Usually, with that many observations, you use kriging in a local neighbourhood, i.e. use only the n nearest observations in a kriging system. If n is pretty large, this is practically equivalent to kriging with a global neighbourhood.
There are several other R packages that provide kriging, some of which provide local kriging, e.g. gstat does not require calculation of the full distance matrix. -- Edzer
______________________________________________ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
