NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Lunars - Oblateness Correction
From: Frank Reed
Date: 2008 Jul 12, 21:02 -0400
From: Frank Reed
Date: 2008 Jul 12, 21:02 -0400
Since this has come up in a several recent messages, I thought I would spell out the way it's handled in my online lunars calculator. The approach is straight out of Chauvenet. What we call "oblateness" is the slight flattening of the Earth at the poles relative to the equator converting it into an "oblate spheroid" rather than a perfect sphere. The degree of flattening is approximately one part in 300 (1/297 is closer but not significantly so for these purposes). Chauvenet usually refers to this property as the "compression" of the Earth's spherical shape. There are two pieces to the correction. First, you correct the HP taken from the almanac (this is Chauvenet's "Table XIII" correction): HP=HP0*(1+(sin(Lat))^2/300). This is a small correction. On average, the error from ignoring it is roughly 0.05 minutes of arc in the clearing process which corresponds to an error in longitude of about 1.5' or about 1 nautical mile on average. The correction is larger at higher latitudes, but of course this has a smaller impact on the position fix since the longitude lines converge. Second, you correct the lunar distance (after clearing it, but it doesn't really matter whether you do it at the beginning instead) with a small increment: inc_LD=sin(Lat)*[(HP/150)*(sin(Dec2)/sin(LD)-sin(Dec1)/tan(LD))] where Dec1 is the declination of the Moon and Dec2 that of the Sun or other body. The error in longitude, converted to nautical miles, that would result from ignoring this second small correction would be, on average, about equal to (1.35 n.m.)*sin(2*Lat). By the way, Chauvenet has a factor of "A" in his equation and tells the reader to look it up in a small table giving log(A) as a function of latitude. If you're trying to figure it out from that table, bear in mind that the table actually shows log(A)+10 which was normal back then. The small variation of A with latitude is not at all important. Chauvenet had the clever idea that most of this small correction could be included in the lunar distance tables in the almanac. Then the correction would have been simply inc_LD=sin(Lat)*x where x is equal to everything in the square brackets above. I have considered adding this little addition to the predicted lunar distance tables on my web site. Anybody want it? PLEASE NOTE: these corrections have been included in the calculations of the lunar distance clearing tool on my web site for over three years. You can directly assess the significance of the total oblateness correction by selecting "Ignore Oblateness" in the "Options" section. If you're trying to get an exact assessment of your skill or your sextant's arc error, there's no reason to turn off the oblateness calculation. But for historical NAVIGATIONAL calculations or just for general understanding, there may be times when you want to turn off the oblateness correction. Incidentally, there's no real reason to bother with the remaining details of Chauvenet's own, rather idiosyncratic method of clearing lunars. It was rarely used, and it offers no really significant advantage. But the general discussion in his book is definitely worth reading. -FER www.HistoricalAtlas.com/lunars --~--~---------~--~----~------------~-------~--~----~ Navigation List archive: www.fer3.com/arc To post, email NavList@fer3.com To , email NavList-@fer3.com -~----------~----~----~----~------~----~------~--~---