NOTICE: is no longer in use from 1st March 2024. Please use the OpenStreetMap Community Forum


Should landuse=forest be reserved only for where the woodland is actively harvested - if so, how can this be determined from the ground (especially with deciduous woodland, where there may be many years between harvests); What about copses?

asked 19 Jul '10, 13:16

Rowland's gravatar image

accept rate: 0%

12next »

landuse=forest is user where a forest has been planted by humans while natural=wood is used where the forest is planted naturally. landuse=forest is often more uniform in type, age and size of trees while natural=wood are often a combination of different type of trees of different size and age.

This is because in an actively harvested forest (even where harvest can be generations apart) the trees are all planted at the same time and of the same type while in natural forests the trees are all planted at different times and by different types of seeds.

Note that natural forests may have soil conditions that keeps some types of trees from growing allowing for less variation in types of trees making them difficult to distinguish from a planted forest.

permanent link

answered 19 Jul '10, 14:19

Gnonthgol's gravatar image

Gnonthgol ♦
accept rate: 16%


My point is that it is virtually impossible to tell the difference on the ground; which was why I was seeking clarification. All the woodlands I know of are all managed to a certain extent, but none follow the description of trees being laid in a uniform type, and all include a mixture of species and ages.

(20 Jul '10, 08:59) Rowland

As far Is I know, it is usual in France to let the harvested forest grow again, by itself, very young trees being not harvested. From time to time, some young trees are destroyed, as to give space to others. This is usefull with slow growing trees, such as oaks (barrels for wine maturing are made of oak).

(20 Apr '12, 18:28) HelenePETIT

The distinction often used between landuse=forest for a managed forest and natural=wood for a natural wood is misleading. There is not much un-managed forest on this planet, certainly not in Europe or other places that have sizeable population. Even though a forest might look natural to you, it probably is not. The clear-cut forest farming that is being done in some places of the world is not the only way to manage forests. In many places trees are selectively cut and removed and the forest might look somewhat natural, but it is not.

So currently there is no good way to distinguish these. Maybe somebody can come up with a better tagging scheme.

permanent link

answered 19 Jul '10, 14:38

Jochen%20Topf's gravatar image

Jochen Topf
accept rate: 31%

edited 21 Jul '10, 12:12

Andy%20Allan's gravatar image

Andy Allan


That distinction was what I was hoping to find - for the moment, I'l carry on using natural=wood unless I know that the woodland is harvested each season (the likes of Thetford Forest and Rendlesham Forest). Also flag with an operator=... if I know a woodland is actively managed by an organisation such as the Forestry Commission

(20 Jul '10, 09:03) Rowland

As a newbie I just discovered this natural=wood against landuse=forest. I now fully understand the difference. Hoverer, I have been using landuse=forest for a while, just because I use Potlach2 and under the "Natural" tab I only see "Forest" which produces landuse=forest.

(06 May '12, 18:16) gerdami

My personal solution is to add ''landcover=trees'' to all tree-covered areas. This way you can map what you see on the ground without having to guess or research unmanaged vs. managed. Also this distinction doesn't fit into the tagging scheme (and wasn't even there at first), where "natural" is about distinct features, while landuse is about the use of the land (both keys are orthogonal and can be applied at the same time).

(20 Jun '14, 11:07) dieterdreist

It is pretty difficult to pick between them at times.

I generally take the view that landuse=forest identifies a planted crop of trees, typically all the same species, which will be harvested at some future date.

By contrast, a natural=wood may be managed (e.g. it may have paths cleared, fences erected, signposts, and may have trees planted), but it is not a crop.

permanent link

answered 23 Jul '10, 13:21

GrahamS's gravatar image

accept rate: 28%


That's the definition I go by, as it's the only one that can be done fairly reliably. Replace "landuse=forest" by "landuse=forestry" in your mind and it's easy.

The corrolary is that forestry areas that have been cut and will take years to grow back can additionaly be tagged natural=scrub.

(05 Dec '13, 11:50) Vincent de P... ♦

Copses can be unplanted - and in the UK frequently are (or at least so far as recorded history goes), but are not managed by the clear-fell and replant system characteristic of forestry. They can be classed as 'ancient woodland' for planning purposes. Therefore, natural=wood would seem appropriate. Copses are managed and cropped, however (or at least were before they fell out of management).

If landuse=forest is an actively managed forestry operation involving clear-felling and replanting, then natural=wood is quite wide. Of course, ecologically, natural=scrub turns into natural=wood, and, if not actively managed as heath, natural=heath will turn into natural=scrub then natural=wood. Equally, if landuse=forest is not actively managed, it will naturally turn into natural=wood.

permanent link

answered 16 Jul '11, 17:14

paulbiv's gravatar image

accept rate: 0%


IMHO the definition on Map Features is very clear on this issue:-

Natural woodland (trees). Only for completely unmanaged/wild ares. For forests that are owned or managed by someone, use landuse=forest instead.

I am not aware of any woods in the UK that are not owned or managed by someone and as such should be tagged landuse=forest

permanent link

answered 17 Jul '11, 13:01

gumpa's gravatar image

accept rate: 0%

This whole thing is really difficult.

One point of view:

All woodlands are landuse=forest across the whole world because even subsistence tribal communities will use and manage woodland for fire wood and materials etc. Even highly inaccessible forests can be seen as being managed at carbon sinks and balancing CO2 emission in developing countries. i.e. all natural= features are managed and therefore should be landuse= features. Also 'pristine' woodland can be cut down at the whim of loggers/governments so have been managed to be non-managed.

Second point of view:

All landuse=forest should be natural=wood because all woodland contains 'natural' wood - even if it has been planted. And may also include other 'natural' plants and animals that have moved in.

Personally, I use natural=wood on 'amenity woodlands' and small woodlands that have low levels of management. Where there are wide areas of monoculture woodland with forest access tracks and signs of planting in rows etc I use landuse=forest.

I'm leaning towards tagging the area of highly managed forest with landuse=forest and the individual packets of woodlands as natural=wood. In the same way I would tag an industrial estate with landuse=industrial but tag the buildings, waterways and areas of grass etc within an industrial estate separately.

In many ways, don't worry too much. The tagging can be rendered in different ways and people will make sense of what you have done one way or another.

Other tags may be more useful, and I will try to work on getting consensus e.g. type=broad-leaved

or wood=decideous but there is real problems at the moment with lots of different ways to use these. Also species= or taxon= or perhaps other ways to record species will also help define woodlands better. operator=forestry_commission etc would also help understand the context of the type of woodland.

Well, I you can't see the woods for trees on this - I don't blame you...

I'm organising a woodland mapping party in March - so hopefully I'll have a better answer by then:

permanent link

answered 04 Jan '12, 22:13

Hawkeye's gravatar image

accept rate: 0%

I think partly the problem is that the map renders landuse=forest with a coniferous/pine tree icon.

For people reading the map, they would expect to see a pine wood not and semi-natural broad-leaved wood used for collecting small amounts of timber. Also this is rather temperate zone centric. A rubber tree plantation is probably best not represented with by a pine tree icon.

The plain green rendering of natural=wood is more associated with broad-leaved amenity planting/semi-natural woods.

Also, I think natural=wood was in the potlatch editor for a while but now the only 'simple' option is landuse=forest . This will probably increase the amount of landuse=forest being added.

Not that I support tagging for rendering. But its just an observation. I'll add my observations to the relevant discussion pages.

I am coming around to the idea of only using landuse=forest.

permanent link

answered 05 Jan '12, 12:29

Hawkeye's gravatar image

accept rate: 0%

I've ask for an ITO maps displaying different uses of these tags. Hopefully this will be useful:

permanent link

answered 05 Jan '12, 15:49

Hawkeye's gravatar image

accept rate: 0%

The problem with the Map features definition is that it doesn't match at all well to botanical usage in the English language. It leads to classification of things as forests that have never been called such since the days when forest meant a royal hunting preserve with no woodland implication.

See for example "Trees and Woodland in the British Landscape" by Oliver Rackham. Rackham, Oliver (1976). Trees and Woodland in the British Landscape. Archaeology in the Field Series (First ed.). London: J.M. Dent & Sons Ltd. ISBN 0-460-04183-5. Rackham, Oliver (1990). Trees and Woodland in the British Landscape. Archaeology in the Field Series (Revised ed.). London: J.M. Dent & Sons Ltd. ISBN 0-460-04786-8. Rackham, Oliver (2001). Trees and woodland in the British landscape: the complete history of Britain's trees, woods & hedgerows. London: Phoenix Press. ISBN 1-84212-469-2.

It would be a respectable historical view that there are NO completely wild/unmanaged woodlands on the planet, if by that you mean NEVER having been managed (see discussions on Biochar and the Amazon Rain Forest for example).

If all woodland is therefore secondary, then it comes down to whether currently managed, and the nature of the products.

The Rackham distinction is roughly that forest is managed for timber, as in tree-sized products, and may involve clear-felling and replanting. Wood is managed for poles (coppice poles and firewood) or unmanaged because management has stopped.

In the UK context that basically means that forest is conifer and wood is broadleaved, but this will vary between countries

permanent link

answered 08 Jan '12, 18:05

paulbiv's gravatar image

accept rate: 0%

One distinction occurs in the US. There are the national forests, which are managed woodlands (landuse=forest), though they often are not replanted by hand. In some places trees are replaced by natural regrowth. Also there is now less clear-cutting. Then there are the national and state wilderness areas, where no trees are cut commercially. They might fit natural=wood. Also someone mentioned natural=scrub as turning into a wood eventually. Well, not if the scrub is in a dry place, such as a desert, as it is many places in the world. It is always scrub.


permanent link

answered 22 Apr '12, 19:05

techlady's gravatar image

accept rate: 0%

Follow this question

By Email:

Once you sign in you will be able to subscribe for any updates here



Answers and Comments

Markdown Basics

  • *italic* or _italic_
  • **bold** or __bold__
  • link:[text]( "title")
  • image?![alt text](/path/img.jpg "title")
  • numbered list: 1. Foo 2. Bar
  • to add a line break simply add two spaces to where you would like the new line to be.
  • basic HTML tags are also supported

Question tags:


question asked: 19 Jul '10, 13:16

question was seen: 31,471 times

last updated: 20 Jun '14, 11:07

NOTICE: is no longer in use from 1st March 2024. Please use the OpenStreetMap Community Forum