This page last modified 2002 September 11

False Retrodiction

Some Comments on the Misuse of Planetarium Software for Analysing Ancient Events

There is a growing trend, probably triggered by the availability of excellent planetarium software at little or no cost, for this to be misused in relation to ancient events. This short article examines this misuse in general.

It is usually the case that the people who are guilty of this are woefully ignorant of the limitations of the software they use and, consequently their claims are specious at best. They most often fall into the camps of pseudo-historians or pseudo-archaeologists (they themselves prefer the term "alternative" to "pseudo"). They are part of a growing trend to misuse astronomy to justify outrageous claims for which there is no tangible evidence.

The better amateur planetarium programs use the full VSOP87 (Variations Seculaires des Orbites Planetaires)[1] theory to calculate planetary positions over long periods [2]; lesser programs use less precise methods. The accuracy of VSOP87 probably deteriorates with increasing time; it is thought to be precise to better than an arcminute for dates later than ab out 6000 years BP. The point is that no-one actually knows how accurate it is, even as far back as that. Prior to 6000BP , the errors increase markedly — the accuracy of all planetary theories does this outside their proper time range.

Another factor affecting the observed position of solar system objects in the skyor on the horizon is the obliquity of the ecliptic. Bill Gray, the author of Guide, warns against using his program outside the range of 10,000 years either side of the present because the rapidly increasing errors outside this range.

The retrodiction [3] of cometary positions into the past is fraught with difficulties. Comets are relatively low-mass objects whose orbits have a high eccentricity. If a comet passes close to a planet, its trajectory will be changed. (In the case of Comet Shoemaker-Levy 9, it was changed so that it impacted the planet!) Hence, the retrodiction of cometary positions for periods in excess of a couple of millennia is generally foolhardy at best.

[1] Full precision VSOP87 (and PS-1996 [2]) is definitely used by Guide and Skymap; most other planetarium programs use a lower-precision model from Jean Meeus's Astronomical Algorithms.
[2] PS-1996 theory is used over shorter periods — up to about 100 years for the naked-eye planets.
[3] Backward prediction.


I wish to acknowledge my reliance on some notes on accuracy published by Bill Gray, the author of Guide, and on Jean Meeus's Astronomical Algorithms as sources of information for this article.