[R] How to use your own data in gstat and sp?

Roger Bivand Roger.Bivand at nhh.no
Mon Aug 17 12:34:24 CEST 2009

First re-read the definition of what a SpatialPixels object is -
?"SpatialPixels-class" - note that it says: "class for defining a pixels,
forming a possibly incomplete rectangular grid of arbitrary dimension". When
you try to coerce a SpatialPointsDataFrame to a SpatialPixelsDataFrame, it
is checked to see that the points form a "possibly incomplete rectangular
grid". Your points are random within a 1,10 square, so only regularly spaced
for a very special setting of the seed of the RNG, right? Since you are
unlikely to hit that very, very rare seed at random, you are not likely to
generate a regular grid in this way. Did you examine the class(), summary()
and str() of the input data in the examples?

Hope this helps,

Roger Bivand 

PS. consider following this up on the R-sig-geo list.

Paul Heinrich Dietrich wrote:
> This seems pretty basic, but I can't get any data to work except for the
> documented examples.  When the goal is to get to SpatialPixels, here is
> what I see...
>> x <- runif(10,1,10)
>> y <- runif(10,1,10)
>> z <- rnorm(10,0,1)
>> MyData <- as.data.frame(cbind(x,y,z))
>> library(gstat)
>> coordinates(MyData) <- ~x + y
>> gridded(MyData) <- TRUE
> suggested tolerance minimum: 0.893182669294186 
> Error in points2grid(points, tolerance, round, fuzz.tol) : 
>   dimension 1 : coordinate intervals are not constant
> I've spent way too much time trying to get my data to work.  Please help. 
> Thanks.

View this message in context: http://www.nabble.com/How-to-use-your-own-data-in-gstat-and-sp--tp24994744p25004460.html
Sent from the R help mailing list archive at Nabble.com.

More information about the R-help mailing list