[Tagging] Feature Proposal - RFC - boundary=forest(_compartment) relations

[David Marchal]

Sent with [ProtonMail](https://protonmail.com) Secure Email.

‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
Le mercredi, 10. février 2021 12:27, Mateusz Konieczny via Tagging <tagging at openstreetmap.org> a écrit :

> It would be worth explicit documenting that forestry includes not only logging
> Active forestry may be happening in area without logging, without plans for
> commercial exploitation of forest (or use of forest for purposes other than
> production of timber and related materials).

I took care of that in the proposal; you can find details here: https://wiki.openstreetmap.org/wiki/Proposed_features/boundary%3Dforestry(_compartment)_relations#Forestry_area

> What about privately owned forests?

The proposal mainly talk about public forests because they are the ones I found the most documentation about. That being said, the proposal also applies to privately owned forests; nothing prevents that, nor motivate preventing it.

> Long time ago someone imported areas officially considered as forest parcels in Poland,
> result was often a bizarre mess (barcode pattern of narrow strips of land was very popular
> - 100m long, 5m wide plot owned by National Forestry, 5m wide plot owned by
> private owner, 5m wide plot owned by National Forestry, repeated 50 times).
>
> Especially if in some areas boundary=forestry would be worth rendering,
> how it should be handled that in some areas it would merely record land ownership
> and nothing else? And rendering it would be useless and harmful there?

In the case you gave, the user considered land lots as forest compartments; these are two different notions, and the proposal clearly states that they are not to be confused. If the users incorrectly mapped the area, applying the new tagging scheme to it would of course include correcting the existing modelling.

> Also, how it is supposed to be mapped/verified?
> If one sees 45 years old trees - how one can recognise is it maintained
> commercial forest, nowadays protected area, abandoned private forest without
> forestry, private forest without forestry because it is not needed right now
> or something else?

This is why the proposal renders precise limits mandatory for a forestry area (https://wiki.openstreetmap.org/wiki/Proposed_features/boundary%3Dforestry(_compartment)_relations#Boundaries): they must be either materialized in a different way than plain land lot boundaries (as explained in the proposal, this is routinely done for practical reasons), or available from an open forestry data record. In addition, there must be forestry works in it; there may be reasons for not doing forestry works for decades, but the limits should still match in one of the two criteria described. If there are no distinguishable nor open source limits, and no forestry works, then it is not to be mapped as a forestry area. More detailed reasoning is available in the proposal page and its discussion page.

> Would it be OK to map this also when this boundaries has no traces at all on the ground
> and importing them is the sole source of obtaining this data, with independent
> verification being impossible?

Yes, it would be OK if the data source is somewhat official or credible enough: it is OK if it is a public forest manager database, like the US Forest Service or the French Office national des forêts, or even an established private operator as a concession holder, but if it is the website of your local bakery, it is not. The credibility of specific data sources should be decided by local OSM communities, which are the most relevant decision makers about local operators.

>> Let's take the above schema, with ║ symbolising the boundary=forestry relation, and ─ symbolising natural=wood, the only remaining tag for wooded areas:
>
> What about landcover=trees? Would it be also deprecated?

Hmm. I didn't thought about it, but it seems it would also be deprecated, because saying "natural=wood now simply say the area is covered by trees" would, IMO, simply void landcover=trees. But perhaps you have a different reasoning?

>> I'll try to explain through ASCII art (to be quick; should what I explain be merged in the proposal, I would take time to draw nice schemas). Let's say there is a pond (│ symbol) in the forestry area, modelled with landuse=forest (─ symbol), and the pond is considered part and parcel of the forestry area. Currently, the landuse=forest and water=pond polygons would overlap:
>> ─────────────────────
>> ─────────────────────
>> ───────┼┼┼┼┼┼────────
>> ─────┼┼┼┼┼┼┼┼┼───────
>> ───────┼┼┼┼┼┼┼┼┼─────
>> ───────┼┼┼┼┼┼┼┼──────
>> ─────────────────────
>> This mapping allows to model the fact that the pond is considered part and parcel of the forestry area.
>
> This is an incorrect tagging caused by using landuse=forest to map forestry area

Indeed, it is currently the case, but I was using an hypothesis here: if the pond is considered part and parcel of the forestry area, and if landuse=forest is used to model this forestry area, then the tagging isn't incorrect, just counterintuitive. That is not good mapping for me either, but I was just describing what could be done, and why it should be considered wrong.

>> Unfortunately, this often leads to a somewhat buggy rendering, with trees symbols rendered in the water.
>
> This is deliberate to show that tagging is buggy.
>
> See
> https://github.com/gravitystorm/openstreetmap-carto/pull/1242
> https://github.com/gravitystorm/openstreetmap-carto/pull/1728

OK, I take note of that.

>> The proposal could be about "landuse=forest is managed, natural=wood is unmanaged. Period.", but, as I understand the situation, such proposal would likely stall or being uneffective, because most mappers who use another approach of the landuse=forest/natural=wood distinction would likely keep their approach, voiding the proposal, even approved. That wouldn't resolve the pointed rendering issues either.
>
> Alsowhat about
> - people mapping forest from aerial images
> - not interested in this distinction
> - unable to survey situation
> - ones overwhelmed by complexity
> ?

Then they would simply use natural=wood and leave the forestry area mapping to users who can manage it. This way, they would simply add the wooded area, which is still an improvement of OSM data; if they are interested, they can care about boundary=forestry, but it is not required for mapping the wooded area itself, and they can simply not bother about it.

> (also note that it is extremely hard to fing forest that is not managed - typical protected forest
> is also managed in some way)
>
> There must be a way to mark "I mark that there are trees here" without making
> mandatory to specify anything else.

Yes, there is a way to do it: in the currently discussed amendment of the proposal, one could simply use natural=wood. If you know the area is managed, use boundary=forestry in addition; if you don't know or don't care, don't use it.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/tagging/attachments/20210210/5761d1aa/attachment.htm>