2

We use the Dallas-Ft Worth rasters for our internal processes. We use publicly-accessible data to superimpose markings representing runways and runway ends (the dark gray line and green dots on the uploaded image, respectively). The coordinates for runway ends, for instance, perfectly match the data found here. But for some reason, these markings don't match up with the FAA map. Not even an .osm file from OpenStreetMap (rendered as the transparent purple airport) matches up with the FAA map.

Why is that?

Is this an issue with how ArcMap renders the Dallas-Ft Worth raster from the FAA?

Is there some sort of configuration we need to tweak in importing the raster into ArcGIS?

osm, runways and runway ends

  • I looked at your raster compared to OSM, and it looks like it's simply not well-orthorectified. (Or the airport has changed since the raster was created). You'll probably have to re-orthorectify it, which isn't too difficult considering the many number of control points you'll have. – Jon Aug 2 '19 at 22:40
  • I should note, though, that they were much better-aligned than what you show in your photo. I used QGIS, dragged your raster in first, then loaded an OSM layer. You might have a rendering issue on top of the orthorectification one. – Jon Aug 2 '19 at 23:05
  • I'm admittedly new to GIS development, so I had to research on what you mean. Is this what you're referring to? If so, then it seems perfect for what I need.. I just need to: (1) note down the parcel points for each runway on the FAA raster, (2) note down the control points of the runway ends using the FAA runway information, (3) link the parcel points with the control points, (4) run the adjustment tool Does this make sense? Is my understanding correct? – CJ Virtucio Aug 3 '19 at 21:25
  • Also, I tried loading the raster and runway ends unto QGIS, and the result was identical to the uploaded image. – CJ Virtucio Aug 4 '19 at 21:27
  • You are correct--the alignment is the same. I was wrong in my second comment. Orthorectification is the process of choosing control points on your raster that map to points in a known-accurate CRS. Basically, you're distorting your image so that the control points between your image and a known base image align. Plenty of resources for how to do this in Arc and QGIS. – Jon Aug 5 '19 at 17:26
1

I got in touch with the FAA since it really does appear to be a problem on their end. Turns out, it was:

Thanks very much for your inquiry. VFR Charts currently remain a manually compiled product. Charted items are in relative perspective to each other. This frequently results in features not depicted in their exact latitude/longitude locations. Additionally, the historical manual printing process was used for many editions (approximately 80) for the older charts. The photographic process from positive film to negative film for each edition resulted in a minor shift to the north. The cumulative effect over many editions would result in a noticeable shift. We are now at the threshold of producing VFR charts from a data-driven position. This will eliminate many of the issues that you notice. However, there will always be manual adjustments to maintain relative feature positions (obstacles/airports/etc.) on the proper side of linear features (roads/railroads/rivers/etc.). We will shift the DFW airport pattern to its correct position on the Dallas - Ft. Worth Sectional chart. Additionally, we will address those most egregious items when brought to our attention, or when discerned in the course of our compilation processes. Thanks again for your inquiry.

So the real fix is to wait for the FAA to release a proper API for their data.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.