Approach surface is approximately 3.5m short across 15km

[Marc_Graham]

Hi there,

I just generated a 15km long approach surface using the customise json tool:

and it is short by approximately 3.5 meters.

The blue line is the extent generated by the ArcGIS Airports GP tool:

The red area is from a dwg generated using an alternative CAD based method:

All Data is projected in NZTM/NZVD2016, both in ArcGIS Pro and in the input CAD dataset.  The measurements are made in ArcGIS Pro 3.3.5 and all scenes/maps are also in the same coordinate system.

Any ideas what could be causing this difference?

Thanks,
Marc

[Koeller_Christine_A_-_DOT · Accepted solution]

When we first started using the Airports extension to create surfaces, we were noticing distance differences in some of our surfaces as well (between hand drawn and tool generated surfaces). It wasn't as large as 3.5 meters, but we also weren't drawing anything as long as your surface. When we met with Esri, they explained that their tools output in Geodesic, not Planar distances. I thought it was unusual, but I don't really know that it's wrong to use geodesic outputs because I don't think FAA specifies that detail. I also couldn't find this detail in the documentation for the Airports tools, so that's why we asked Esri. I'm curious if CAD would even draw in Geodesic distances now that you mention it. Every time we get CAD drawings from a consultant, the distances appear to be planar so I'm guessing most people aren't using geodesic. When you measure, see if you can switch from Planar to Geodesic, I'd be curious if that gives you the right measurement. 

[SCourter]

Hello Marc,

@Koeller_Christine_A_-_DOT is correct about the measurement. 

When measuring the approach and take-off surfaces, you want to use geodesic instead of planar. Planar measurement is taken on a flat, two-dimensional surface.  Geodesic measurement takes into account the curvature of the Earth and is better over long distances.

When you set up your customize OIS, what Runway Specification and Runway Classification did you use? See the example below. 

And when you ran the geoprocessing tool, did you input the same Runway Classification?

[Marc_Graham]

Hi @SCourter and @Koeller_Christine_A_-_DOT 

Of course you are both right, when measuring in geodesic I get the correct distance.

We created this surface using Precision Approach Category I Code Number 3 4

I am interested to know what the standard approach for this sort of work is.  We have traditionally built our surfaces in CAD, and I don't know if this has come up before.  To be honest 3.5m across 15km isn't worth worrying about, however it is good to know this is the cause.

Thanks for your help,

Marc

[Marc_Graham]

Hi @SCourter 

Is there any way to override the geodesic calculations to use the projected coordinate system of the input data?

The geodesic nature of the calculations and the resulting slight differences in the output from what is expected in the PCS is causing issues with our workflows.

Example:

In NZ the inner horizontal surface and conical surfaces are defined from the edge of the strip, not from the runway centerline. I can approximate this quite well by offsetting the runway to either side by half the strip width (eg. 280/2 = 140 either side). However all the ArcGIS editing tools work in planar coordinates, not geodesic, and so my new runway centerlines don't line up with the generated OIS runway strip edges.

These issues just compound at greater distances, as I mentioned in my original post.

Another example:

say we want to generate labeled points showing distances from the end of the runway strip. This measures from the geometry, which uses the PCS:

In my previous response I said 3.5m over 15km isn't enough to be concerned about, but given that there are going to be a bunch of little issues that potentially arise from it, I'm starting to think otherwise.

Thanks,

Marc

[SCourter]

@Marc_Graham, Considering the complexity of your workflow and issue, it would be best if we addressed this via email. Could you please send me an email at scourter@esri.com?