[Talk-GB] Large swaths of "heath" on Dartmoor

[SK53]

Moor (or possibly fell) covers a decent amount of Corine data imported
across Europe as natural=heath. In effect natural=heath on OSM no longer
means heath. It may mean any of the following:

   - Upland vegetation in its broadest sense: unimproved upland grassland,
   drier blanket bogs (covered by heather), Racometrium heath, Bilberry
   dominated heath, Shrubby vegetation dominated by brooms (at least in France
   & Spain), and no doubt a few others I've missed.
   - Moorland in Britain, which is probably a slightly smaller subset of
   the above
   - Lowland heathland: places like the Surrey Heaths, Suffolk Sandlings,
   Norfolk Brecks etc.
   - Other less obvious lowland areas known as heaths: particularly with
   large swathes of bracken and patches of birch.

When this thread first started I thought we could work to remove these
multiple meanings, but having seen what places with natural=heath from
Corine imported-data in the Cevennes,  suspect that this is an unrealistic
objective.

The alternatives are to start sub-typing natural=heath, with heath or
heath:type. The main category to identify in the short-term are the classic
lowland heaths which are scarce & threatened in the UK.

Wikipedia has a partial tabulation
<https://en.wikipedia.org/wiki/Heaths_in_the_British_National_Vegetation_Classification_system>
of the formal heath categories in the National Vegetation Classification,
which may help as background reading. I'm sure that pretty much all
communities in the U-group (calcifugous grassland & montane), several Mires
(e.g., M15 & M16), and even some calcicolous upland grasslands are included
in current natural=heath.

At a more practical level the JNCC Phase 1
<http://jncc.defra.gov.uk/PDF/pub10_handbookforphase1habitatsurvey.pdf>
guide recognises 6 heath categories, of which 4 are relatively common: wet
& dry heaths, and their respective mosaics with grassland. Anything where
the peat depth in the soil is NOT regarded as a heath, but will be a Mire
community (pennine moorland will be largely blanket bog in this
terminology).

Both NVC <https://wiki.openstreetmap.org/wiki/Key:plant_community> & Phase1
<https://wiki.openstreetmap.org/wiki/Proposed_features/Key:habitat> have
relevant pages on the wiki for (slightly) further info. NVC is clearly far
too technical for just about everyone, but Phase1 is probably usable with a
small bit of guidance.

Probably the best way to take this forward is to compile good examples of
places people are likely to know (particularly in National Parks) which
have a known classification AND a reasonable number of usable images on
Geograph. Wales is the easiest place to do this because the whole of the
country was mapped using Phase1.

Regards,

Jerry




On 25 September 2017 at 12:28, ael <law_ence.dev at ntlworld.com> wrote:

> On Thu, Feb 09, 2017 at 10:10:07AM +0000, SK53 wrote:
> > than anything they reflect that OSM as a project lacks good tags for many
> > of these boreo-temperate upland features, and whilst that is true there
>
> I have been changing some "heath" areas of Dartmoor to "moor". But I
> notice that the wiki claims that this is deprecated.
>
> Since most of these are large areas which really cover a variety of
> vegetation, I can't see that any of the "established" tags are really
> appropriate. "Moor" is exactly right.
>
> If forced to use the documentated tags, I would go for
> natural = grassland;wetland as the best approximation despite the
> fact that not everything is wet nor is grass.
>
> Of course, it only makes sense for coarse-grained approximate mapping,
> and more localised accurate tags are the ideal.
>
> Should "moor" or something similar be restored and supported by
> renderers?
>
> ael
>
> _______________________________________________
> Talk-GB mailing list
> Talk-GB at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-gb
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-gb/attachments/20170925/6d747425/attachment.html>