Forum: Questions And Answers About Geocaching
Topic: Techno Geek asking too many questions
started by: Volvo Man
Posted by Volvo Man on Nov. 17 2003,5:58 pmhow's the distance between 2 sets of coordinates calculated? is it as follows:
(difference in latitude) squared + (difference in longitude)squared = distance squared
naturally, degrees are converted to miles, in the above.
Which brings me to my next question, what is the exact distance of a single minute travelled (in Miles & Metres)., ie if travelled in lattitude, assuming no travel in longitude
I'm writing my own little database to plot my own personal travels, this way I can combine my finds from both sites and get accurate as the crow flies milage. I'm not asking for recomendations for software downloads because I like writing my own.
PS, this database program will be available for all free once I've finished it.
Posted by team_tar on Nov. 18 2003,6:03 am
Well, I don't think it's so simple. sqrt(lat^2+lon^2) is valid on SHORT distances, as it assumes a flat surface, while the earth is (approximated in coordinate system as) an triaxial ellipsoid with some given parameters (whic in fact define the ellipsoid used in your map datum). In fact, what your GPS (or the computer programs) does when calculating the distance is using some kind of algorythm to integrate along the shortest path (pseudomeridian, it seems to me is called) along the elipsoidic surface. The way surface is flattened is one of the core problems of mapping (while a map is flat). I don't work in geodesy so are unable to give you a figure of what "short distance" means... if you want to preserve the precision of the GPS measurement, which in general is in the order of 10 m, you probably could not go over 100km, likely less... in some book (or site) on geodesy and mapping you may likely find exact figures on this regard, and also the formulae to calculate distances along given elipsoids... which at the end is merely a geometrical problem, but likely in a specialized site or book you will find them in the best form to be applied to geographical coords.
About the length of a minute in lat/lon, we live in a better world because we have Joe Mehaffey ! Go to:
< http://home.online.no/%7Esigurdhu/Grid_1deg.htm >
As you see, the metric of a arc minute changes depending of the position...
Stop! Joe is even better... found also the formulae:
< http://gpsinformation.net/main/transformation.htm >
Posted by Volvo Man on Nov. 18 2003,5:12 pmThanks for the pointer.
Distance calculation isn't a problem, as that will come from arc traversed on a great circle and the mean circumference is accurate enough for my purposes (+/- 0.5% or less).
However, I had completely forgotten that the poles are stationary in my calculations, so mine would only work adequately within a few minutes of the equator.
fortunately the formula will be fairly easy to put into my database as it's written almost in spreadsheet layout.