Skip to content
Prev 4950 / 29559 Next
Original Raw

Kriging in gstat - error: cannot allocation vector of size

Richard Feldman 
Hello all,

I have been using gstat to krige temperature data with elevation as 
external drift.

feb01.meantemp.krig<-krige(MEANTEMP~elevation, locations= 
feb01.meantemp, newdata=elevation, model=feb01.meantemp.r)

I keep encountering the warning message:

Error: cannot allocate vector of size 20.7 Mb
In addition: Warning messages:
1: In as.list.data.frame(X) :
   Reached total allocation of 1535Mb: see help(memory.size)

I have been kriging separate datasets for each day of the year. Usually 
I can successfully krige 10 days or so before I run into these memory 
problems. I then have to close R and begin again.

To alleviate the problem, I reduced the surface onto which I am kriging. 
Right now it is a 1300km x 600km rectangle comprised of 250m x 250m grid 
cells. I am using about 80 weather stations for the interpolation. Is my 
grid really too big? I ask this because for the grids I do successfully 
interpolate I then use writeAsciigrid in Maptools and I run into the 
same problem: reached total allocation of 1535 Mb. But I'm not even 
interpolating here...just exporting the result as an Ascii grid. I find 
it hard to believe that my interpolation is out of the ordinary compared 
to what other gstat users do...or perhaps I should not be using a 
Windows computer with 2.00Gb of RAM?

Any ideas?

Thank you very much,

Richard