[R] variograms and kriging

[Steve_Friedman at nps.gov]

Hello

Trying to develop variograms and kriged surfaces from a point file. Here is
what I've done so far.


library(gstat)  # also loads library(sp)
library(lattice)

 soilpts$x <- soilpts$UTM_X
 soilpts$y <- soilpts$UTM_Y
 soil.dat <- subset(soilpts, select=c(x, y, Area, BulkDensity, LOI, TP, TN,
TC, Total_Mg))

dim(soil.dat)
[1] 1292    7

 coordinates(soil.dat) <- ~ x+y

 gridded(soil.dat) <- TRUE

Warning messages:
1: In points2grid(points, tolerance, round, fuzz.tol) :
  grid has empty column/rows in dimension 1
2: In points2grid(points, tolerance, round, fuzz.tol) :
  grid has empty column/rows in dimension 2

 class(soil.dat)
[1] "SpatialPixelsDataFrame"
attr(,"package")
[1] "sp"
> bbox(soil.dat)
      min     max
x  476819  575981
y 2785749 2948128

> soil.dat[1:3,]

suggested tolerance minimum: 0.165318957771788
Error in points2grid(points, tolerance, round, fuzz.tol) :
  dimension 1 : coordinate intervals are not constant


The last error message and the warning returned above,  leads me to think
that the spatial sampling locations must be regular equally spaced.  My
data thou is not

I have spent the morning trying to figure this out - going back and forth
among many spatial packages that can do variograms and krigging.  Without a
good road map to follow however, I've had to a number of about faces.  Not
sure which way to turn now.

Can anyone provide guidance?

Using Windows WP and R 2.11.1 packages updated today.

Thanks
Steve


Steve Friedman Ph. D.
Spatial Statistical Analyst
Everglades and Dry Tortugas National Park
950 N Krome Ave (3rd Floor)
Homestead, Florida 33034

Steve_Friedman at nps.gov
Office (305) 224 - 4282
Fax     (305) 224 - 4147