OpenStreetMapping of caves

[Martin Laverty]

The Ordnance Survey marked the course of a small number of caves on its maps (eg Ingleborough Cave, Yorkshire and Fairy Cave, Glamorgan) many years ago, but this never became a standard practice.
I have just noticed that fairly large parts of Agen Allwedd and some other caves in the Llangatocjk area of South Wales have recently been added to OpenStreetMap by someone identifying as 'conifermapper':

OpenStreetMap

OpenStreetMap is a map of the world, created by people like you and free to use under an open license.

www.openstreetmap.org

The mapping symbol used is that for a tunnel [ https://wiki.openstreetmap.org/wiki/Key:tunnel?uselang=en-GB ] which seems inappropriate but a proposal for a natural:cave [passage] category was dismissed many years ago [ https://wiki.openstreetmap.org/wiki/Cave ].

I would be interested to hear other cavers' views on the desirability and implementation of such mapping...

 

[aricooperdavis]

Wow, that's an interesting development!

OSM (as a foundation) is pretty agnostic when it comes to what goes on the map - if it's geographic data it goes in. As such I think it would probably be happy with people adding passage data to OSM, as long as it doesn't run up against any other rules (such as respecting the intellectual property rights of the publishers of any survey).

However, the real world implications of this (which OSM as a foundation pretty much ignores as not in its remit) is that these caves will get more footfall, potentially increase callouts etc. This is already an issue for MR teams who end up rescuing people from "goat tracks" etc.

Personally, I don't like hiding information from people to protect them from it without them even knowing I'm doing so, so whilst I probably won't be uploading any data myself until I've decided how I feel about it, I also don't object.

 

[Tricky Dicky]

I think originally OSM required its surveyors to provide digital photos to show that they'd actually been to a location rather than just copying a map (or survey), preumably 'conifermapper' has been doing this. Certainly an interesting development....................

 

[ChrisB]

There are many specialist forks from Openstreetmap, such as openskimap, opencyclemap, etc. There a proposal here for opencavemap but it doesn't seem to have come to anything. I don't understand the different types of open data licences well enough to know what is permitted.

 

[asheshouse]

I use OpenstreetMap a lot, for cycling and trekking. I also began adding data to it when I realised that a lot of footpaths were not shown in the area where I live. Having looked at the Aggi details in the link above shown I cant see how the level of detail shown would be of any use to anyone. The detail shown is obviously based on copyright publications and no credits are provided, as far as i can see. Just serves to confuse the clarity of the surface information shown.

 

[hannahb]

Mapy.cz is good for cave entrances - someone said it uses OSM and other open map data but I don't know if that's true. It's great for holidays abroad, too.

 

[Martin Laverty]

I think asheshouses, for one, will agree with me that on https://www.openstreetmap.org/way/110284383#map=18/51.83548/-3.18058 conifermapper has mixed up the original and Ogof Gam enrances: what this shows of their caving background is unclear. Also, if you swap to OpenCycleMap you will see that the caves are apparently cycleable - no doubt a side-effect of the use of being classed as tunnels.

I agree with aricooperdavis that openness is good, but should be used carefully in a useful context (I would suggest that the Cambrian Cave Registry does this on its map). I also share asheshouse's view that the effect here in OSM (and man, but not all, its derivatives) is not useful to anybody, indeed the symbols used are far more intrusive than for the tunnels of the London Underground.

Showing cave entrances on OSM is not a problem, indeed it may curb the casual walker's interest in probing further into a hole - or at least find out more details first. hannahb seems to be correct that mapy.cz uses OSM as its base, but it has not (yet?) been infected by this digrammatic cave passage problem (which on further investigation goes as far west as Ogof Tarddiad Rhymni via Ogof Ap Robert, Chartist's and Crescent Caves, Ogofs Cynnes, and Blaen Onneu Quarry Pot (but omits Carno Adit - a valid tunnel, perhaps even cycleable - and its cave)...

 

[Martin Laverty]

I think asheshouses, for one, will agree with me that on https://www.openstreetmap.org/way/110284383#map=18/51.83548/-3.18058 conifermapper has mixed up the original and Ogof Gam enrances: what this shows of their caving background is unclear. Also, if you swap to OpenCycleMap you will see that the caves are apparently cycleable - no doubt a side-effect of the use of being classed as tunnels.

I agree with aricooperdavis that openness is good, but should be used carefully in a useful context (I would suggest that the Cambrian Cave Registry does this on its map). I also share asheshouse's view that the effect here in OSM (and man, but not all, its derivatives) is not useful to anybody, indeed the symbols used are far more intrusive than for the tunnels of the London Underground.

Showing cave entrances on OSM is not a problem, indeed it may curb the casual walker's interest in probing further into a hole - or at least find out more details first. hannahb seems to be correct that mapy.cz uses OSM as its base, but it has not (yet?) been infected by this digrammatic cave passage problem (which on further investigation goes as far west as Ogof Tarddiad Rhymni via Ogof Ap Robert, Chartist's and Crescent Caves, Ogofs Cynnes, and Blaen Onneu Quarry Pot (but omits Carno Adit - a valid tunnel, perhaps even cycleable - and its cave)...

Actually, zoom in far enough and see that mapy.cz has passage plague too...

 

[aricooperdavis]

The fact that the passages appear as cycleable and confuse the surface map is because there is no suitable way tag for cave passages. This isn't the fault of the contributer, I'm sure they'd use an appropriate tag if one existed.

 

[Martin Laverty]

The fact that the passages appear as cycleable and confuse the surface map is because there is no suitable way tag for cave passages. This isn't the fault of the contributer, I'm sure they'd use an appropriate tag if one existed.

Indeed, the use of highway and tunnel tags seems inappropriate: way and cave have been suggested in the past but apparently not approved (albeit they do seem to be usable, as in the examples in your link, which also show that information can be recorded at a level below what users will normally be able to see... )

A further example (legitimate as it is public access for the path, but with another track linking entrances) is from Mulu: https://www.openstreetmap.org/query?lat=4.0256&lon=114.8249#map=16/4.0261/114.8262
and what led me there: https://overpass-turbo.eu/?w="natural"="cave"+global&R - navigate to Mulu and you will see the Deer Cave survey outline overlaid, but not, I think, visible on OSM.