Martin Koppenhoefer <dieterdreist@...> writes: > > 2011/9/1 Johan Jönsson <johan.j <at> goteborg.cc>: > > > I have also looked at the UN-organization FAO that reasons about a scheme > > for > > tagging land cover... > > http://www.fao.org/docrep/003/X0596E/x0596e01f.htm > > > It's indeed interesting how they do landcover, they work in 2 phases. > In the first phase they use 3 simple steps to differentiate. > > 1. "Primarily vegetated=yes/no" and a criterion for vegetated with at > least 4% vegetation cover for at least 2 months of the year. > > this results in 2 classes, each of which is analyzed in a second step: > > 2. "Edaphic Condition=terrestrial/aquatic or regularly flooded" > aquatic includes marshes, swamps, bogs and all areas where water is > present for a substantial period regularly every year. > > the 4 resulting classes are then divided by their > > 3. "artificiality of cover" in "Artificial/managed" and in "(Semi-)natural" > > Those classes are then further refined in the second phase, when more > detailed landcover and environmental attributes like climate, erosion, > landform are added. > > A scheme is here: > http://www.fao.org/docrep/003/X0596E/X0596e10.gif > > cheers, > Martin >
It is an ambitious project and it would be nice if we could try to do something similar. By there scheme bare_rock goes like this: Vegetation=no Wetland=no man_made_cover=no Further differentiating could be done based on the surface structure, something in the line of fragmented=yes/no /Johan Jönsson _______________________________________________ Tagging mailing list [email protected] http://lists.openstreetmap.org/listinfo/tagging
