Spatial autocorrelation on a sphere
Assuming this is currently not possible, I would try modify nlme::dist to accept a spherical distance measure option, and have it call sp::spDists; corSpher has a metric parameter.
On 05/15/2012 09:55 PM, Dan Bebber wrote:
In the case where data are sampled on a sphere (e.g. the Earth) and the spatial locations are denoted by latitude and longitude, is there a way of generating a spatial autocorrelation structure that can derive Euclidean distances from these coordinates? I imagine that corSpher(form = ~ latitude + longitude, nugget = TRUE) would work over short distances, but would deteriorate as distance increased. One potential solution might be to derive distances from the spDistsN1 function in package sp, or the rdist.earth function in package fields, or the distm(x, y, fun=distHaversine) in package geosphere, and somehow to supply these to corSpher. Very grateful for any suggestions or comments. Dan Bebber
_______________________________________________ R-sig-Geo mailing list R-sig-Geo at r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Edzer Pebesma Institute for Geoinformatics (ifgi), University of M?nster Weseler Stra?e 253, 48151 M?nster, Germany. Phone: +49 251 8333081, Fax: +49 251 8339763 http://ifgi.uni-muenster.de http://www.52north.org/geostatistics e.pebesma at wwu.de