NavList:

Frank,

You "wondered why they had not done a complete modern analysis of the occultations".

There is no doubt that lunar limb features can be a source of interest and fascination (see https://academic.oup.com/mnras/article/276/2/476/998827). Nor is there any doubt that they make a difference if you know your position precisely - but Worsley did not. To extract a chronometer error (CE) you need to fold in your latitude and local mean time both determined by observation with a sextant or theodolite and the occultations can only be timed to half a second at best.

The elephant in the room for the longitude of Endurance is the errors in the Moon's position in the Nautical Almanac and the catalogued positions of the occulted stars. Exactly what we chose to include in our calculations is clearly stated in our paper which allows anyone to check our results and include any additional corrections or adjustments as they see fit. For very high accuracy refraction needs to be taken into account. It might seem that refraction would affect the Moon and the star in same way but the rays that enter an observer's eye would strike the Earth at a different location if the atmosphere were removed. We do not include that.

Whether the inclusion of lunar limb features adds extra useful information is not clear, at least to me, but it might be a partially decidable question. Does their incorporation produce a significant improvement in the R² of the fit of CE against time? That might be interesting to know but whether they could have had any material impact on the hunt for the wreck of Endurance, given the several other uncertain factors that influence its longitude, is quite another matter.

Regards,
Robin