Hello all, I am working on a set of bird locations from which I have calculated a kernel density estimate using kde (package: ks). I use kde because I can use the Plug-In bandwith calculation (see script below), which gives me good results.
I would like to know if it is possible to export the 50% and 95% contours to an Esri shapefile. I know that it is possible to do so using kernelUD (package: adehabitat): Corrie Curtice and Mathias Tobler gave an explanation in this same mailing list (http://www.mail-archive.com/animov@faunalia.it/msg00115.html). -> Does any one of you know if there is a way to do something similar with kde ? Thanks, Yvan Library(ks) bird1<- data.frame(read.table(file='C:/bird1csv',sep=',',header=TRUE,)) # Setting bandwidth to PI (Plug-In) H <- Hpi(x=bird1) # Running Kernel Density Estimator based on bandwidth = H fhat <- kde(x=bird1, H=H) # Plotting with 25,50,75% contours plot(fhat, display='filled.contour2', main = 'Plug-in', cont = c(50,90)) points (bird1, cex=0.5, pch=16) ###### # Adapted from Corrie Curtice's script (see link above) Library(adehabitat) data <- data.frame(read.table(file='C:/bird1csv',sep=',',header=TRUE,)) xy<-data[,c("x","y")] id<-data$GLS href <- Hpi(x=xy) kudl <- kernelUD(xy, id,h="href") kver<-getverticeshr(kudl) spol <- kver2spol(kver) df <- data.frame(GLS="2575") spdfBaja <- SpatialPolygonsDataFrame(spol, df, match.ID = FALSE) writeOGR(spdfBaja,"C:/Locations","bird1",driver="ESRI Shapefile") # Works but result of KDE using kernelUD doesn't give as much details as kde # direct transposition of kernelUD by kde does not work since the classes are different
_______________________________________________ AniMov mailing list AniMov@faunalia.it http://lists.faunalia.it/cgi-bin/mailman/listinfo/animov
