Jonathan Greenberg <greenberg <at> ucdavis.edu> writes: > > Ok, I've placed the input files and the PDF on a website > http://www.cstars.ucdavis.edu/~jongreen/temp/ > This an aliasing problem that disappears partially when using higher resolution, as noted in the quoted
# http://finzi.psych.upenn.edu/R/Rhelp02a/archive/90744.html of my first message (did you check it?). Dieter library(akima) # Providing data on the net as you did is fine, but you can make # life easier by directly providing the file url urlt = url("http://www.cstars.ucdavis.edu/~jongreen/temp/tahoedata02.csv") tahoecontourdata<-read.csv(urlt,header=TRUE) x=tahoecontourdata$Elevation_Bin y=tahoecontourdata$PRR z=tahoecontourdata$MaxStems # For easier reading, don't put too much into one expression, # but split it steps = 600 # fine grid intp = interp(x,y,z,duplicate="strip", xo=seq(1800,3200,length=steps), yo=seq(120,280,length=steps)) col=grey.colors(33,start=0,end=1,gamma=1) pdf("tahoe.pdf") # even at this resolution, the pdf picture has ugly kinks filled.contour(intp,col=col, nlevels=33,zlim=0:650) dev.off() # This was the workaround in # http://finzi.psych.upenn.edu/R/Rhelp02a/archive/90744.html pdf("tahoe1.pdf") image(intp$x,intp$y,intp$z,col=col) dev.off() # or use lattice (but needs some additional massaging) gr = expand.grid(x=intp$x,y=intp$y) gr$z= as.vector(intp$z) gr = na.omit(gr) library(lattice) pdf("tahoe2.pdf") p = contourplot(z~x+y,data=gr,region=TRUE) print(p) dev.off() ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.