-
I see what you mean, but I don't think the average elevation in itself is the correct parameter to use as it just tells us how high above sea level the average is.
You want something like the lowest deviation from the median. :)
We need more data, but I can't provide it.
Edit: am just looking forward to some real mountains and proper racing.
drøn-
As a statistician by trade, I could talk all day what the best measure is. Let's leave it at this, tho
-
I think what you want is a metric like "mean topographic roughness". Topographic roughness would be calculated per pixel and captures the maximum elevation difference between a central pixel and all immediately surrounding pixels. You would then take the average value across the landscape of interest. Of course this kind of metric is heavily influenced by the scale of analysis and so you could then move the discussion along to an enjoyable debate about the appropriate scale of the pixels to be used. However, just for reference, last time I did this kind of work (>10yrs ago) most open access digital elevation models were available at 90m resolution.
Edit: It looks like this is essentially what this database does, calculated at a 30 arc second grid scale (~930m at the equator). On that metric, the Netherlands wins the "flatness contest" by a considerable margin... Netherlands 3.7m Vs Denmark 18.9m.
Whopper
ewanmac
It doesn't make sense to take the highest and lowest points because that doesn't take into consideration the "lumpiness" of the terrain. What if a country has a lower spread between highest and lowest points but goes constantly up and down vs another that is constantly flat (let's say in 99% of the surface), but has a highest maximum elevation (in 1% of the surface)?
Surely the latter is more flat, if the average elevation is lower than the former.
Hence, the average is a more significant measure of flatness