Proposal:Boundary=forest( compartment) relations (v3)

From OpenStreetMap Wiki
Jump to navigation Jump to search
boundary=forest(_compartment) relations (v3)
Proposal status: Approved (active)
Proposed by: Penegal
Tagging: boundary=forest
Applies to: area relation
Definition: Tagging scheme for forest boundaries (managed or unmanaged forests)

Draft started: 2021-11-17
RFC start: 2022-01-24
Vote start: 2022-02-13
Vote end: 2022-02-27


Overhead imagery of a forest compartment
A forest compartment within a forest.

A delimited forest is a land which is predominantly wooded and which is, for this reason, given defined boundaries. The lands inside the boundaries are often subject to different rules than the surrounding lands. The delimited area is typically named, and have materialized, visible, typically cleared, boundaries. It may cover different tree stands, non-wooded areas, highways… but all the area within the boundaries are considered and managed as a single forest.

A forest compartment is a numbered sub-division within a delimited forest, physically materialized with visible, typically cleared, boundaries.

Proposed tagging changes

Summary of Tagging Changes
Tag Action Description
boundary=forest Approve Describes the boundary of a delimited forest. The delimited forest may include multiple disjoint parts and may encompass non-wooded areas such as glades, screes, or ponds as long as the collective area is administered (possibly also managed) as a whole. This tag is combined at a minimum with name=*.
boundary=forest_compartment Approve A delimited forest may be divided into forest compartments. A forest compartment is a referenced area wholly within a boundary=forest, which has visible boundaries, typically evidenced by the removal of trees along the boundary line or the presence of physical markers. The tag ref=* is used to indicate the compartment's unique identifier in the delimited forest, as it is displayed on the ground.
boundary=forestry_compartment Deprecate This tag is de facto synonym of boundary=forest_compartment. The term "forestry" means "the science or practice of planting, managing, and caring for forests", but compartments are not necessarily managed through forestry; there are many reasons for not managing the tree stand of a compartment, and yet maintaining the compartmentalization for the purpose of managing the whole delimited forest.
No Change These tags continue to be used to tag physical, tree-covered areas, regardless of whether they are located within a delimited forest or not. Resolving the dispute between theses tags is out of the scope of the current proposal, which is only about a tagging scheme for consistent modelling of delimited, heterogeneous areas which are considered constituting a single forest.


What is usually called forest and forest compartment encompass wooded areas, but not only: it may encompass screes, ponds, glades… It also typically encompass various tree stands: needleleaved trees, deciduous trees, plantations… All these features are currently mapped with their own OSM feature, but there is no way to accurately map their belonging to a single forest.

Identifying delimited forests

Though delimited forests are mainly wooded, they are not totally so. Delimited forests may contain screes, ponds, scrublands, or grasslands which are not covered by trees, and are consequently not tagged as wooded areas. Such non-wooded areas are still part of the delimited forest, as they may be:

  • managed by the manager of the delimited forest;
  • under reforestation;
  • lands with a distinct, wood-related ecosystem (glades for instance, which essentially exist because they are in a wooded area, else they would be called grasslands);
  • legally considered part of the surrounding forest, i.e. they are subject to the same subset of laws, rights and obligations ( Régime forestier in France, for instance).

Some forests may even be predominantly scrubland (natural=scrub), which often forms after a clearcut followed by natural regeneration (i.e. no plantation). In this case, the land will not be again tree covered until the seedlings grow, which can take decades.

Here are some examples of delimitation of forests and forests compartments in France:


A wild wooded area is considered a delimited forest when its limits are clearly materialized, for instance with paint, boundary markers, or cutlines.

The markings used to delimit forests are more significant than the standard markings used for common land lot boundaries. Standard land ownership boundaries in wooded areas typically lack signs, marks or paint highlighting the difference between wooded area boundaries and non-wooded area boundaries, but local laws and customs may prescribe carved marks or other methods for the materialization of delimited forest boundaries. If a wooded area is only delimited with land lot boundary stones, without any signs, marks or paint highlighting the difference between its boundaries and standard land lot boundaries, the area should not be modelled as a delimited forest.


A forest compartment designates a single piece of a given delimited forest, which is identified by a reference, often a number, which is unique inside the delimited forest. Forest compartments are not a land lot nor a cadastral parcel. Though boundaries of forests and their compartments on one hand, and land lots on the other hand, frequently overlap, a given delimited forest or forest compartment may be composed of multiple land lots, and a single land lot may be divided in multiple delimited forests or forest compartments. Each compartment should be modeled as a single boundary=forest_compartment entity, without consideration for any land lot boundaries which may encompass or be contained within the compartment.

Forest compartments should be mapped if present, but will not be present in all delimited forests. Some delimited forests are too small to justify the creation and maintenance of compartments, or may be managed without compartments.

In some countries, such as Poland, forest compartments may be defined in public GIS but not materialized on the ground. Since these "compartments" do not have verifiable and distinguishable existence and limits on the ground, they are effectively cadastral information and should not be mapped.

Forestry activities

This proposal is not about forestry activities: delimited forests may be the place of forestry activities, but this is not necessarily the case. Materialized boundaries may be mandated by law, current practices or necessity, but do not necessarily imply that the encompassed area is subject to forestry works.

Forests vs protected areas

Delimited forests and protected areas may overlap or coincide. If they overlap, they can be modelled separately, with a boundary=protected_area entity and a boundary=forest entity.

In some cases, though, a delimited forest may also be a protected area (most likely, IUCN Category VI): for instance, in France, the Réserve Biologique Dirigée (RBD) and Réserve Biologique Intégrale (RBI) can legally only encompass public forests, which are always delimited forests. In such a case, the facts that the area is, on one hand, a forest, and, on the other hand, a protected area, cannot be separated.

In the case that a whole delimited forest is also a protected area and these two characteristics apply to the same entity (a RBD/RBI, for instance), they must be mapped with a unique OSM entity, which would be tagged with boundary=forest + protect_class=*, consistently with the tagging scheme for IUCN Category II national parks (tagged boundary=national_park + protect_class=2). It is appropriate to also tag this entity with other relevant tags commonly used with protected areas, such as related_law=*, access tagging, or leisure=nature_reserve if applicable, to reflect the dual status of the area.

If, on the other hand, the protected area and the delimited forest exactly encompass the same area but are distinct features (that is to say, the legal protection status of the area is not attached to the area being also a delimited forest), they should be mapped with separate entities: a boundary=forest entity for the forest, and a boundary=protected_area for the protected area.

Areas named "Forest"

The fact that an area has the word "forest" in the name does not necessarily mean that the area should be mapped with the proposed tagging scheme: the wooded state of the land must be a predominant characteristic, and there must be defined boundaries. For instance, US National Forests have mostly unmaterialized boundaries, and are consequently not in the scope of the proposed tagging scheme.

Named natural forests, such as Germany's  Black Forest, are indeed large wooded areas which are considered as a unique forest, but they lack clear boundaries, are not effectively managed as a whole and, as such, do not qualify for the proposed tagging scheme.


A delimited forest should be mapped as follows:

  • The overall delimited forest should be mapped as an enclosing polygon or relation. Do not add land cover tagging (such as natural=*) to this boundary or "cut out" minor non-wooded areas such as ponds or screes that may be present.
  • A delimited forest may include smaller delimited forests, themselves either divided or not in compartments, so a boundary=forest entity may enclose smaller boundary=forest entities.
  • Natural features (such as natural=wood, natural=water, natural=scree, etc.) should be separately tagged with their own polygons delimiting actual areas where these features are present, and as an inner multipolygon member of the enclosing natural=wood/landuse=forest features if they are not wooded. Woodland-specific tagging such as leaf_type=*, leaf_cycle=*, etc., should be tagged on natural=wood/landuse=forest features.
Example modelling, with two different wooded areas (tagged natural=wood) and the overall delimited forest entity (tagged boundary=forest); the pond is considered part of the delimited forest and therefore included in it.

The overall delimited forest (entity tagged boundary=forest) can be tagged as follows:

Suggested tagging for boundary=forest
Key Value Discussion
type=* boundary Mandatory
boundary=* forest Mandatory
name=* The forest name Recommended. If applicable, the delimited forest can be named with this tag, and, maybe, its subtags, like name=Forêt de Fontainebleau or name:de=Wald von Fontainebleau.
ref=* The forest ref Optional. Such reference is probably country-based, so one should use a subtag, like ref:FR:ONF, whenever applicable; the choice of this subtag for a given country is up to the local OSM community.
operator=* The forest operator Optional.
owner=* The forest owner Optional.
operator:wikidata=* Wikidata ID Optional. Use for giving the wikidata ID of the operator; for instance, a French forêt communale, mandatorily managed by the national Office National des Forêts, would be tagged operator:wikidata=Q3260938.
owner:wikidata=* Wikidata ID Optional. Use for giving the wikidata ID of the owner; for instance, the French forêt communale d’Épinal, owned by the commune d’Épinal would be tagged owner:wikidata=Q173695.
source=* The source used Optional.
protect_class=* Protection class ID Optional. Use if the delimited forest has also been designated a protected area as defined by IUCN protected area categories, and if the protected area and the forest form a single entity.
protection_title=* The forest owner Optional. Use if the delimited forest has also been designated a protected area as defined by IUCN protected area categories, and if the protected area and the forest form a single entity.
related_law=* The related law Optional. Use for giving the title of the edict or legal text which establishes or regulates the delimited forest.
leisure=* nature_reserve Optional. Use if the area is also a nature reserve; if there are access restrictions, don't forget to accordingly add access keys on the highways allowing access to the area.

Forest compartments

The tag boundary=forest_compartment is used to describe a physically materialized forest compartment. There must be a 1:1 correspondence, that is to say one boundary=forest_compartment entity by compartment, and one compartment by boundary=forest_compartment entity. Consequently, boundary=forest_compartment relations should never overlap. Note that a delimited forest is not always divided into compartments.

The link between a delimited forest and its compartments is spatial, as for all boundary entities: a compartment is considered part of a delimited forest if and only if its boundary=forest_compartment entity is spatially fully encompassed by the boundary=forest entity. Do not include boundary=forest_compartment entities as Role subarea members of the corresponding boundary=forest relation; such modelling is disputed and widely considered redundant.

It may happen that the compartment materialization signs only show the compartment ref, and don't allow to know to which forest the compartment belongs. It is also possible that delimited named forests don't exist in a region and, therefore, that boundary=forest is not applicable. It is thus accepted to map the boundary=forest_compartment entities without mapping the boundary=forest entity to which they belong. The enclosing boundary=forest entity should be mapped if this information is available.

A boundary=forest_compartment entity should have the following tags:

Suggested tagging for boundary=forestry_compartment
Key Value Discussion
type boundary Mandatory
boundary forest_compartment Mandatory
ref The compartment ref displayed on the ground Recommended. According to the Map what's on the ground principle, this tag's value must be the one displayed on the ground. For a compartment, this must be the unique ref of the compartment within the encompassing delimited forest, not a ref which is valid in, or applies to, a broader area.

It may happen that official databases record, or local signs display, this local ref along with additional data (the ref of the surrounding forest, an owner ref…); if so, one should remove these additional details, which are not related to the compartment itself, and only keep the compartment ref inside the delimited forest, as displayed.

name A name Optional. If and only if applicable, the compartment can be named with this tag. Don't confuse name and ref! A name is a word or group of words with an intrinsic meaning, whereas a ref is a sequence of numbers/letters whose meaning exists only in the context of a larger delimited forest. If you want to use this tag, think twice: it is likely you're trying to tag a ref, not a name, as typical compartments have only a reference, not a name.

As a delimited forest is managed as a whole, it is not useful to tag each compartment belonging to a delimited forest with tags applying to the whole delimited forest; such tags should be applied on the enclosing boundary=forest entity. For instance, the operator of the delimited forest is most probably unique, so an operator tag should not be applied on each compartment, only on the boundary=forest entity.

Transition of existing tagging

The following table explains how existing tagging can be replaced with the new tagging scheme. Note that automated edits should be conducted only in consultation with the community. See the Automated Edits code of conduct for more details.

Current tagging New tagging
  • Re-tag as boundary=forest_compartment
  • If a ref is given, check that the value is really the compartment ref inside the delimited forest, as it is displayed on the ground.
natural=wood or landuse=forest, when the entity represents a delimited forest
natural=wood or landuse=forest, when the entity represents a forest compartment
landuse=forest, where the area is neither a forest compartment nor a delimited forest landuse=forest (no change)
natural=wood, where the area is neither a forest compartment nor a delimited forest natural=wood (no change)


French forêt communale

The forêt communale de Chauffecourt would be modelled by an entity (area or relation) tagged this way:

Compartment #6 of this forêt communale would be modelled as an entity tagged this way:

German state forest

The Einsiedler Wald relation would be tagged this way:

Compartment #159 of this forest would be modelled as an entity tagged this way:

Is this a delimited forest?

Here are some special cases where the applicability of the proposed boundary=forest relation may need clarification.

Case Applicable? Why?
Illegal agricultural deforestation area No It most likely lacks precise boundaries, which are a prerequisite to be modelled as boundary=forest. In addition, once the deforestation is started, the area likely does not keep its wooded state for long and is thus out of scope for OSM.
Legal forestry concession with illegal logging Yes Even if the logging is illegal, the area is wooded, and the concession should have verifiable boundaries. Of course, if the area becomes totally, permanently deforested, without letting natural regeneration take place or replanting, the area is no longer a forest.
Wooded land without verifiable boundaries No If the area limits are not verifiable, it should not be modelled as a boundary=forest relation.
The  Black Forest No Even if it is a predominantly wooded land, which is commonly known as a whole, single wooded area, its boundaries are neither distinct nor verifiable, hence disputable, so it is not to be modelled with a boundary=forest relation. That being said, the underlying physical, wooded lands themselves are still eligible to modelling with natural=wood/landuse=forest.
Mostly wooded area without verifiable boundaries No The area is predominantly wooded, but its boundaries are unverifiable.
A unique, whole area with verifiable boundaries and predominantly wooded, but fragmented in distinct, separate areas Yes 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.
A wooded area with distinct existence and verifiable boundaries, which contain screes and glades Yes The boundary's verifiability and the area being predominantly wooded qualifies it for a boundary=forest relation. As this relation is about delimited forest boundaries, if the screes and glades are considered part and parcel of the delimited forest, the relation will not treat them as exclaves of the said delimited forest. These non-wooded areas are to be mapped by their own entities, and the wooded land around with natural=wood/landuse=forest entities; once again, do not overlap natural=wood/landuse=forest entities and non-wooded entities.
A wooded land lot whose boundaries are mere boundary stones No The materialization of boundaries in a way which is distinct of standard land lots is required.
A wooded land lot whose boundaries are marked in a way distinct of adjacent land lots Yes The materialization of boundaries in a way which is distinct of standard land lots is enough, unless you are sure the specific marks used are unrelated to forests.
A US National Forest Likely, no Typically, US National Forests have no materialized boundaries, so they are not in the proposal scope.

Taginfo comparison

Approved Tag Deprecated Tag

Applies to

This tagging scheme applies to relation boundary relations, but you may use an area area if the entity's geometry is simple enough.


For renderers which wish to render boundary=forest and/or boundary=forest_compartment, it is suggested to render the perimeter of the area, and to not use a fill.

Features/Pages affected

External discussions


Please comment on the discussion page.



Voting closed

Voting on this proposal has been closed.

It was approved with 18 votes for, 0 votes against and 0 abstentions.

I, Penegal, will take care of creating/editing the necessary Wiki pages in the following weeks.

  • I approve this proposal I approve this proposal. I support most of what is proposed. This doesn't prevent us to look for solutions for points I raise in Talk Fanfouer (talk) 16:06, 13 February 2022 (UTC)
  • I approve this proposal I approve this proposal. I can understand why boundary=forestry is not preferred for the same reason as boundary=forestry_compartment. --- Kovposch (talk) 16:15, 13 February 2022 (UTC)
  • I approve this proposal I approve this proposal. --Penegal (talk) 16:41, 13 February 2022 (UTC)
  • I approve this proposal I approve this proposal. --JeroenHoek (talk) 19:32, 13 February 2022 (UTC)
  • I approve this proposal I approve this proposal. A tip of my hat to the authors of / contributors to this proposal: hard work and patience is quite evident here. -- Stevea (talk) 19:49, 13 February 2022 (UTC)
  • I approve this proposal I approve this proposal. --Peter Elderson (talk) 21:15, 13 February 2022 (UTC)
  • I approve this proposal I approve this proposal. I echo Stevea's sentiments---lots of hard work here. --Dr Centerline (talk) 22:13, 13 February 2022 (UTC)
  • I approve this proposal I approve this proposal. this is probably quite useful for forests which are administered, legally defined etc., while it still leaves gaps for forests that are neither legally defined nor consisting of only trees or forest land, but are culturally distinguished (have a common forest name as an ensemble). But it seems capable to solve the administrative part. --Dieterdreist (talk) 10:33, 14 February 2022 (UTC)
  • I approve this proposal I approve this proposal. I do not fully agree with the strict guidelines of materialisation and verifiability. I am afraid this will prevent usage of the scheme in regions where on the ground marking is poorly implemented and maintained, vandalism or encroaching is more common. However this issue also applies to other boundary tags, could be addressed with additional sub-tags. I hope the approved proposal page will be less "strict". Nevertheless, this tag fills a gap, took a lot of effort and consolidation and deserves support as a base from where we can build. Being politely persistent should be awarded.--Bert Araali (talk) 14:12, 14 February 2022 (UTC)
  • I approve this proposal I approve this proposal. --Wetterauer (talk) 17:15, 14 February 2022 (UTC)
  • I approve this proposal I approve this proposal. --Adavidson (talk) 23:32, 16 February 2022 (UTC)
  • I approve this proposal I approve this proposal. --EneaSuper (talk) 11:22, 20 February 2022 (UTC)
  • I approve this proposal I approve this proposal. High time a boundary=forest tag is standardized to discourage using landuse=forest for it. --JB (talk) 12:27, 20 February 2022 (UTC)
  • I approve this proposal I approve this proposal. --Lkw (talk) 14:50, 20 February 2022 (UTC)
  • I approve this proposal I approve this proposal. --501ghost (talk) 00:57, 21 February 2022 (UTC)
  • I approve this proposal I approve this proposal. I've been following this proposal since its first version and it keeps getting better and better. Thank you Penegal for sticking with it and helping us get to a clear and consistent scheme for tagging these important and useful boundaries. --Adamfranco (talk) 01:16, 21 February 2022 (UTC)
  • I approve this proposal I approve this proposal. For me, I can hardly see any possibilities for capturing such things on the ground. In part, however, it offers elegant possibilities for recording certain contiguous forest areas. In addition, a good basis for Polish forest divisions is now certainly laid here: for example: and surroundings. --streckenkundler (talk) 20:40, 21 February 2022 (UTC)
  • I approve this proposal I approve this proposal. --ZeLonewolf (talk) 01:48, 24 February 2022 (UTC)