Dear Frank, I am not so convinced of a big difference, at least between JPL and IMCCE. I used the JPL Horizons simulator with a geocentric position because if you choose a town (on the surface) the results are different. I admit that the center of the Earth is not the best place to watch the Sun. The results seem close to those of IMCCE.
Horizons simulator gives : ************************************************ Date__(UT)__HR:MN R.A. (a-apparent)__DEC ************************************************** 2019-Sep-23 07:40 11 59 58.47 +00 00 09.6 . . . . . . . . 2019-Sep-23 07:48 11 59 59.66 +00 00 01.8 2019-Sep-23 07:49 11 59 59.81 +00 00 00.9 2019-Sep-23 07:50 11 59 59.96 -00 00 00.1 2019-Sep-23 07:51 12 00 00.11 -00 00 01.1 2019-Sep-23 07:52 12 00 00.26 -00 00 02.1 . . . . . . . . 2019-Sep-23 08:00 12 00 01.46 -00 00 09.8 The RA is increasing by 2.99 s in 20 min, which means 0.01 s in 4,01337 sec). Calculating it from 07:40, this means that RA reached 12h at 07:50: 14.04s for JPL according to the table and IMCCE gives 07:50 15.58s The Dec is decreasing by 19.4 " in 20 min from 07:40 to 08:00 which means 0.1" in 6.18556 sec. Calculating it from 07:40, this means that Dec reached 0° at 07:49 53,81s for JPL and IMCCE gives 07:49: 51,80 This is just an extrapolation of data over 20 min and not the best in terms of accuracy. There is a difference, indeed (1.5 and 2s) but it is not huge. This could come from some of the selected parameters (there are many) or/and the planetary theory used (VSOP or DE406, etc..) that could be different from JPL and IMCCE. This does not mean that one is better than the other one. I have not investigated Sol et Umbra yet, because of lack of time, but this could come from the selected location not being the center of the Earth. I am open to any other better explanation. Very best regards Hervé Guillemet ----- Mail original ----- De: "Frank King" <[email protected]> À: "Hervé Guillemet" <[email protected]> Cc: "Sundial list" <[email protected]>, "Frank King" <[email protected]> Envoyé: Vendredi 4 Octobre 2019 14:58:54 Objet: Re: Equinoctial Puzzle Dear Hervé, Congratulations on your comments on my puzzle about the September Equinox last month... > It seems that the answer to your question > can be found in the attached picture > inclosed in a recent information letter > issued by the French IMCCE institute > specialised in celestial mechanics and > ephemerides calculations This gives us three times of interest on 23 September 2009: 07:49:51.80 Right Ascension = 12h 07:50:11.81 Solar Longitude = 180 07:50:15.58 Solar Declination = 0 Call this the FRENCH solution. If you have an Android cell 'phone you can look at Sol et Umbra which gives these times on 23 September 2009: 07:49:49.40 Solar Declination = 0 07:50:09.25 Solar Longitude = 180 07:50:32.50 Right Ascension = 12h Note that the events occur in reverse order! Call this the ITALIAN solution. Now use the JPL Horizons program: https://ssd.jpl.nasa.gov/horizons.cgi I don't have a proper computer at the moment but here are three values I found (using my 'phone) for the single time: 23 September 2019 07:50:12.00 Solar Declination = 00:06:13.3 Solar Longitude = 180.0019964 Right Ascension = 11:59:01.94 The declination has not yet dropped to zero. The longitude has gone past 180. The Right Ascension has not yet reached 12h. Call this the U.S. solution. Moral: never believe a single source of information :-) If you think you can see the pattern, try using the Horizons program to investigate the March Equinox in 1718. Using the Gregorian Calendar, we find: The Right Ascension went to zero late on 16 March (just before midnight). The declination went through zero about the same time on 16 March. The solar longitude reached zero on 21 March. FIVE DAYS LATER!!! So you see: there is still a little bit more of my puzzle to unravel!! Very best wishes Frank
--------------------------------------------------- https://lists.uni-koeln.de/mailman/listinfo/sundial
