[Tagging] Refining landcover/natual landuse WAS Re: A name for stony ground?

[Martin Koppenhoefer]

2011/9/2 Johan Jönsson <johan.j at goteborg.cc>:
> It is an ambitious project and it would be nice if we could try to do something
> similar.


+1


> By there scheme bare_rock goes like this:
> Vegetation=no
> Wetland=no
> man_made_cover=no


yes, in the first phase


> Further differentiating could be done based on the surface structure, something
> in the line of fragmented=yes/no


in the second phase (they call it the modular/hierarchical phase) they do:

I surface aspect
II macropattern
landform,climate
altitude,erosion,vegetation
soil type/lithology

cheers,
Martin

PS: The FAO document is really interesting, what do you think about a
tag "vegetation_structure" to be applied to vegetated areas like
meadow, scrub and forest with the suggested values "open" and
"closed"? This is inspired by this scheme:
http://www.fao.org/docrep/003/X0596E/X0596e13.gif
and could help to differentiate between woodlands and dense forests or
between thickets and shrubland. It could also be used on beaches (some
of them are bare, others have sparse vegetation, e.g. there could be
"vegetation_structure=open" and "vegetation_structure=none" (where
"none" could include also be very sparse vegetation, almost not
present) and in mountaineous regions (where there is often areas which
are mixed pebbles and grass).

Another key to indicate the same could be "vegetation_density" with
values "dense" "sparse" (or "low").

Obviously this would also require to define that it is related to the
main vegetation form, i.e. the one in the "landuse" (or "landcover")
key.