[jkorchok]

This is just a heads up, not a bug or problem. I've been noticed anomalys in terrain after applying HGT terrain maps:
1. Hills, where there are none in real life. I finally figured out that HGTs can include the height of highways, not just the ground.
2. HGTs don't necessarily align with other maps, they can be off by around 30-40 meters.
This picture illustrates both problems.
http://www.johnkorchok.com/images/HGTerrors.png
You can see what appears to be a hill in the green-gold area, leading down to the track on the right. In real life, the ground is very flat former swampland. The "hill" the the Highway 80 overpass that runs over where the tracks will be. So after applying the HGT, I'll have to excavate to restore the original terrain height in 1876.
You can also see Highway 80 as shown on the map and the same rise on the HGT. They don't match, with the HGT showing the overpass rise as being about 30 meters too far south.

HGTs are very convenient, but if you're trying to be precise, a good topo map is still a necessary cross-reference.

[ebnertra000]

I've noticed this as well. I've had to eyeball a bridge or two, and even a tunnel. In Northern Minnesota, the terrain seems to be about 50-75m west of a marker at the same position. The Erie Mining overpass over DMIR's Hinsdale Spur was a dead giveaway when the marker for one end of the bridge was halfway into the valley the trestle spans, and the other was well outside the valley on the other end

[Goku]

I think that HGT has offset 1/2 of resolution. So, if resolution is 30m, then it's 15m off and when resolution is 90m, then it's 45m. It was never tested.

[SP 0-6-0]

Can it be fixed though?

[ebnertra000]

I doubt it, but it wouldn't be hard to compensate for once you know it's there

[jkorchok]

I made this alternative terrain.ace file that is a simple grey grid. It makes it easier to see terrain features when route building. Enjoy!

Terrain Grid

http://www.johnkorchok.com/images/terrain.png

[jkorchok]

I found a source of better-quality HGT maps at USGS EarthExplorer. A higher level of detail than what you can download from ViewFinderPanorama, the files are also about 9 times larger than the usual ones. Because they're larger, I'd save these for terrain close to the route. You do have to create an account, then puzzle out the user interface:

Under Search Criteria tab, it's simplest zoom in and click on the map for the area you want to download.
On the Data Sets tab, expand NASA LPDAAC Collections, then NASA SRTM (SRTM 3) Collections and check NASA SRTM3 SRTMGL1. The 1 stands for 1 arc-second.
Then click on the Results tab. Each map will have a small download icon. Choose Standard Format which will have a 7 - 10 MB file size.

[Genma Saotome]

1 arc second? That's 90m resolution. You can get 1/9 arc second from USGS (in IMG format) which is 10m resolution and they're slowly working towards 1m resolution.

1 arc second resolution is like, oh 20-30 years old. IMO worthless for anything that isn't almost flat.

[jkorchok]

 Genma Saotome, on 03 May 2018 - 04:48 PM, said:

1 arc second? That's 90m resolution. You can get 1/9 arc second from USGS (in IMG format) which is 10m resolution and they're slowly working towards 1m resolution.

Thanks for the tip. I'm aware that other formats have higher resolution, but I was writing about the available HGTs. 1 arc-second is approximately 30 meters and the current data is 18 years old.

[Lindsayts]

In Australia the highest resolution terrain data one can get for all of Aus is 1/3rd of an arce second (30m aprox), one needs to select the format when ordering the data ( from http://www.geoscience.gov.au/ ), the best format availible being geotiff (the others are all vector formats). Unlike HGT files the position of data is the centre of the 30metre aprox square, where as in HGT files its at the corner of the square.

There is 1m resolution Lidar data availible but only over a VERY VERY ;limited area.

Lindsay