landcover=trees is used to mark the presence of trees. It does not imply the use nor origin of the trees.
The tag is distinct from landuse=forest (Managed forest or woodland plantation) and also distinct from geographical features as described with natural=wood (Natural primeval woodland)
https://wiki.openstreetmap.org/wiki/Tag:landcover%3Dtrees
https://wiki.openstreetmap.org/wiki/Forest (See Approach 5)
We do already have a preset for natural=wood
landcover=trees seems redundant, and not really used very much.
I kind of think we should just stick to the one that is already most widely accepted, but happy to hear other opinions.
The need to have the landcover=trees preset arose because we detected mappings of tree groups within urban areas. These were mapped as mini woods (natural = wood) when they are not.
In that case the community decided to change the tag to landcover=trees, since the definition on the wiki is the one that best fits the reality even though the render does not support it. (See https://www.openstreetmap.org/changeset/51432496)
Anyway I understand your point, maybe the best would be to map these areas as gardens and each tree individually (natural=tree), but what is certain is that natural=wood is not right in this case.
I guess we could just add a landcover=trees tag to the existing natural=wood and landuse=forest presets. There's no harm in that right?
According to taginfo landcover=trees is mainly used in combination with natural wood or landuse=forest, but landcover=trees is essentially redundant in this case. Data consumers can imply a landcover=trees in a forest or wood anyway, where no other landcover is tagged.
A forest might contain areas where landcover=trees is wrong, e.g. a cutline. Therefore we must not add landcover=trees to every forest.
The real intention of the landcover=trees tag is a separate use for an area of trees outside a wood or forest, even if this use isn't the most frequent one. This needs a separate preset, but the forest or wood presets need to get priority in case of the combined use. E.g. a forest covered with trees is still a forest and not just a area of trees.
Here is why I don't really want to add this tag as a new preset:
tag | uses
--- | ---
https://taginfo.openstreetmap.org/tags/natural=wood | 4,346,557 uses
https://taginfo.openstreetmap.org/tags/landuse=forest | 3,306,152 uses
https://taginfo.openstreetmap.org/tags/landcover=trees | 11,128 uses
natural=wood is used, in practice, for all kinds of tree cover, (not just "primeval woodland" - why do people think this?). I see it used pretty frequently for small groups of trees even in urban areas.
landuse=forest is also used, in practice, for all kinds of tree cover, but preferring towards places where the trees are managed by forestry.
I just don't see how the new landcover tags solve any problem not already handled by the natural tags.
I think you are right @bhousel, landcover=trees is too generic and on the other hand natural=wood is used as a de facto for all kinds of tree covers.
I kind of think we should just stick to the one that is already most widely accepted, but happy to hear other opinions.
馃憤
Most helpful comment
Here is why I don't really want to add this tag as a new preset:
tag | uses
--- | ---
https://taginfo.openstreetmap.org/tags/natural=wood | 4,346,557 uses
https://taginfo.openstreetmap.org/tags/landuse=forest | 3,306,152 uses
https://taginfo.openstreetmap.org/tags/landcover=trees | 11,128 uses
natural=woodis used, in practice, for all kinds of tree cover, (not just "primeval woodland" - why do people think this?). I see it used pretty frequently for small groups of trees even in urban areas.landuse=forestis also used, in practice, for all kinds of tree cover, but preferring towards places where the trees are managed by forestry.I just don't see how the new
landcovertags solve any problem not already handled by thenaturaltags.