[R] (no subject)

[Jonathan Greenberg]

I'm trying to hunt down an appropriate kriging package for my specific
application, and I was hoping someone on the R list might have some pointers
--  I'm interested in performing kriging and related spatial interpolations
with one of the R packages, but I need to be able to provide my own
point-to-point distances (e.g. I do not want to use standard between point
distances, as calculated by Euclidean or other similar distance measures).
Is there an R package that *I* can provide the matrix of distances between
every pair of points (e.g. for 10 points, I would have a 10 x 10 matrix of
distances)?  Similarly, if this is possible, can I then provide this package
a vector of distances from each point to an arbitrary unknown location (in
my example, this would be a 1x10 vector) and apply the model to this vector
to predict a single unknown point?  It seems most (if not all) of the
kriging packages I'm finding for R take the x,y,z location as the inputs and
calculate these distances themselves.  Thanks!

Sincerely,

Jonathan Greenberg

--
Jonathan A. Greenberg, PhD
Postdoctoral Scholar
Center for Spatial Technologies and Remote Sensing (CSTARS)
University of California, Davis
One Shields Avenue
The Barn, Room 250N 
Davis, CA 95616
Cell: 415-794-5043
AIM: jgrn307
MSN: jgrn307 at hotmail.com