Now I bet you all thought that a contour is a line that joins points of equal height! Now read on.
Message to Tracklogs support:-
Reply from TracklogsHi. I have recently bought the Dark Peak map 1:25000 scale.
I find that the heights of the contours do appear to be digitised very accurately. For example, if I run the cursor along a contour, the digital height displayed bottom right should be fairly close to the contour height. Instead it seems to be up to 15 m out (in some worst cases). One spot height (the Knott at SK046893) is over 20m out, displaying a digital height of 434m for a printed spot height of 454m.
This cant be right. How can the calculated climbs and descents be considered anywhere near accurate with these kinds of discrepancies? Indeed, if I run a track along a 1 mile stretch of the Marple canal I get an ascent of 13m and a descent of 15m. (15m is the height of a five storey building).
Hi,
The whole elevation model vs contour lines is quite a sticky debate. We don't actually digitise the contour lines, instead we use a satellite elevation model from NASA. The reason we do this, and hold onto your hat here, is that the OS contour lines aren't quite as perfect as they are often assumed to be. In fact the contour lines have a claimed accuracy of around 5% which is not quite as good as the NASA data. A bit shocking at first but not too suprising when you realise the data being compared is 21st century satellite vs man with a stick in the 1930's. It's not just us either, the OS have discontinued their contour based elevation digital data pending a new satellite derived product.
Now the complicated part. Plan a regular route up and down hills and the rolled up elevation totals will be about the same however you calculate them. That's the beauty of the statistical errors. In fact take a GPS out and measure physically everything will be about the same. Good so far. Problems arise when you do either of two things. If you attempt to traverse along a known contour of constant height you will get a discrepancy as the statistical errors will tend to magnify, that's what happens on your canal run. The second issue is spot heights, the trig points from the OS stand out as beacons of accuracy within a sea of averages and this is not matched in the NASA data. We do post process the NASA data to distort the model around these known good points but it's imperfect.
So that's the situation at present, we make best endeavours to keep the information as accurate as possible and hopefully we can improve it all over time as more and more accurate information is gathered by the satellites. In the meantime don't worry too much about the canal route, the errors there won't be reflected in your hilly stats.
Personally, I would trust the man with a stick as much as the satellite.
Reply With Quote
