I’ve been having a discussion with another mapper about how to properly tag US national forests, specifically with regards to inholdings. For context US national forests contain both public and private land, and inholdings are the private land within the boundaries of a national forest.
This leads to there being two sets of borders within for the forest:
The forest boundary as defined by the forest service
The lands owned by the forest service (USDA) within the forest boundary
I can see two possible solutions for this (and I have seen in practice)
The forest boundary is boundary=protected_area, the owned land is something else (I tagged it as boundary=legal, which has decent use elsewhere, but maybe there is something better to use)
The owned land is boundary=protected_area, the forest boundary is something else (I have seen boundary=administrative)
There’s the recently created boundary=forest, which seems to cover your case:
A unique, whole area with verifiable boundaries and predominantly wooded, but fragmented in distinct, separate areas: If the fragmented area is still considered in its entirety as a unique delimited forest, then it is to be modelled with a unique boundary=forest relation, which would enclose all the separate pieces of wooded land; use natural=wood/landuse=forest for the physical wooded land.
Thus, I would model the inholdings as separate areas (closed ways or multipolygons as you prefer), each tagged with landuse=forest but distinct operator and/or owner tags – I don’t see a need to further complicate things with boundary=legal.
Now, I fail to see the bright line distinguishing boundary=forest from boundary=protected_area (unless the forest in question falls short from the definition of a “protected area”). Use whichever is more appropriate. I suppose that boundary=forest, being a recently voted proposal, does not render everywhere…
As noted in more detail on the previous (regected) proposal, the US National Forest boundaries themselves are generally not materialized with cut-lines/blazes. The “timber stand improvement areas” are the closer equivalent for boundary=forest.
The outer boundary of the US National Forests is the “proclamation boundary” where congress has authorized land acquisition, whereas the forest-service-owned lands are where public use is allowed/controlled by the Forest Service.
However, doesn’t that [mapping both] violate “one feature, one OSM
element” ? I
believe we should stick with the inholding method, because separating
national forests into different relations complicates search features,
rendering, etc.
For most users, the proclamation boundary would be pretty useless if
ownership is already there. As Kevin noted, the proclamation boundary shows
an area that the government has been authorized to acquire land, and has
little impact on actual protection and land cover.
I believe the result of that and other discussions (such as in the #protected-lands channel of the OSM-US Slack) wrapped up with the consensus that only the ownership boundaries would be mapped for National Forests and BLM lands.
Regarding the mapping of the inholdings themselves, if they were organized administratively, then you could use some sort of administrative tagging on a boundary relation. I would caution that the community generally opposes parcel mapping, at least for non-protected things such as residential property.
Regarding Adam‘s comment, I fully agree that it is only prudent to map USFS boundaries based on ownership of the land. I would add one legal caveat, however, that the ownership could be in fee-simple, or could be in the form of a conservation easement.
