topic Re: Distance between almost antipodes cities in Coordinate Reference Systems Questions

Distance between almost antipodes cities

GB — Sat, 16 Sep 2017 08:54:14 GMT

Hello!

I would like to calculate distances between different pairs of cities that are almost at their antipodes. I was wondering if ArcGIS takes into account that given that the globe is not flat, there is two ways to reach the antipodes and one option is shortest of the other. For example I can calculate the distance from Lisbon to Auckland with a line that crosses Asia, or via a line that crosses America. I believe that the distance calculated via Asia, in this case, is longer that the one I would have via America. Does ArcGIS take into account this issue? And how can I address this issue in my analysis? Thank you.

Re: Distance between almost antipodes cities

DanPatterson_Retired — Sat, 16 Sep 2017 11:09:54 GMT

Great circle distance calculations can be done within ArcMap or Pro.

It you want to incorporate code into your analysis, then I have an example of implementing Vincenty's calculation on GitHub, and discussed here. If your data are in tabular form then you can use the 'geodesic' option for the Add Geometry Attributes tool in ArcToolbox

Re: Distance between almost antipodes cities

GB — Sat, 16 Sep 2017 17:30:29 GMT

Thank you for your reply.

My original data come from excel, basically I have a table with the lat lon of a series of locations (more than 1000). I would like to calculate the distances of these locations with a specific european capital (which I can identify with lat lon coordinates).

Do I need any special shapefile when using the Add Geometry Attributes toolbox? If yes, where can I find them? I would like to add that as I have to calculate Great circle distance distances between this benchmark european capital and some locations in Antarctica. Your help is appreciated!

Re: Distance between almost antipodes cities

DanPatterson_Retired — Sat, 16 Sep 2017 18:38:24 GMT

make a featureclass by using the Excel to Table tool then use the X Y to Line tool to get geodesic distances

your excel table needs to be generically set up as

FromX FromY ToX ToY

that is, the X, Y (longitude latitudes) of each pair needs to form a column

Re: Distance between almost antipodes cities

GB — Mon, 18 Sep 2017 20:16:48 GMT

I followed your advice but I have a problem: it is not calculating the lines between my reference point and the locations that are in its left (all the cities in the Americas are not considered). Why does it happen?

Re: Distance between almost antipodes cities

GB — Mon, 18 Sep 2017 20:16:48 GMT

With reference to my previous question - I solved. I don't know why it happened, but in order to obtain the distances for all the locations, I have to put in the FROM X and FROM Y column the coordinates of the reference european capital, and in the column TO X and TO Y the coordinates of the other cities I had in my database. I was doing the opposite.

Re: Distance between almost antipodes cities

DanPatterson_Retired — Mon, 18 Sep 2017 20:31:45 GMT

So the suggestion worked, but you switched up the from to. Shouldn't matter however, 'not calculating' means it returned an error? a zero? a negative distance? Use the Vincenty suggestion then

Re: Distance between almost antipodes cities

GB — Fri, 03 Nov 2017 18:10:46 GMT

Can I add a new question? I would like to create straight lines between point, it is just a graphical issue, I am not interested in the accurateness of the distance now. Do you have any solutions? I think that xy to line doesn't allow me to do that.

Re: Distance between almost antipodes cities

DanPatterson_Retired — Fri, 03 Nov 2017 20:44:48 GMT

Did the previous solution work? you gave no indication

Re: Distance between almost antipodes cities

GB — Fri, 03 Nov 2017 21:49:54 GMT

Hi,

I tried all the options that xy to line permits, they work but I cannot obtain straight (Euclidian) lines. The problem is expecially for the line from a place in France and Fiji, If I am not wrong, as I would like to have a line crossing Asia, while xy to line creates a line that crosses the Atlantic Ocean. Thank you.

Re: Distance between almost antipodes cities

DanPatterson_Retired — Fri, 03 Nov 2017 22:17:15 GMT

Shift the coordinates of both locations by a longitudinal amount to put Fiji on the negative side of the date line. It doesn't matter as long as the latitudes remain the same

Re: Distance between almost antipodes cities

GB — Mon, 06 Nov 2017 09:38:55 GMT

Sorry but I cannot understand.

This is an extract of my excel file:

Location tx toy fromx fromy

(point in nz) 174,7633315 -36,8484597 -1.051945 43,708608

How should I modify it?

Thank you.

Re: Distance between almost antipodes cities

DanPatterson_Retired — Mon, 06 Nov 2017 09:53:43 GMT

move 174 longitude by 7 degrees to the east, it becomes -179 degrees longitude since it crossed that imaginary 180/-180 line. Move your other longitude by the same amount (the amount isn't relevant) and their 'relative' positions will remain the same. Do the calculation

Re: Distance between almost antipodes cities

GB — Mon, 06 Nov 2017 11:59:56 GMT

Thank you. I tried, but it doesn't work for two reasons: 1) it goes already on the east side; 2)I'll obtain a line that does not depart from the point I want.

Re: Distance between almost antipodes cities

DanPatterson_Retired — Mon, 06 Nov 2017 15:04:06 GMT

Good luck then

Re: Distance between almost antipodes cities

GB — Tue, 07 Nov 2017 14:22:18 GMT

Sorry I cannot understand your answer. Does it mean that there aren't alternatives?

Re: Distance between almost antipodes cities

DanPatterson_Retired — Tue, 07 Nov 2017 14:28:06 GMT

I have tried to explain alternatives, but I will give someone else in the queue a chance to jump in with suggestions.