# Large objects of fixed size that can be used for measure control points

I am looking for some real world objects that has rather large sizes where the exact size is known such they can be used for controlling different implementation of length measurements in GIS application.

The quesstion started from here: Should we always calculate length and area in lat,lng to get accurate sizes/lenghts of real world measurements

So if anyone has a list of objects that has lenghts above 1km in the real world that has easily identifiable points in maps such the length of them can be measured easily.

• I think the best thing to do is define your own coordinate system using an equidistant, equal area, or conformal base and modifying the central meridian to pass near the centre of your area of interest with a couple of parallels defined about 2/3's of the way from the centre of the area of interest. This would give you 3 projected systems one for bearing, one for length, and one for area. Not sure what "measurement" you are getting in a geographic system, but it should be in degrees or radians I would have thought. I have done this on a feature by feature basis before, but have seen a script – Mike Oct 14 '16 at 21:15
• our discussion started in the linked question, where we simply just need some argument/evidence for which model fits the real world best when asking the question: From Point P, where is P2 15km from P1 a given bearing. We used a local projection UTM, but a question was raised if using the great circle is more accurate then pytagoras in local coordinate system – Poul K. Sørensen Oct 15 '16 at 13:34
• @mike Thinking a bit more about what you wrote, creating our own local coordinate system sounds like a really good idea. Can you share a bit more on how I would do this, or link to some starting matrial. I am working with proj4 in javascript, so maybe there is something there – Poul K. Sørensen Oct 15 '16 at 14:43
• Basicly what I have is that I have a point of interest and everything we work on is within 25km radius form this point. So maybe its as simple as creating a proj4 string that uses this point of interest somehow – Poul K. Sørensen Oct 15 '16 at 14:47
• I am sure I saw a script for doing this when I was looking at it, but I can't find anything now. What you are saying is essentially it though. In my case I always already had a feature. In your case you would have to estimate a centre and extent, define your central meridian at your centre longitude and your parellels 2/3 of the way to the north and south extents. Create a conformal projection and create a point out your distance at the bearing you need. Then create a equidistant projection (same meridian/parallel) and adjust your point to the actual distance along the new bearing. Maybe :) ? – Mike Oct 17 '16 at 22:35