Proposal:Leaftype

The Feature Page for the approved proposal leaf_type/leaf_cycle is located at Key:leaf_type

The Feature Page for the approved proposal leaf_type/leaf_cycle is located at Key:leaf_cycle

leaf_type/leaf_cycle Proposal status: Approved (active) Proposed by: Rudolf Tagging: leaf_type=* Data item Applies to: Definition: Describes the type of leaves. Statistics: Rendered as: * Draft started: 2014-04-29 RFC start: 2014-05-02 Vote start: 2014-05-23 Vote end: 2014-06-06

leaf_type/leaf_cycle Proposal status: Approved (active) Proposed by: Rudolf Tagging: leaf_cycle=* Data item Applies to: Definition: Describes the phenology of leaves. Statistics: Rendered as: * Draft started: 2014-04-29 RFC start: 2014-05-02 Vote start: 2014-05-23 Vote end: 2014-06-06

Tag Count leaf_type=broadleaved $1% More details at taginfo leaf_type=needleleaved $1% More details at taginfo leaf_type=mixed $1% More details at taginfo leaf_type=leafless $1% More details at taginfo Tag Count leaf_cycle=evergreen $1% More details at taginfo leaf_cycle=deciduous $1% More details at taginfo leaf_cycle=semi_evergreen $1% More details at taginfo leaf_cycle=semi_deciduous $1% More details at taginfo leaf_cycle=mixed $1% More details at taginfo

Proposal

leaf_type=* describes the type of leaves, such as broadleaved or needleleaved.

leaf_cycle=* describes the phenology of leaves, such as evergreen or deciduous. Phenology is the study of cyclic and seasonal natural phenomena, especially in relation to climate and plant and animal life.

Rationale

Usage of "wood="

Value	Usage
deciduous	269 228
mixed	171 085
coniferous	140 542
quercus	902
yes	541
evergreen	352
broad-leaved	194
eucalypt	126
broad_leaved	97
palm	66
nipa_palm	4
casuarina	2

Usage of "type="

Value	Usage
broad_leaved	299 818
broad_leafed	96 697
broad_leaf	1243
broad-leaved	571
broad-leafed	434
conifer	28077
coniferous	1009
deciduos	2 632
deciduous	3 988
evergreen	230
palm	10782

There are several tags to indicate vegetation with leaves, e.g. landuse=forest, natural=wood, natural=scrub, natural=tree and others.

Today we have one additional key wood=* to describe the type of leaves. This key is arguable.

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.
• 2. (also woods) An area of land, smaller than a forest, that is covered with growing trees."

The usage of wood= for the type of leaves is not suitable because wood relates to the trunc, not to the leaves. Maybe it is suitable for natural=wood, but the key also relates to forest, trees, hedges and shrubs.

The values wood=coniferous and wood=deciduous are ambiguous. Coniferous describes the kind of seeds, not the type of leaves, nor the phenology, but it seems that it is used for the leaftype. Deciduous describes the phenology, not the type of leaves, but it seems that it is used for the leaftype.

Coniferous stands for "cone-bearing seed plants with vascular tissue". See coniferous on Wikipedia. This plants are mostly needleleafed, but there also exist some with broad leaves (e.g. Gingko biloba).

There exist also the tag wood=evergreen to describe the phenology, but not the kind of leaves. This applies to needleleaved and broadleaved plants.

There exist also the tag wood=yes to mark the presence of trees.

Other values in use are palm, nipa_palm, eucalypt and casuarina.

There is also an extended usage of type=broad_leaved; type=broad_leafed; type=broad_leaf; type=conifer; type=coniferous. The usage of type=* for non-relation elements should be avoided. It seems this is the most extended non-relational usage of type=*, dedicated mainly to natural=tree.

There exist three undocumented tags trees=broad-leaved, trees=coniferous and trees=mixed. The tag trees=* is used for the kind of trees in orchards. The values broad-leaved and coniferous are mismatching. It seems these tags refers to a CORINE import in Spain.

There is no tagging for evergreen broadleaved plants (e.g. most trees in the rain forest), nor for deciduous needleleaved plants (e.g. larchs).

It is very complicated to include all combinations of leave type and phenology in one key, especially in consideration that the phenology is not common knowledge in every case.

We need two distinct keys to describe the type and phenology of leaves, suitable for all kind of plants and all kind of keys.

It is advantageous to use new keys, to avoid confusion with changing values of an already used key. The key wood=* will be deprecated. It seems there is no proposal nor approval for this key.

Tagging

Leaf type

Tag	Comment	Wide-view picture	Close-up-view sample pictures
leaf_type=broadleaved	Broadleaved vegetation.
leaf_type=needleleaved	Needleleaved vegetation. Commonly coniferous.
leaf_type=mixed	Habitat with broadleaved and needleleaved vegetation.
leaf_type=leafless	Leafless vegetation.

Leaf phenology

Tag	Comment	Sample pictures
leaf_cycle=evergreen	Plants that are never entirely without green foliage. See evergreen on Wikipedia
leaf_cycle=deciduous	Plants that are leafless for a certain period during the year. See deciduous on Wikipedia
leaf_cycle=semi_evergreen	Plants that lose most, but not all, of their foliage for a fraction of the year.
leaf_cycle=semi_deciduous	Plants that lose their foliage for a very short period, when old leaves fall off and new foliage growth is starting.
leaf_cycle=mixed	Habitat with evergreen and deciduous vegetation.

The values relate to the Land Cover Classification System (LCCS) by FAO

The Global Forest Map (GFM) by UNEP-WCMC and the Global Land Cover 2000 Project by EU use the same terminology.

Applies to

• landuse=forest - Managed woodland.
• natural=wood - Woodland with no forestry.
• natural=scrub - Uncultivated land covered with bushes or stunted trees.
• natural=tree - Single tree, often significant.
• natural=tree_row - Line of trees.
• barrier=hedge - Line of closely spaced shrubs.
• landuse=orchard - Plantation of trees or shrubs maintained for food production.

Features/Pages affected

• wood=* - Will be deprecated!
• The values quercus, palm, nipa_palm, eucalypt, filao and casuarina can be moved to trees=* or species=*.
• trees=broad-leaved should be migrated to leaf_type=broadleaved
• trees=coniferous should be migrated to leaf_type=needleleaved
• type=broad-leaved (and others) should be migrated to leaf_type=broadleaved
• type=conifer (and others) should be migrated to leaf_type=needleleaved
• type=deciduous should be migrated to leaf_cycle=deciduous
• type=evergreen should be migrated to leaf_cycle=evergreen

Comparison of tags

New tag	Old tags
leaf_type=broadleaved	wood=deciduous wood=broad-leaved wood=broad_leaved type=broad_leaved type=broad_leafed type=broad_leaf type=broad-leaved type=broad-leafed trees=broad-leaved
leaf_type=needleleaved	wood=coniferous type=conifer type=coniferous trees=coniferous
leaf_type=mixed	wood=mixed trees=mixed
leaf_cycle=evergreen	wood=evergreen type=evergreen
leaf_cycle=deciduous	type=deciduous type=deciduos

Current use

leaf_type=*	leaf_cycle=*

wood=deciduous	wood=coniferous	type=broad_leaved	type=conifer

Possible rendering

Rendering of woodland

Tag	Rendering	Comment	Pictures
leaf_type=broadleaved		Broadleaved woodland.
leaf_type=needleleaved		Needleleaved woodland.
leaf_type=mixed		Mixed woodland.

Signature for tree rows

Tag	Signature	Comment	Picture
leaf_type=broadleaved	File:Signatur Laubbaumreihe.png	Broadleaved tree row.
leaf_type=needleleaved		Needleleaved tree row.

Comments

Comments are welcome. Please keep the discussion on the discussion page.

Voting

The voting is finished. There are 27 approvals, 2 rejections, 3 abstentions. Therefore the proposal is approved. Thanks to all voters.

• I approve this proposal. --Rudolf (talk) 16:32, 23 May 2014 (UTC)
• I approve this proposal. --Janko (talk) 17:25, 23 May 2014 (UTC)
• I approve this proposal. Escada (talk) 18:07, 23 May 2014 (UTC)
• I approve this proposal. --Dieterdreist (talk) 18:57, 23 May 2014 (UTC)
• I approve this proposal. --Tordanik 21:07, 23 May 2014 (UTC)
• I approve this proposal. --PeeWee32 08:35, 24 May 2014 (UTC)
• I approve this proposal. --Nounours77 (talk) 13:49, 24 May 2014 (UTC)
• I approve this proposal. —seav (talk) 20:53, 24 May 2014 (UTC)
• I abstain from this proposal. It has not been explained why common terms "broad-leaved" and "needle-leaved" should be written without a hyphen. I also wish to point out that, should this proposal be accepted, that will not be a license to make any automated tag changes to our existing database. --Frederik Ramm (talk) 06:38, 26 May 2014 (UTC)

There was no complaint about the syntax, therefore there is no explanation about it. The notation "broad-leaved" is used 194 times. The notation "needle-leaved" is used nowhere. I agree that this proposal is no license to automated mass-edits.

If you take a close look you can find the explanation in using the terminology of "LCCS by FAO". --Rudolf (talk) 07:50, 26 May 2014 (UTC)

The tag broad_leaved is actually used on over 300000 elements on OSM, broad-leaved on around 20000 (mainly the French CORINE import). See taginfo. SK53 (talk) 10:08, 6 June 2014 (UTC)

• I approve this proposal. Unixasket (talk) 09:39, 26 May 2014 (UTC)
• I approve this proposal. B108 (talk) 09:52, 26 May 2014 (UTC)
• I approve this proposal. LordOfMaps (talk) 09:53, 26 May 2014 (UTC)
• I approve this proposal.--Boscolindo (talk) 10:11, 26 May 2014 (UTC)
• I oppose this proposal.--KK-O (talk) 11:15, 26 May 2014 (UTC) We have three main types of forest in geographie: deciduous,coniferous or mixed. If we want more details, we should name the trees in the forest by their botanical names, but this should be a part of a botanical database. The name wood=* in OSM is defined as the forest/wood-type and that is okay and everybody knows the OSM-meaning.

The values of "wood=" are ambiguous. It's the same to distinguish warm or heavy. Also it is impossible to tag evergreen broadleaved forests, which are main parts of the worldwide treecover. We cannot consider forests in Europe only. --Rudolf (talk) 12:31, 26 May 2014 (UTC)

• I approve this proposal. streckenkundler (talk) 13:26, 26 May 2014 (UTC)
• I approve this proposal. Seichter (talk) 14:34, 26 May 2014 (UTC) But I doubt whether this tagging will get widespread acceptance.
• I approve this proposal.--Geri-oc (talk) 06:40, 27 May 2014 (UTC) siehe aber Diskussion (Grundtags)
• I abstain from this proposal. -- HillWithSmallFields (talk) 10:51, 27 May 2014 (UTC) I mostly like this proposal, but I don't like the leaf type values; not only would needle-leaved be better than needleleaved, but since the key already mentions that it is about leaves, I think the "leaf" word-element should be left out of the value, so I suggest the values should be simply "broad" and "needle"

The values "broadleaved" and "needleleaved" are consistent with the terminology of the "Land Cover Classification System (LCCS) by FAO". This system is used worldwide. --Rudolf (talk) 12:12, 27 May 2014 (UTC)

• I oppose this proposal. I do not believe that this is a better way to distinguish between different types of wood.--Hfst (talk) 20:40, 27 May 2014 (UTC)

With the existing tags, a definite distinction is all but impossible. --Rudolf (talk) 06:24, 28 May 2014 (UTC)

• I approve this proposal. This is a good way to clean up our tagging mess. But once this gets accepted, you need to push on it: Templates in JOSM, iD, write map creators to implement it... I'm curious about how this will work out. --Derstefan (talk) 20:47, 27 May 2014 (UTC)
• I approve this proposal. --User:nyampire (talk) 06:46, 28 May 2014 (UTC)
• I approve this proposal. Javiersanp (talk) 08:32, 29 May 2014 (UTC)
• I approve this proposal. simlox (talk) 15:55, 29 May 2014 (UTC)
• I approve this proposal. Bulwersator (talk) 20:32, 4 June 2014 (UTC)
• I approve this proposal. --Surly (talk) 04:47, 5 June 2014 (UTC)
• I approve this proposal. --NetWormKido (talk) 06:52, 5 June 2014 (UTC)
• I approve this proposal. FreeExec (talk) 06:11, 5 June 2014 (UTC)
• I approve this proposal. -- and I suggest to add the "scaleleaved" value for cypress and many Cupressaceae plants BushmanK (talk) 06:31, 5 June 2014 (UTC)

This proposal follows the classification of the LCCS. It is advantageous to keep the values as simple as possible. --Rudolf (talk) 06:35, 6 June 2014 (UTC)

• I approve this proposal. Felis_Pimeja (talk) 06:11, 5 June 2014 (UTC)
• I approve this proposal. Ullus (talk) 10:31, 5 June 2014 (UTC)
• I abstain from this proposal. I am wholly sympathetic to the objectives of this proposal, but am unconvinced that the specific tags here leaftype etc. are the best ones which can be used. In particular the existing wood=* has for many years provided basically the same concept. It has, of course, been polluted by wood=deciduous, a problem which needs to be resolved. As a botanist when I see leaf type I tend to think of terms like simple, crenate, entire, pinnate etc., not terms which relate to woodland. (leaftype can be used with trees, but if the species=*, genus=* or taxon=* are used this is redundant.) My feeling is that we should avoid creating new tags when what is needed is to improve the use of an existing tag, and that we should try when possible to find tag and key values which relate closely to normal speech. We talk about broad-leaved woodland, coniferous woodland, LaubWald, NadelWald, and so on which is why I my preference to keep wood for this purpose. Read Andy Allan's excellent description of what the issues are with changing tags. -- SK53 (talk) 10:08, 6 June 2014 (UTC)

As a "parallel inventor" of this scheme (I've started a couple of discussions and proposed almost the same variants of tagging), I can say, that it's impossible to improve/reorder the usage of "traditional" wood=*. First, literal meanings of values are orthogonal, not oppose to each other. And there is no way to tell everybody, that meaning was changed. Second, there is no way to find out, if some particular area was tagged with "old" meaning or "new" meaning (except some stupid thing like using "wood_meaning=new" flag, that is stupid). Third, I understand your point about leaf topology, but different sciences have different terminology.--BushmanK (talk) 23:24, 6 June 2014 (UTC)

• I approve this proposal. This is a better tagging system than the current one and kudos to Rudolf for this effort but this isn't the end of the story - we need to look at render/editor support, creating good examples and use cases and building and exploring edge cases. Copying the LCCS (which does have copyright) may be a good thing but no reason OSM has to stick with existing systems so we could amend to suit our needs in future. Also there's a question of what do we do with existing tags in OSM and writing more rendering rules...but don't get me wrong I support this but SK53 and others raise some important issues. --Hawkeyes (talk) 21:32, 6 June 2014 (UTC)