|
PureBytes Links
Trading Reference Links
|
Tom Pedersen wrote:
> Bob,
> I like it!
> Did you celebrate today?
> I know nothing of Julian but perhaps you (and few others)
> could explain to a programming lamen (lame-man) how it
> functions.
> Thanks,
> See U
This is from the help file of a Scientific Approaches software product:
-------
The so-called Julian Calendar was developed by Julius Caesar with the advice
of the astronomer Sosigenes. The Julian Calendar discards the lunar month
used in more ancient Arabic and Jewish calendars and adopts 365.25 days as
the length of a year. This year is divided into twelve periods (months) of
30 or 31 days. (February doesn't have 28 or 29 days in the Julian Calendar.
) The normal year was 365 days. To make up the extra 1/4 day, an extra day
was intercalated (put into the normal calendar) every four years.
365.25 was amazingly close to the exact actual period of a year. The actual
duration of a tropical year is only 0.0078 day less than Sosigenes'
calculations. Even so, with a 0.0078 day per year error rate, after 1,000
years the Julian Calendar was in error 7.8 days. By 1582 the date of the
vernal equinox was March 11, instead of March 21, so Pope Gregory decided to
return the sun to its proper position by slipping the calendar forward ten
days and by modifying the Julian Calendar calculations in such a way that
the error would not reappear.
We still use Pope Gregory's modified Julian Calendar today. It is identical
to the Julian Calendar, except only such century years are leap years as are
divisible by 400. This is roughly equivalent to dropping 3 days every 400
years, resulting in an average year length of 365.2425 days. That value
differs from the exact tropical year length by only 0.0003 days, so that by
the year 2582 (1,000 years after its creation) Pope Gregory's calendar will
be in error only slightly less than 1/3 of a day. Our decedents can have an
extra leap day 2,000 years after that, in the year 4582, to correct 3,000
years of accumulated error and return the sun back to exactly where it
belongs.
The Gregorian Calendar was immediately adopted in 1582 by all Roman Catholic
countries, but the Greek Church and most predominately Protestant countries
refused to recognize it until long after that time. The confusion following
the change persisted down to the present century. Both calendars were in
common use in different parts of the United States early in its history,
causing confusion for modern historians. Romania did not change from Julian
to Gregorian until 1919. Some parts of the world, such as predominately
Moslem countries still use various forms of ancient lunar calendars.
Astronomers number dates with "Julian Day Numbers" to avoid the need to
correct for the various changes that have been made to other calendars over
the years and because they often are concerned with dates prior to the
beginning of the Gregorian Calendar. Julian day zero was a very long time
ago, so the count is now very high. The Gregorian Calendar date January 1,
2000, will be Astronomical Julian Day Number 2,451,545. 2,451,545 / 365.
2422 = 6,712.1077, so that date will be 6,712 years and 39 days from day
zero of the Astronomical Julian Calendar.
Astronomical Julian Day Numbers are widely used internally by computer
programs to record dates and to perform date arithmetic. Date arithmetic is
greatly simplified with Julian Day Numbers, because the number of days
between any two dates can be determined by simply subtracting one from the
other. Compare that to the problem of determining the number of days
between two Gregorian dates, like April 17, 1998, and November 3, 1983,
where calculation would have to take into account the varying number of days
in each month and the extra February day in intervening leap years.
Julian Day Numbers also make it easy to calculate the day of the week.
Julian Day Numbers change at noon, rather than at midnight, as in the
Gregorian Calendar. If you know the Julian Day Number at noon, the day of
the week can be obtained by adding one and then taking the result modulo
seven. A zero result corresponds to Sunday, one to Monday, two to Tuesday,
etc. Compare that to the difficulty of calculating the day of the week of a
Gregorian Calendar date like May 4, 1999.
For more information about Astronomical Julian Day Numbers, see:
Meeus, J. 1982, Astronomical Formulae for Calculators, 2nd ed., revised and
enlarged (Richmond, VA: Willmann-Bell).
Also, Hatcher, D.A. 1984, Quarterly Journal of the Royal Astronomical
Society, vol. 25, pp. 53-55; and op. cit. 1985, vol. 25, pp. 151-155, and
1986, vol. 27, pp 506-507.
-Bob Brickey
Scientific Approaches
sci@xxxxxxxxxx
|