NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: Pub. 249 and Pub. 229
From: Gary LaPook
Date: 2017 Dec 16, 18:42 +0000
From: Bob Goethe <NoReply_Goethe@fer3.com>
To: garylapook@pacbell.net
Sent: Friday, December 15, 2017 7:30 PM
Subject: [NavList] Pub. 249 and Pub. 229
From: Gary LaPook
Date: 2017 Dec 16, 18:42 +0000
That does not make sense since the same computer does the computation for both tables and probably to ten significant figures. The only difference is in the tabulation of the data which takes place after rounding. Plus and minus 0.5 for HO 249 and 0.05 for HO 229 so those are the maximum sizes of the errors in the tabulated data and the probable error is half of those values. BTW, these errors are not normal but have a square shaped distribution. . It gets more complicated if you have to interpolate. If you just have to do first differences interpolation in HO 229 then the maximum error goes to 0.14 with a probable error of 0.03'. When using the second differences the maximum error varies between 0.12' to 0.31' and the probable error 0.03' to 0.05'. In some weird cases the error can be larger but only if you don't do the second differences interpolation.
gl
From: Bob Goethe <NoReply_Goethe@fer3.com>
To: garylapook@pacbell.net
Sent: Friday, December 15, 2017 7:30 PM
Subject: [NavList] Pub. 249 and Pub. 229
On page 45 of John Karl's book, Celestial Navigation in the GPS age, he says that:
- Pub. 249 is tabulated to the nearest minute (an accuracy of +/- 0.5')
- Pub. 229 is tablulated to the nearest 0.1' (implying an accuracy of +/-0.05')
He goes on to say, "Unfortunately, that's not the case at all. The best H.O. 229 can do is +/-0.2', which compared to H.O. 249's +/-0.5' accuracy, is only marginally better."
I have pondered why this might be so...and have come up with nothing. I would appreciate Karl (or anybody else for that matter) unpacking this for me a bit.
Thanks.
Bob
