The Earth has a flatenning of about 1 in 300.  So this is the rough size of
maximum distance errors you can expect by assuming the Earth to be a sphere
rather than an ellipsoid.  For example, the distance between Washington
D.C. and Los Angeles is 3,711 km (2,004 nautical miles) assuming the earth to
be a sphere with a radius of 6371 km but 3,719 km (2,008 nautical
miles) using the best fitting (WGS 84) ellipsoid.
-- Richard Langley

On Thu, 16 Nov 2000, Fernando Cabral wrote:

>Hello Richard
>
>Minutes after sending my previous message (should this always happen?) I found 
>a
>page on the Internet which seems to give me what I need.
>
>I have checked it with some coarse calculations. Since I don't have my GPS
>here with me, I can't compare the results I got with the GPS'.
>
>Nevertheless, they seemed right to me.
>
>Previously I had been using two formulae given by J. Meeus in his book
>"Astronomical Algorithm" (if my memory is of any avail), one of which
>he says is not very precise. They both have agreed very well with
>the results given by the GPS. With the less precise I have obtained
>results within 0,3% and with the more precise numbers are the same.
>
>Now, in this page by Fiona Vincent, it seems I found a still simpler
>way to calculate both distances and azimuth.
>
>Can you give me an idea on what magnitude of error should I expect from them
>(perhaps due to the Earth flattening)?
>
>Thank you.
>
>- fernando
>
>[Image]
>
>
>
>
>
>Richard Langley wrote:
>
>> On Thu, 16 Nov 2000, Fernando Cabral wrote:
>>
>> >Hello Friends
>> >
>> >It's been a long, long time since I last disturbed you with my novice
>> >questions.
>> >I was just acculating credits do be entitled to ask the following
>> >question
>> >that has more to do with navigation than any other thing:
>> >
>> >a) If I am using UTM coordenates, what is the easiest way to calculate
>> >    the bearing from point A to point B in the chart;
>> >b) Same questions if  am using latitude and longitude
>> >
>> >When using UTM I have come accross a solution that works but I must
>> >confess I hate it because I don't think it is elegant and it takes a lot
>> >of
>> >time so I am sure there must be a better solution.
>> >
>> >For the UTM it is much simpler because I can always create a Pythagorean
>> >
>> >triangule whose sides are the difference of Northing and Easting of the
>> >points,
>> >so I have three sides and an angle. Now, if I make the origin point the
>> >origen
>> >of a Cartesian system I can find the Azimuth adding together the angle I
>> >found
>> >plus 0, 90, 180 or 270 if the destination point is on the first, second,
>> >third or
>> >fourth quadrant.
>> >
>> >I does work, but there must be a simpler way to do it.
>>
>> Nothing realy wrong with your approach but if you use the ATAN2 Fortran
>> function (or the equivalent in other programming languages or the Cartesian 
>> to
>> polar coordinate function on many calculators), you can specify the sign of
>> delta x and delta y and the azimuth will then be automatically reported in 
>> the
>> correct quadrant.
>>
>> >If what I have are the geographic coordinates it takes me much more time
>> >
>> >with the spherical triangles and I am never sure I've done the right
>> >thing.
>> >
>> >I've read some books on celestial navigation and position astronomy. I
>> >can see
>> >the solution is there, but it does not seem I have the expertise to
>> >convert all those
>> >useful information in a simple formula for this specific calculation.
>> >
>> >And I feel very unhappy when I have to fill out a couple of pages with
>> >ugly
>> >calculations just to find out an azimuth I can easily find with a GPS or
>> >with
>> >a protactor and a map!
>>
>> The use of plane trigonometry to find distances and angles is an important
>> feature of UTM.  But for highest accuracies one has to be careful about the
>> differences between geodetic and grid azimuth.  See my article in the 
>> February
>> 1998 issue of GPS World magazine for more details.
>>
>> -- Richard Langley
>>    Professor of Geodesy and Precision Navigation
>>
>> >Best regards
>> >
>> >- fernando
>> >
>> >
>> >
>> >
>> >
>> >--
>> >Fernando Cabral                         Padrao iX Sistemas Abertos
>> >mailto:[EMAIL PROTECTED]              http://www.pix.com.br
>> >Fone Direto: +55 61 329-0206            mailto:[EMAIL PROTECTED]
>> >PABX: +55 61 329-0202                   Fax: +55 61 326-3082
>> >15? 45' 04.9" S                         47? 49' 58.6" W
>> >19? 37' 57.0" S                         45? 17' 13.6" W
>> >
>> >
>> >
>>
>>
>> ===============================================================================
>>  Richard B. Langley                            E-mail: [EMAIL PROTECTED]
>>  Geodetic Research Laboratory                  Web: http://www.unb.ca/GGE/
>>  Dept. of Geodesy and Geomatics Engineering    Phone:    +1 506 453-5142
>>  University of New Brunswick                   Fax:      +1 506 453-4943
>>  Fredericton, N.B., Canada  E3B 5A3
>>      Fredericton?  Where's that?  See: http://www.city.fredericton.nb.ca/
>> ===============================================================================
>
>--
>Fernando Cabral                         Padrao iX Sistemas Abertos
>mailto:[EMAIL PROTECTED]              http://www.pix.com.br
>Fone Direto: +55 61 329-0206            mailto:[EMAIL PROTECTED]
>PABX: +55 61 329-0202                   Fax: +55 61 326-3082
>15? 45' 04.9" S                         47? 49' 58.6" W
>19? 37' 57.0" S                         45? 17' 13.6" W
>
>

                                                                                
=============================================================================== 
 Richard B. Langley                            E-mail: [EMAIL PROTECTED]  
 Geodetic Research Laboratory                  Web: http://www.unb.ca/GGE/
 Dept. of Geodesy and Geomatics Engineering    Phone:    +1 506 453-5142      
 University of New Brunswick                   Fax:      +1 506 453-4943      
 Fredericton, N.B., Canada  E3B 5A3        
     Fredericton?  Where's that?  See: http://www.city.fredericton.nb.ca/
=============================================================================== 

Reply via email to