Very late to this thread, so a few quick comments:
IMO Rock glacier should be a separate tag not a sub-tag. I’ve refrained from mapping some well known ones in Graubunden. There’s a project to create an up-to-date inventory for them in Switzerland.
Usage of natural=scree is a reasonable way to start mapping them which can be refined as their existence becomes better known.
glacier:type=rock is a great example where undocumented subtagging can change the generally understood meaning of the main tag. I really dislike this happening, even if understandable. I think such usage should be discounted when evaluating “most widely used tag”.
In addition glacier:type=rock is disjunct from the other values, which are concerned with position of ice glaciers.
Boulder fields: I’m all for a distinct tag. Citta dei Sassi is anything but scree, the boulders are house-sized (similar boulders exist in the woods under the Goldauer Bergsturz and no doubt elsewhere with massive rock falls since the end of the Ice Age). There are also interesting boulder fields created during peri-glacial times in Pennsylvania (don’t have my Roadside Geology to hand, but Hickory Run is one example) and Sweden.
Absence of rendering for new tags does tend to favour the two current approaches.
Practically there is only one mapper in Chile who created most of glacier:type=rock, that late in a political debate there. I see that as a laudable effort involving openstreetmap in furthering environmental goals. A second mapper joined a bit later, also from South America and created much of the rest of such mappings. Perhaps this political statement already served its purpose, so the mappers might be willing to retag? Someone would have to ask. But retag to what? Might geological=rock_glacier be the answer, what do you think? The way I read what I wrote there, natural=scree is the second tag. (Something that will make live very hard on data consumers?)
It might not be the answer, but it is certainly a step towards one. (Had realised I hadn’t mentioned it in last night’s post and had planned to add it).
My first impression of geological - I found it unwieldy, why not just geology. A bit more processing later: geology is like landuse, geological like natural
When I mapped a few rock glaciers recently, I even went so far to turn a big scree into a multipolygon to carve out the rock glacier parts - completely unnecessary, geological=rock_glacier can overlay natural=scree. In some cases perhaps it can even overlay natural=fell, remember only the lowest part must be unvegetated to mark an active rock glacier.
Yesterday I was hiking at Fontainebleau, a site that is well known for its boulder climbing. I was surprised to find that some blockfields had been tagged natural=bare_rock, while others are just not tagged at all. I checked after coming back home and found an exchange on help.openstreetmap.org where you mention natural=boulder_field. This makes a lot of sense to me.
To be honest, I was a bit surprised that this site is not better mapped given its international popularity.
@StC: yes, completely forgot about Fontainebleau, although the origins are somewhat different from the ones in alpine areas. Also some rock features in the forest tend to appear as little ridges raised above the forest floor (IGN map).
… when actually very few of the blocks rise above the trees, since said trees always find a way to grow on the blocks themselves. I guess this can raise debates between mapping it asa forest vs mapping it as a blockfield.
I saw a few days ago that someone has created a wiki page for natural=blockfield, which seems to match this discussion. What is the perception here about this tag value?
Concretely, I have all these natural=bare_rock at Fontainebleau that I’d like to convert to natural=blockfield but I’d like some confirmation that it’s a good idea.
A blockfield (or felsenmeer) is a different type of formation than a rock glacier. I think the OSM Wiki page for natural=blockfield was well written.
Blockfields form in place where exposed bedrock cracks and is weathered to form individual boulders. The rock itself does not move, but erosion along the cracks (or “joints”) breaks the rock apart into separate boulders.
A rock glacier is a type of “mass wasting” (i.e. landslide) where broken rock slides down a slope. In the case of a rock glacier, the movement is aided by a base layer of permafrost that the rock cover helps to maintain.
Because of their movement downslope, rock glaciers often have a distinctive appearance in aerial imagery, with one or more rounded lobes at the foot and a rounded, wrinkly surface. There’s a good example of this appearance on Wikipedia: Rock glacier - Wikipedia
Blockfields don’t move, so they don’t get those lobes or wrinkles.
As for converting areas of natural=bare_rock to natural=blockfield, I’d say as long as you’re using good judgement and verifiable sources, the change is appropriate!
Thanks for the reminder.
What I was trying to describe in that article was “bog” (wetland), not just “woods”, so I’ve corrected it by using the “wetland=bog” tag.
There were a couple of other things I wanted to clarify, which I’ve done on the ‘Talk’ page, so please add them to the discussion.
