NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: Latitudes by lunar distance. was: Lunars with and without altitudes
From: George Huxtable
Date: 2006 Nov 26, 15:47 -0000
From: George Huxtable
Date: 2006 Nov 26, 15:47 -0000
DW has sent an interesting posting, [NavList 1750], with threadname "Lunars with and without altitudes". In fact, though, it's more relevant to the original thread, "Latitudes by lunar distance", so I have tried reverting its threadname back again. It may indeed be the answer to my request for a detailed and numerical account of how to work Frank's proposed method. But there are some snags... One is the unfamiliarity (to me, and no doubt, to others) of the "CA" programming language he has used. I can see it has some virtues, in allowing algebraic expressions to be constructed using numerator and denominator. Though no programming pundit, I have managed to deduce much of what that program does, as others have too, perhaps. But it would be useful to have available some definition of the syntax of that language. Can DW point us to where that can be found? It would also be helpful to see some comments added to the prgram, to flesh out the description he has given. A particular difficulty, for me, was in expression (%i10), which I didn't manage to translate into its equivalent in normal algebra, without knowing the detailed rules for the notation.. DW's final results are in the form of two position-lines, based on his assumed position of N38d, W70d. One, using star1, Pollux, gives a position line at right angles to a displacement through 0.61094 d in the direction 63.83353 degrees. The other, using star2, Aldebaran, gives a position line at right angles to a displacement through 0.69511 degrees in the direction 5.789598 degrees. By roughly plotting I get their intersection to be at about N38 39', W69 33', not that far from Frank's own deduction of N38 46', W69 25'. But it's somethig of a surprise that there should be any difference at all, based as they are on the same observations. One likely reason for that difference is the use of Astronomical coordinates taken from the Nautical Almanac, given there to the nearest 0.1 arc-minutes. Where Frank's data comes from, he hasn't told us yet, but my guess is that he is presuming that more precise coordinates have been somehow made available to an observer. But there are other details, that need to be thought about, when such great precision is being sought. 1 Moon semidiameter. DW has appled a correction for this, outside what his program does, before he has entered the details. If the HP is as stated, 58.4', then the SD associated with that should be 15.9', as I make it. To that has to be added a figure for "augmentation", which varies with the sine of the altitude, and has a maximum value, at the zenith, of about 0.3'. From his stated numbers, DW appears to have applied an overall SD correction of 16.2', which may indeed be correct, if the Moon altitude was high. But Moon altitude wasn't given (the horizon wasn't visible), and has to be calculated or estimated, if only very roughly. That step (allowing for semidiameter, augmented suiably) should really be part of the program, using a true Moon altitude calculated within the program. 2 Refraction. Here, DW has called on the expression given in Meeus 16.2, though neglecting a third-order term. The trouble with that, is that Meeus states that below 15 degrees it will "give inaccurate, or even completely meaningless, results". Yet it has been used to correct star altitudes, but really, the program should check first whether those (calculated true) altitudes are within the usable range. Or better, perhaps, use some adaptation of Meeus 16.4, which is much less limited in angular range.. I am not saying here that using formula 16.2 gives rise to significant error in this case; just that, under some circumstances, it could. 3. Parallax. DW uses an adaptation of the simple parallax formula (HP cos alt) which is rather more complex, and presumably this is to make the correction "backwards", on the basis of true calculated altitude rather than observed altitude. I seem to remember that we have discussed that matter, on Nav-L, in the past. Can he offer a reference? However, there's also a "standard" correction, which is often neglected in navigation, but needs to be taken into account when working to maximum precision (as we certainly are here). This is the reduction in parallax on account of the Earth's ellipsoidal shape, which increases from zero at the Equator to 0.15' at 60 deg latitude. I can't find that in the program, but haven't delved into every cranny. Is it hidden somewhere, implicitly, in one of the equations? It may be that DW's program (perhaps with a bit of further development) will indeed provide the guidance to Frank's method that I have been seeking, as an implementation of it, and I wonder whether Frank will give it his endorsement as such. George. contact George Huxtable at george@huxtable.u-net.com or at +44 1865 820222 (from UK, 01865 820222) or at 1 Sandy Lane, Southmoor, Abingdon, Oxon OX13 5HX, UK. --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to NavList@fer3.com To , send email to NavList-@fer3.com -~----------~----~----~----~------~----~------~--~---