Tom White, May 2025
My family walked up Tor y Foel in the Brecon Beacons when the children were small, and they weren't very impressed by the number of false summits (points that looked like they were the top, but weren't). The fact that I forgot to bring the chocolate didn't help either.
This got me thinking - what shape would a hill with the most number of false summits be? And also - how many false summits were there on Tor y Foel?
To answer the second question first, I wrote a program to analyse a GPX track of a recent walk up Tor y Foel. First, the code converts the GPX data into an elevation profile - basically a plot with distance on the x-axis and height on the y-axis.
Then, for each point on the route, it calculates the highest visible point on the elevation profile. This part is tricky and relies on the idea of a visibility polygon. Luckily, there are libraries to compute visibility polygons, but they are quite computationally demanding.
You can see the result in the interactive visualization below (best viewed on a large screen!). As you move your mouse over the diagram from left to right to simulate climbing up the hill, it shows a line connecting the point on the hill to the highest visible point. This is what you would see as a summit if you were out on the hill. The last one is the true summit, all the others are false summits.
The chart also shows a number of thin black peaks in the bottom right. These represent the locations of the most prominent summits for the walker: the higher the peak, the more points on the route have this point as an apparent summit.
By counting the black peaks, we get an idea of how many false summits there are. For Tor y Foel there seem to be five, plus the final, true summit. The most prominent false summit with the largest peak is not the last, and there are two more false summits after that one, which may explain why there seem to be so many false summits when you are climbing it.
Route Upload: It would be great if you could upload a GPX file from a walk and generate a chart like the one above. Unfortunately, the current implementation relies on some quite compute-intensive Python code that doesn't run in a browser, so this would need more work.
3D View: This way of finding false summits converts the route to a 2D elevation profile, which may not always work well. For Tor y Foel the ascent is roughly a straight line, but for a route which loops round, like a horseshoe, this technique is not very accurate and 3D methods would be needed. This is much harder from a computational point of view, but would be a great follow-on project.
False Summits App: Another thing I would love is to somehow use the visualization when you are out in the hills. Using augmented reality to superimpose computed false summits on a display, perhaps a bit like the brilliant PeakFinder App? This too is a challenging project, and relies on getting 3D working first.
To go back to the beginning - when I started thinking about false summits, I was drawn to the idea of what the theoretical "worse hill" would look like. Read about the horrible hills you certainly would not enjoy climbing on this page.