Hi Tom,
Yes all I need is a simple list of overlaps. So it sounds like
gOverlaps() would be useful.
Thanks for the citation. That is a paper I have missed seeing.
Global distributions polys generally, with the exceptions of some
localized endemics exceed the country polys. So do not tile complete
areas and in most cases there may be large areas w/o overlap at all when
looking a selected countries.
A sample for a single species new shp that was created and the extracted
attribute tbls from ArcGisPro looks like this:
So the species taxa code is followed by the overlap (occurrence) of the
habitat/ecological attributes for Peru.
Prodav Sabana de Palmeras
Prodav Puna
Prodav Mar Frio de la Corriente Peruana (Humboldt)
Prodav Desierto Costanero del Pacifico Peruano
Prodav Paramo
Prodav Selva Alta (Yungas)
Prodav Serrania Esteparia
Prodav Mar Tropical
Prodav Bosque Seco Ecuatorial
Prodav Bosque Tropical del Pacifico Peruano
Prodav
This is being used for the new Interactive ID keys to the New World bat
calls that will eliminate species that do not occur in many habitats
I am making these keys freely available to everyone along with call Fact
Sheets since the automated black boxes in current software iterations
are not yet ready for Prime Time. For example the "classifiers" used by
K'Scope for Neotropcial areas are to data <35% correct and none can
really deal with multiple species in a single file. I provided 99.9% of
the base data for these classifiers and when challenged with novel data
sets not part of the training data they are still not robust. Same for
the U.S. with Florida on a current project I am working on.
Bruce
[[alternative HTML version deleted]]
_______________________________________________
R-sig-ecology mailing list
R-sig-ecology@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
