NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Accuracy of 1767 almanac
From: Peter Monta
Date: 2013 Aug 3, 20:47 -0700
From: Peter Monta
Date: 2013 Aug 3, 20:47 -0700
Geoff South's post got me wondering---how accurate were the Nautical Almanac lunar distances? Let's take the 1767 almanac as an example. Maskelyne's preface claims one degree of longitude, or 2' of lunar distance, while Wikipedia says the early almanacs were good to 0.5'. Annoyingly, the time coordinate is apparent time; but after sorting that out, and typing in a couple of weeks of data, it looks from the attached plot as though it's a little worse than 0.5'. Peak errors are hitting 1.0' in just this short sample (taken from January, February, October, and November to cover the equation-of-time extremes). Are there transcriptions of the complete data sets for some or all of these almanacs? I'll attach the code and data in case there's interest in running it on other data sets. I've run across several discussions of all this in the navlist archives (from e.g. March 2008). Sorry for any duplication. Cheers, Peter