vertically and extending to one side from the vertex at an angle equal to the latitude down to intersect the horizontal forming the base from the vertical line.''

''The time taken for the Sun to reach the critical depth is then represented by the hypotenuse and is calculated as 3.5 mins /cosine angle of latitude. The analysis is approximate, and works well only for the Equinox period.''

Height of the horizon on both the sunrise and sunset sides changes the day's length. Going up into the mountains will lengthen the day, while standing in a valley with hilltops on the east and the west can shorten the day significantly. This is why settlements in east-west running valleys are more favourable (daylight-wise) than north-south running valleys. O K -

2.8     == Further points at foot of the Wikipedia page: ==

Equinoxes are points in time, but equiluxes are days. By convention, equiluxes are the days where sunrise and sunset are closest to being exactly 12 hours apart. This way, you can refer to a single date as being the equilux, when, in reality, it spans sunset on one day to sunset the next, or sunrise on one to sunrise the next. As an example, for a city 45°N and 123°W (Portland, Oregon), the 2006 autumnal equilux was on September 25 when sunrise was at 7:01 am and sunset was at 7:02 pm. The 2006 autumnal equinox was on September 22 at 9:03 pm (all times in Pacific Daylight Time). For the Northern Hemisphere, the autumnal equilux lags behind the equinox, and the reverse is true in the spring. As you might suspect, the whole situation is reversed for the Southern Hemisphere. NOTE needed here ''It can be premature to reverse all ideas for the South! Using the "incorrect" March and September Equinox nomenclature, because it can be confusing to establish by mental reasoning what actually happens around the Equinox near to the Equator when assessing the behaviour in each Hemisphere, the following can be established.''

''The behaviour of the "terminator" at times other than that of the Equinoxes is a source of confusion, because then the terminator exhibits asymmetrical behaviour about the Equator (a source of retained associated memory of behaviour). The behaviour of the Equinox - Equilux pair exhibits symmetrical behaviour about the Equator. For the Northern Hemisphere, the September Equilux lags behind the Equinox. For the Southern Hemisphere, the September Equilux also lags behind the Equinox. (It is nevertheless true that, similar to statements in Wikipedia, the Equilux is on the nearest Summer side of the Equinox!). It is possible for an investigator to by-pass the difficulty of the "24-hour Equilux" mentioned here, by considering "like with like" and temporarily infer the usage of equivalent 12-hour periods at the time of the Equilux.''

Exercise for the investigator interested in detail!

Portland: Equatorial Equinox 22 Sept latish, 21:00 PDT (evening) on stated Equilux day 25th Sept of 12 hours-1 min approx (Both manifestations at 07:01 PDT would have indicated a midday point for the Equilux). There is no reason in analysis to ignore the actual hours (approx). As an exercise, it could be considered that for the following day 26th, the day length decreased by just more than two minutes (at the time of the Equinoxes, this factor seems to be a constant for all appropriate latitudes). Is it likely, therefore, that the 26th was a day with 11 hours and 59 minutes? While that might contradict the idea that an Equinox must involve twelve hours, I believe that in the process of analysing the situation, it can be regarded that the Equilux really occurred overnight (just as the Official Equinox did on the 23rd). Therefore the difference in time between the two types of Equinox then would have to be assumed to be 3 days and 4 hours approximately. This provokes a query! The difference seems too small for 45 degrees North? I have found (approximately, it is true) that the difference appears to be about 3 days at 36 degrees North (Cadiz, Spain). Therefore, should the difference not be greater for Portland?''

My Analysis for CADIZ, Spain, differed at the time!

''DATA for CADIZ.(San Fernando OBS) Equinox at 23rd Sept 04:03 UTC, Equilux at 26th Sept apparently midday UTC. Actual 12 hours zero mins reported with fairly symmetrical differences on days before and after Equilux. Difference therefore 3 days 8hours!! Something is amiss!''

''Dare I suggest that the Equilux in September at Portland was also nearer the 26th that night, because the next day to the "Equilux" of 25th might have been 11 hours and 59 minutes? How much does the progress of time over the eight hours to the PT Time Zone affect the outcome? Also, my instances were in accordance not only with local newspaper reports plus a report also for Gibraltar nearby, but also with my own observations of sunset at the time?''

W E G Plumtree, M.Phil.(Lond - Imp. Coll), B.Sc.(Spec Phys), Formerly in Uk M.Inst P.