From OpenStreetMap Wiki
Jump to navigation Jump to search

Limited/incomplete example values?

The current examples of coniferous/deciduous/mixed seem very limited, and also pretty incomplete. There are coniferous trees that drop their leaves (needles) every year, and there are broadleaved trees that keep their leaves during winter, only to shed them during the spring.

  • coniferous/broadleaved describes the type of leaves
  • deciduous/evergreen describes the behaviour to shed leaves every year or not

The distinction between coniferous and broadleaved trees is not that hard to make while mapping, but the distinction between deciduous and evergreen is not so clear cut, unless you know your trees. --Ldp 16:25, 22 November 2009 (UTC)

Wood density

The examples given for wood=* should probably also suggest ways to tag open vs. dense woods. In my opinion, wood=coniferous, does not provide a good description of the vegetation in a wood, as pine woods and spruce woods are quite different when it comes to the soil and freedom of movement, or "runnability" in orienteering. The nature/vegetation type may be a more relevant way to describe the wood.

Ways to tag logging areas should also be discussed. Also whether the lumbering area have been replanted.

Expressed as tags:

  • wood="coastal pine wood",
  • wood=logging (with start_date=2010 for the year the lumbering took place)

vibrog 19:39, 4 April 2010 (UTC)

More values

To comply with INT 1, we need more values for wood, see point C31.#, all of these values are rendered differently on nautical maps. Suggested values: wood=evergreen, wood=palm, wood=napa_palm, wood=casuarina, wood=filao and wood=eucalypt --Skippern 03:13, 26 August 2010 (BST)

Tried to rationalise values

As woods can be both evergreen and broad-leaved or evergreen and coniferous - I have tried to separate these with wood= and type= tags

wood= broad-leaved, coniferous, mixed, palm (this can be established easily on first survey - even from aerial photos). type= evergreen/deciduous (needs good identification skills or more than one survey).

Ultimately, species data would be best. But I haven't got my head around the best way to do this...

Your changes to the wood tag are unacceptable without prior discussion, starting with this one. The three values coniferous, deciduous, mixed have been used for years in the database, have made their way into editor presets, and are used by map styles.
I'm sorry that I have to revert your edits and I don't want to judge the merits of your suggested tagging, but the wood key is too established to be changed unilaterally. --Tordanik 01:47, 5 January 2012 (UTC)
The existing tags would still remain in the database and people could still use them, but new data entered would make more sense. This has been discussed from 2009 in the first comment. We are talking about a small number of tags which people and renders could quickly incorporate. None of these tags have been voted on, so would it be best to write a new proposal and vote on them? I have map styles using these tags, but I don't mind changing them to incorporate both old and new ways of tagging. I just feel we can do better than the current tags.
Sorry Tordaniak, it was foolish of me to change it so rashly. --Hawkeyes 12:59 5 January 2012
Traditionally mappers have discouraged using type as the key for anything. Would foliage=decidous/evergreen/* be fitting? Thinking about it very briefly I couldn't come up with any other use for such key, and it's "available". And IMO there's no problem in using wood=deciduous: it can be assumed to mean that it consists of (mostly) broad-leaved deciduous trees, unless tagged with other tag combinations - an evergreen broad-leaved wood would then be wood=broad-leaved+foliage=evergreen, and anything like a Larch forest (do they exist in patches large enough naturally?) could wood=coniferous+foliage=decidous. Alv 10:53, 5 January 2012 (UTC)
Yes, I see now the reasons for type= not being used. I'll try to contact the tagging mail-list with further discussion of these topics. foliage= sounds potentially useful. I really just want some idea of the best way to map woodlands for mapping party I'm helping to organise in March New Lanark, Scotland. I am happy to put some serious enough into finding the best way to tag and give guidance for people wanting to map their local orchards, woods, hedgerows etc for access, foraging, maintenance, navigation, conservation, biodiversity etc. I'm keen to hear from anyone with examples, ideas around woodland mapping.Hawkeyes 12:59 5 January 2012
Current scheme is obviously controversial and does not reflect the actual features of forest. In the same time, keeping the old scheme and adding the new will also confuse people by presence of the same value in different meanings.

I could propose something like foliage=deciduous/evergreen/semi-deciduous/* and leaf=broad/needle/*. It reflects the significant part of options, described in UNEP-WCMC forest classification for temperate/boreal forests [1]. But, there is still intersection with wood=deciduous, for example. Another problem is mixed case. We should avoid using "mixed" value in case of more than one variant of mix, while putting several values divided by comma or something else is, again, not the best practice in OSM.--BushmanK (talk) 19:56, 19 January 2014 (UTC)

In relation to LCCS I propose leaftype=broadleaved/needleleaved/mixed/leafless. You need special knowledge to tag additonal leafphenology=evergreen/deciduous/semideciduous. foliage may be an alternative, but doesn't clear the distinction to leaftype. --Rudolf (talk) 09:16, 17 April 2014 (UTC)

ITO map for wood tags

To make up for my rash editing of this page, I have asked Ito maps to produce a map showing wood= tagged areas etc. Please chip in to the discusison here:

ito map wood=

Transformation wood= to forest= ?

While there are is a discussion about creating new keys I will complain to transform wood=* to forest=*.

The oxford dictionary says: wood = 1. The hard fibrous material that forms the main substance of the trunk or branches of a tree or shrub, used for fuel or timber. 2. (also woods) An area of land, smaller than a forest, that is covered with growing trees.

IMHO, this doesn't match the using of wood to describe the foliage of a large area covered chiefly with trees. The term forest matches better. To get a suitable expression, a change will be appropriate.

--Rudolf (talk) 11:49, 9 April 2014 (UTC)

I believe the reason for using "wood" as the key name was originally the concept of "subclassing", which is sometimes expressed in OSM by using one tag's value as the key of the subtag. So just like we have amenity=parking + parking=underground, people invented natural=wood + wood=deciduous.
Now this is obviously not a good solution, but renaming does little to improve it. If we change it (which involves changing existing data in the database, and the code of editors and applications), then we should also separate the concepts of "how does the foliage look" and "does it fall off in winter", as discussed above. --Tordanik 13:34, 9 April 2014 (UTC)

Draft of proposal "leaftype" and "leafcycle"

To get some clearance in the tagging of leaftypes I started a draft Proposed features/leaftype. Maybe the term foliage is the better choice than leafcycle. Comments are welcome. --Rudolf (talk) 15:13, 29 April 2014 (UTC)

Thanks for creating a proposal to solve this longstanding problem! I think your keys are fine, foliage or some other key would not necessarily be better. However, you should add underscores (leaf_type and leaf_cycle), because as far as I know these are not a single English word. --Tordanik 16:33, 30 April 2014 (UTC)
Good hint. Thank you.--Rudolf (talk) 16:35, 1 May 2014 (UTC)
I started the RFC. Comments are welcome at Talk:Proposed_features/leaftype. --Rudolf (talk) 18:57, 3 May 2014 (UTC)

The key "wood=" is deprecated

The key wood=* is deprecated. Please consider to use the approved key leaf_type=broadleaved/needleleaved/mixed instead of wood=deciduous/coniferous/mixed. Please consider also updating the tagging of existing elements that you have mapped yourself. --Rudolf (talk) 22:12, 6 June 2014 (UTC)

Lesser used, but distinctive uses of wood=*

I understand you have a goal of deprecate {{tag|wood} with this tag, but I cannot read from your documentation how to tag special cases, like wood=eucalypt, wood=casuarina, wood=nipa_palm, wood=palm, wood=filao. I understand that you want wood=evergreen to be tagged with leaf_type=broadleaved + leaf_cycle=evergreen, which I find more complicated (but maybe there is another reason behind?). Please explain how to mark woods of types Eucalypt, Casuarina, Palm, and Filao. Also explain how to tag Nipa-Palm, though they might be classified under natural=wetland similar to wetland=mangrove. The reason for this is that certain types of maps (i.e. Nautical Charts according to INT 1) renders this with different symbols. --Skippern (talk) 16:27, 29 June 2014 (UTC)

Please look at Proposed_features/leaftype. The goal is a definite distinction of broadleaved and needleleaved vegetation. The tag wood=evergreen is ambiguous, because there exist evergreen needleleaved and evergreen broadleaved woodland. Therefore we need two keys for all combinations. The values quercus, palm, nipa_palm, eucalypt, filao and casuarina can be moved to trees=* or species=* or genus=*. These keys are already in usage to define the species of plants. It is a good idea to make a decision about this question. (I propose to move this discussion to Talk:Key:wood, because this issue is not about leaf_type.) --Rudolf (talk) 12:33, 7 July 2014 (UTC)
Some examples:
  • genus=palmae
  • genus=nypa
  • genus=eucalyptus
  • genus=casuarina (same as filao tree?)
--Rudolf (talk) 15:05, 7 July 2014 (UTC)
Thanks for reply, though I find this system quite more complicated than the previous. I am not too familiar with casuarina and filao, but guessing from images and the fact that they have different symbols on nautical maps I guess they have quite different visual appearances. --Skippern (talk) 20:04, 11 July 2014 (UTC)
The original denotation of wood=* was to distinguish broadleaved and needle-leaved woodland, not to define the genus of plants. There is genus=* to define the low-level taxonomic rank of natural=tree, natural=wetland, natural=wood or landuse=forest. Why should we ignore an established tagging-scheme?
wood=filao is used nowhere in the database today. When you look at wikipedia:Filao_tree you will find "Genus: Casuarina". Every picture of a filao tree is also a picture of casuarina. Therefore I propose to use genus=casuarina instead of wood=filao. Everyone who needs a finer tagging, can use species=Casuarina equisetifolia. --Rudolf (talk) 07:47, 14 July 2014 (UTC)