MathGroup Archive 1996

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Julian Day Calculation - Plea for Help

  • To: mathgroup at
  • Subject: [mg3983] Re: Julian Day Calculation - Plea for Help
  • From: Ken Starks <straton at>
  • Date: Tue, 14 May 1996 02:11:35 -0400
  • Sender: owner-wri-mathgroup at

searchme at (Matthew K.) wrote:

> I need , if possible, to get the equation for calculating the Julian Day
> well into the future (we're talking 1000 years here). 

There is no need to worry overmuch about being '1000 years hence'.
The Julian Calendar repeats exactly every 400 years, which 
is also an exact number of weeks.
( 365 x 400 ordinary days + 97 leap days 
= 146097 days
=  20871 weeks )

Once you have a conversion for any 400 year interval, it is trivial to 
add or subtract a few multiples of 146097 on one side of your function 
and a matching number of 400 years on the other.
The fact that it is an exact number of weeks means that you 
also get the day ofthe week without even trying.
 ( Exercise: what proportion of '13th of the month'
happen on a Friday ? )

Beware of using Julian dates as a measure of scientific/absolute
 time ( as measured on an atomic clock ) because civil time 
sticks in extra 'leap seconds' whenever the powers
that be happen to find it convenient. They are neccessary because
 of the gradual slowing in the rotation of the earth on its axis,
 and that is a bit hard to predict accurately  - in the long
 term - because it depends on how much icecap melts Etc.


  • Prev by Date: Re: Installation
  • Next by Date: Re: multinomials: a programming question
  • Previous by thread: Re: Julian Day Calculation - Plea for Help
  • Next by thread: mathematica in C