# Length difference between Google Earth and ArcGIS for Desktop?

I have drawn a complex path with multiple vertices in Google Earth (GE) Pro and its measurement property is shown as being 2,460 nm. It is located East of Australia.

If i import this KML into ArcGIS and try to measure it, i get different results: Using the Calculate field and the formula !shape.length@nauticalmiles! using WGS84 GCS, i got 2,554 nm.

Thinking that i should better use an equidistant projected coordinate system to get the length of my path, I redid the measurement and got 1,966 nm with the World Equidistant Cylindrical PCS.

I also played with the Plate Carree PCS and got 2,554 nm.

Am i wrong to use an equidistant projection for this calculation as it is normally supposed to maintain the distance?

Can i also consider GE length as being the correct value? I used external tool such as KML path measurer and got 2,467 nm.

What would be the correct process to do such a basic calculation?

• Every projection is a model so the measurements will differ slightly between them. You shouldn't be attempting to get a measurement from WGS84/Geographic. If you want to get a true 'accurate' measurement you should be considering the geoid, which is maths far to complex for me! – Michael Stimson Feb 11 '15 at 5:55
• Try running it through the Great Circle mapper - gc.kls2.com - it accepts lat/lon (see the FAQ). – GIS-Jonathan Feb 11 '15 at 15:32
• – Chris W Feb 11 '15 at 21:58
• Yes, it is incorrect to use an Equidistant projection. Unless your path is a straight line emanating from the origin of the projection, the distance will be wrong. Plate Carree (which you used twice, under two names) is even worse. The link offered by @ChrisW addresses the use of GE for distance measurement. – whuber Feb 14 '15 at 15:39
• Thanks Whuber. Which projection should i use in ArcGis in that case which would give me a correct value? My path is around the 170E, 0N. – Sylvain C. Feb 15 '15 at 23:19