Dear Jos,

You ask:

> Has anyone an idea how to use a sundial for
> calculating if the current year is a leap
> year?

Yes, been there, done that, got the T-shirt.

The Noon Mark on the new London Stock Exchange
in Paternoster Square more or less does what
you want but once you understand it, you will
probably decide to move on to a different
project!

The FIRST thing you should do is to get a
feel for the leap year cycle and how it
affects solar declination.

Using some kind of solar calculator, such
as:

  http://www.esrl.noaa.gov/gmd/grad/solcalc/

you should do the following:

  Choose a date when the declination is
  changing quite rapidly.  We are near
  the autumnal equinox and I suggest you
  choose 22 September 2014.

  Then choose 12:00:00 as the time.  It
  really doesn't matter what time you
  choose, or what longitude, so long as
  you stick to the same time and place.

  If you are on the Greenwich Meridian
  then you will see that the declination
  is +0.23 degrees at 12h on 22 September
  this year.

  Then change the year to 2015, but don't
  change anything else.  You will see the
  declination is +0.33 degrees.

  Continue like this for about 36 years.
  Here are the first few values:

    2014   +0.23
    2015   +0.33
    2016   +0.04
    2017   +0.13
    2018   +0.22
    2019   +0.32
    2020   +0.02
    2021   +0.12
    2022   +0.21
    2023   +0.31
    2024   +0.01

OK, do you see the pattern?  Starting in a
leap year, say 2016, the declination increases
for the next three years and then jumps BACK to
(but not quite to) where we started.  In 2020
we are not quite where we were in 2016.

If you have a large enough nodus height you
can track tiny changes in declination.

If you mark out a giant analemma you can
divide it up into individual days, 366
days including 29 February.

If you concentrate on 22 September you can
see that the declination is lowest in a
leap year and then gets higher for the
next three years.

All this will work without ambiguity for
about 36 years.  Then, the long-term
drift [the Julian drift] results in you
drifting into the previous day and the
whole thing breaks down.

I could get away with this on the new
London Stock Exchange since the design
life of the building is only 40 years
and the plan will work most of the time.

This scheme works, in theory, for almost
any date but near the solstices the
changes are too small to notice and you
won't be able to detect them.  Worse, on
the day of a solstice, the declination
changes direction and it's a mess!

Fortunately, 29 February is at a time of
year when the declination is changing
quite fast so you can mark that in quite
nicely.

For completeness, Leap Year Day is really
24 February but there are probably only
two other people on this list who agree
with that assertion!

[I distinguish here between the extra
DAY and the extra DATE in a leap year.]

Very best wishes,

Frank

Frank King
Cambridge, U.K.

---------------------------------------------------
https://lists.uni-koeln.de/mailman/listinfo/sundial

Reply via email to