Sunday, 2 February 2014

Mapping a Hole

I don't get to do much countryside mapping these days, so when I learnt that my sister and friends had rented the Youth Hostel at Bretton it seemed like a good idea to pay a visit for the day. My motivation was that just before Christmas Day a very large hole had appeared in a field close by (~ 800m).

Sinkhole from Bretton Ridge:
© Copyright Graham Hogg and licensed for reuse under this Creative Commons Licence.

I added it to OpenStreetMap from a couple of pictures on the the BBC News site, along with a fair amount of detail such as field boundaries in the immediate area. However, I fancied seeing such an interesting phenomenon in the field. Also the ridge called Eyam Edge looked to have some interesting looking scrub vegetation: and I'm on the lookout for useful sites for trying out more detailed mapping of this type of thing.

Sinkhole from field footpath
© Copyright SK53 and licensed for reuse under this Creative Commons Licence.
We followed the road towards Foolow from Eyam Edge. The byway which runs along the foot of the edge was somewhat ineffectively closed, but having driven up the road we knew the best views were further down the road. We were able to follow a footpath a couple of fields over from the hole. Even in that field there were obvious examples of mining workings (these turned out to be from a sough which drained a lead mine higher up the hill).

My own picture shows the type of detail which enabled me to estimate the size of the hole more accurately: field boundaries (mainly drystone walls), the position of the power pole(s), and the patches of gorse behind the hole. I mapped it with natural=sinkhole and cheating a little added a second way with natural=cliff. I needed to add area=no to the latter to get it to render.

One intriguing thing was that the power poles either side of the hole had been disconnected from the line and the power line is now supported by two higher poles each quite a bit further away from the hole.

Whilst examining the area closely it became apparent that there were several places which are probably older sinkholes caused by mining.

The scrub on the ridge proved to be just as I hoped. It is a complex mosaic with patches of Hawthorn, Gorse, Bracken and Briar. Some areas are wooded (not shown on the large scale Ordnance Survey maps), and others consist of tall herbs. The most notable of the latter were large stands of Rosebay Willowherb (Chamerion angustfolium).

When we reached the edge of the scrub we discovered that about 4 hectares had been bought by the local community and was been run as a combined nature reserve and heritage area. Originally this was the site of a couple of lead mines. It is now predominantly Hawthorn scrub, with patches of rough grassland with perennial herbs, such as Knapweed. At the top of this area the engine house of Silence Mine (nice name!) has been excavated and preserved. The paths in this area are not shown on my 1:25k OS map, of course they had been put on OpenStreetMap!

My plan to survey the scrub as we walked back along the edge were put on hold when a sudden squall appeared in the west. In the end we walked back in driving rain with thunderbolts flashing down close to the mast a couple of kilometers to our east.

I did get plenty of images of the countryside below the ridge which should help in more detailed mapping of the multitude of field boundaries. It's quite interesting to compare the OS map with OSM, with the extra detail which I added:

Excerpt from Ordnance Survey White Peak Explorer Map (1:25k) under Bretton Edge
compared with excerpt from OpenStreetMap for the same area.
Top map : © Copyright Ordnance Survey
Lower map: © Copyright OpenStreetMap contributors
The basic level of detail is similar: the Ordnance Survey have a little more information about former mining locations, whereas OSM has extra paths, information boards, more field boundaries, and the minor power lines. The interesting difference is in the mapping of the vegetation on Eyam Edge: this may be a feature that the OS don't revise that frequently, although the trees in the sharp angle of the two roads must be at least 20 years old (see the picture below).

Scrub and Woodland near Bradshaw Lane
© Copyright SK53 and licensed for reuse under this Creative Commons Licence.

One feature shown on the OS map, but not visible with OSM is Access Land. However, this patch of Access Land has been recorded on the OSM database.

This was an enjoyable little excursion, even if we got wet at the end: I got to see the new hole, found out a bit about mining in the area and established that Eyam Edge is a great place for tackling problems of mapping scrub on a fine scale. It is also very satisfying to see that the cumulative efforts of many OSM mappers we can achieve similar levels of detail to the Ordnance Survey's 1:25k maps. 

One other thing I noticed was that many roads and paths need adjusting. Many of these would have been mapped several years ago using old maps, old imagery or ropey GPS traces. Today we have many more GPS traces, much better aerial imagery, and a wider range of old maps, all of which mean that in places which are apparently well mapped it is often easy to improve the accuracy of what has already been mapped.

Whilst writing this up, a rather more spectacular event occurred South of Bolzano/Bozen when a section of cliff collapsed with a couple of huge boulders rolling right down the hillside : the first crushing a 300 year old barn and second only stopping a few feet from a house. The local authorities used a UAV to survey the whole area as shown in this neat video.



Of course everything is nicely mapped on OpenStreetMap too, although not everything is rendered:


Even the individual boulders have been added! Including the one which obviously rolled down the hill long before the house was built and the vinyard created.

This highlights OSM's ability to respond to events (as I've pointed out before), and encourages a deliberate focus on mapping a lot of detail around the principle locus of an event. As OSM becomes more detailed, it becomes easier and easier to incorporate the more dramatic changes in geography.

No comments:

Post a Comment