Updating Heights

Written 27/08/18

We've been working behind the scenes on updating the height data which we use in our GPS mapping application. This has involved merging NASA SRTM height data for Ireland and the Isle of Man with the latest release of Ordnance Survey's OS Terrain 50 data for Great Britain (previously we were using SRTM data for Ireland and an older release of the Ordnance Survey's height data for Great Britain and Isle of Man).

As a result of this exercise, and one the main reasons we did it, is improved relief shading on our WayMaps tiles including all of Shetland for the first time. Before we were relying solely on SRTM data when processing relief shading and that stops at 60° North. As a result Shetland looked like this:

60° North crosses the southern end of Shetland so, as you can see, only the southern tip had relief shading - the rest was plain grey. With the new release we've switched to relying on OS Terrain 50 so we have height data for the whole of Shetland and it now looks like this:

One consequence of this work is that if you query the height of a particular point you may find that it has changed slightly.

This is not because land has really changed in height but because OS, in their infinite wisdom, have changed the 50m x 50m height grid they provide us with. It was previously given to us at easting and northings ending 00 or 50 and we estimated the heights at other locations from that. It's now coordinates ending 25 and 75 so with slightly different heights each time. Again we're estimating heights for other locations but as the underlying data has changed so have the results.

They're also now giving us heights to the nearest 10cm rather than the nearest metre.

However the differences are small, in the order of 10m maximum and typically more like 1m or 2m, and the overall effect on profiles, and total ascent and descent seem to be negligible - certainly in all the test cases we're tried - so you shouldn't notice any significant differences if you look back on past tracks or routes (and if you do see a difference the new figures are likely to be more accurate).

«