Dear Steve, I will share the data for the whole year once I get it (I only have April). Your preliminary results sound too good to be true. I did a simulation using TimeAndDate.com for three completely different locations on the European continent: Constanta (Romania), Gorlitz (Germany) and London (UK). I chose these because of their proximity to the time zone meridians, this way the old solar time is more easily found (I just exclude 1h daylight saving time).
The stated duration of the day of 13h20min for April is recorded in those localities on April 13, April 7, respectively April 7 (all gregorian, the last two cities have approximately the same latitude). Sunrise on these respective dates (in solar time): 5:24 in Constanta, 5:23 in Gorlitz, 5:23 in London. Sunset (in solar time): 18:47 at Constanta, 18:42 at Gorlitz, 18:43 at London. In the calendar we've got sunrise at 5:20, which is a good enough fit for all the above examples. On the other hand, sunset is at 7:14 (p.m.) and this time doesn't fit any of the examples above. If we consider it to be the civil twilight, we have 19:16 for Constanta, 19:16 for Gorlitz, respectively 19:17 for London (old hours). These correspond quite well with the sunset given by the calendar. Dan Uza On Fri, Aug 9, 2019 at 10:15 PM Steve Lelievre < [email protected]> wrote: > Dan, > > Using only your April data, and assuming: > > 1. day length is the difference of the sunset and sunrise (as opposed to > the daylength stated), > 2. sunrise and sunset are when the center of the sun is on the horizon > 3. my modern source of solar declination data is "good enough" > 4. your table of values is for the Julian calendar, which for the year in > question is offset from the Gregorian calendar by 11 days, > > then, by varying latitude to minimize the Sum of Squares of the > differences between true day lengths and the representative day length > stated .... I get a latitude of 44.413N, which would correspond to > Bucarest. > > If I could use your table of data for the full year, the result would of > course be different - better, I would hope, but possibly not! > > As yet, I have no idea why the stated day length is not the same as the > difference of the sunrise and sunset. > > Steve > > > On 2019-08-09 1:06 a.m., Dan-George Uza wrote: > > Hello all, > > I have seen an old calendar from 1793 which lists for every month sunrise > and sunset times as well as day and night duration. For example, taking the > month of April: sunrise at 5 h 20 m, sunset at 7 h 14 m; day length 13 h 20 > min, night length 10 h 40 m. > > Somebody asked me if it would be possible to establish the approximate > geographical area for these predictions. > > I'm pretty sure it's not possible. Back then they used true solar time (or > perhaps mean solar time?) so I guess these hours would have been valid for > a whole parallel of latitude, with variations once you go north or south. > > Nevertheless, I made a simulation and realized that I cannot get close to > these numbers. I don't know why. Perhaps because back then sunrise and > sunset was not counted by solar limb, but by geometric center of the Sun? > How did they do it? > > Regards, > > -- > Dan-George Uza > > ---------------------------------------------------https://lists.uni-koeln.de/mailman/listinfo/sundial > > > --------------------------------------------------- > https://lists.uni-koeln.de/mailman/listinfo/sundial > > -- Dan-George Uza
--------------------------------------------------- https://lists.uni-koeln.de/mailman/listinfo/sundial
