NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: The Online Nautical Almanac- beware!
From: Frank Reed CT
Date: 2004 Jul 6, 22:42 EDT
From: Frank Reed CT
Date: 2004 Jul 6, 22:42 EDT
George H wrote of Reis's online almanac:
"He is investigating the problem further, and hopes to correct it or at the
very least to post a warning if users stray into error-prone dates."
I've puzzled out the source of the larger portion of the errors in his star positions. He has apparently not included the "proper motions" for the stars.
Many stars, especially distant stars, have proper motions that are so small that they are of on consequence even after a couple of hundred years. But for those stars which have large proper motions, this is a big issue even for Reis's recommended period of time: 1950 to 2050. For example, the position for Arcturus in 1950 given by Reis's almanac site is off by 0.8 minutes in SHA and 1.4 minutes in Declination, and his 1950 position for Rigel Kentaurus (alpha Centauri) is off by 5.4 minutes in SHA and 0.5 minutes in Declination. If you take it back to 1804, his Rigil Kent position is off by nearly half a degree, and the Arcturus position is out by more than 7 minutes of arc.
In addition, there is still a precession problem (that's a best guess on my part) in Reis's star positions. This is not a serious problem in the recommended period, but it leads to additional errors of up to 1.5 minutes in most cases and errors over 5 minutes in exceptional cases when you go back 200 years.
And George, if you used any data from his site to analyze those Lewis & Clark lunars back in April, as I did, you can see that you may want to revisit those calculations.
None of these errors are present in my online almanac data:
http://www.HistoricalAtlas.com/nadata_v5.html
Frank R
[ ] Mystic, Connecticut
[X] Chicago, Illinois
"He is investigating the problem further, and hopes to correct it or at the
very least to post a warning if users stray into error-prone dates."
I've puzzled out the source of the larger portion of the errors in his star positions. He has apparently not included the "proper motions" for the stars.
Many stars, especially distant stars, have proper motions that are so small that they are of on consequence even after a couple of hundred years. But for those stars which have large proper motions, this is a big issue even for Reis's recommended period of time: 1950 to 2050. For example, the position for Arcturus in 1950 given by Reis's almanac site is off by 0.8 minutes in SHA and 1.4 minutes in Declination, and his 1950 position for Rigel Kentaurus (alpha Centauri) is off by 5.4 minutes in SHA and 0.5 minutes in Declination. If you take it back to 1804, his Rigil Kent position is off by nearly half a degree, and the Arcturus position is out by more than 7 minutes of arc.
In addition, there is still a precession problem (that's a best guess on my part) in Reis's star positions. This is not a serious problem in the recommended period, but it leads to additional errors of up to 1.5 minutes in most cases and errors over 5 minutes in exceptional cases when you go back 200 years.
And George, if you used any data from his site to analyze those Lewis & Clark lunars back in April, as I did, you can see that you may want to revisit those calculations.
None of these errors are present in my online almanac data:
http://www.HistoricalAtlas.com/nadata_v5.html
Frank R
[ ] Mystic, Connecticut
[X] Chicago, Illinois