Is this possible?

Greetings!

Is there an algorithm, or could one be formulated, to calculate the
phase of the moon throughout the the entire 7980 year Julian cycle -
4713 BC to 3268 AD?

I'm developing a historical calendar tool and want to be able to show
the phases of the moon for each day in the Julian cycle.

I've been doing some research on the subject, including lots of Web and
Usenet searches as well as getting the first edition of Jean Meuus'
"Astronimical Algorithms".

My findings so far indicate that such an algorithm does not exist.
It seems that Meeus' algorithms are only good back to about 2000 BC.

So I guess my next question is: is there an algorithm, or could one be
formulated, to calculate the phase of the moon from 4713 BC to 2000 BC?


Respectfully,
Jim Showalter

"Jim Showalter" wrote in message
news
Greetings!

Is there an algorithm, or could one be formulated, to calculate the
phase of the moon throughout the the entire 7980 year Julian cycle -
4713 BC to 3268 AD?

I'm developing a historical calendar tool and want to be able to show
the phases of the moon for each day in the Julian cycle.

I've been doing some research on the subject, including lots of Web and
Usenet searches as well as getting the first edition of Jean Meuus'
"Astronimical Algorithms".

My findings so far indicate that such an algorithm does not exist.
It seems that Meeus' algorithms are only good back to about 2000 BC.

So I guess my next question is: is there an algorithm, or could one be
formulated, to calculate the phase of the moon from 4713 BC to 2000 BC?


The problem is, the ancient values for Delta T (variations in
the period of the Earth's rotation) are not well known, and
neither is the period of the Moon's orbit. These things depend
upon many physical factors including weather and climate on
the Earth, which effects tidal friction, and so forth.

For the historical period, astronomers try to pin down the
Earth and Moon configuration by referring to records of
eclipse sightings (when and where they are seen can pinpoint
the positions of the Sun, Moon, and Earth). There are no such
checks for prehistoric periods.

Jim Showalter writes:

Greetings!

Is there an algorithm, or could one be formulated, to calculate the
phase of the moon throughout the the entire 7980 year Julian cycle -
4713 BC to 3268 AD?

I'm developing a historical calendar tool and want to be able to show
the phases of the moon for each day in the Julian cycle.

I've been doing some research on the subject, including lots of Web and
Usenet searches as well as getting the first edition of Jean Meuus'
"Astronimical Algorithms".

My findings so far indicate that such an algorithm does not exist.
It seems that Meeus' algorithms are only good back to about 2000 BC.

So I guess my next question is: is there an algorithm, or could one be
formulated, to calculate the phase of the moon from 4713 BC to 2000 BC?

DE406 is a numerical integration that covers the time period from -3000
to +3000 and includes the Moon. While not an algorithm itself, one could
in principal develop an algorithm that fits the integration and then use
it to extrapolate back to -4712. Unfortunately, I know of nothing against
which you could test the accuracy of the extrapolation, but a 1700 year
extrapolation from a baseline of 6000 years isn't too bad. The biggest
problem might be the unknown value for the correction from the ephemeris
time variable (atomic time, a uniform time interval) to Earth time (UT,
for example, which is not uniform due to the tidal evolution of the
Earth's rotation rate).

On Sat, 13 Sep 2003 10:35:40 +0000, Greg Neill wrote:

The problem is, the ancient values for Delta T (variations in
the period of the Earth's rotation) are not well known, and
neither is the period of the Moon's orbit. These things depend
upon many physical factors including weather and climate on
the Earth, which effects tidal friction, and so forth.

For the historical period, astronomers try to pin down the
Earth and Moon configuration by referring to records of
eclipse sightings (when and where they are seen can pinpoint
the positions of the Sun, Moon, and Earth). There are no such
checks for prehistoric periods.

Ok, I see now that I was being naive to assume that just because
the moon can be calculated back thousands of years, that it
could be accurately tracked indefinetly.

Thanks for enlightening me, Greg.

On Sat, 13 Sep 2003 18:53:51 +0000, thole wrote:

DE406 is a numerical integration that covers the time period from -3000
to +3000 and includes the Moon. While not an algorithm itself, one could
in principal develop an algorithm that fits the integration and then use
it to extrapolate back to -4712. Unfortunately, I know of nothing against
which you could test the accuracy of the extrapolation, but a 1700 year
extrapolation from a baseline of 6000 years isn't too bad. The biggest
problem might be the unknown value for the correction from the ephemeris
time variable (atomic time, a uniform time interval) to Earth time (UT,
for example, which is not uniform due to the tidal evolution of the
Earth's rotation rate).

Could you give me some links where I can learn about DE406?
I think it's worth pursuing.

Thanks much, tholen!