A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
From: Robin Stuart
Date: 2019 Jan 13, 09:41 -0800
You are correct. What I labelled as “GHA at T0” is computed from Skyfield as R.A. – GST and is therefore sloppy labelling on my part but it was never really supposed to be seen by the user. Glad that there is agreement on the nitty gritty digits. Thank you.
If you could confirm the TT of geocentric conjunction of the Moon and star (T0) in R.A. then I think we could draw some concrete conclusions.
When provided with correct input data the only approximation in the spreadsheet is in the use of linear Besselian elements for the Moon. The calculation of local circumstances and the calculation of longitude from the occultation time use separate pieces of code and different numerical algorithms but produce results that are exactly consistent.
It would seem therefore that the linear approximation is the source of the few second differences being observed. Although including the quadratic terms in the Besselian elements goes beyond what was available in Nautical Almanacs and I originally declared I would not do it, it’s relatively straightforward and might be worth adding them in to put the whole matter to bed,
For reference here’s the relevant data that comes to Skyfield. It is as it appears in the original Hipparcos catalog.
Although we expect the RA and dec to be different presumably the proper motions come from different surveys.
Also I queried the TYCHO catalog for the ICRS J=2000.0 position http://vizier.u-strasbg.fr/viz-bin/VizieR-3 and it seems to be slightly different to what you see.