Hello, I am interested in gstat package in R, and really would like to use it to do some kriging related functionality. but I am not sure how to achieve variogram modeling in gstat.
From the following example, it looks like vgm is used to achieve variogram modeling, my Q is how do I know that whether "Exp", "Sph", "Gau", or "Mat" is the right model for my data, or which parameter i should pick?.. Is there some set of functions I should use to find these parameters? or I should look at the data, figure it out myself?
--------------------------------------------------------------- "data(meuse) data(meuse.grid) m <- vgm(.59, "Sph", 874, .04) # ordinary kriging: x <- krige(log(zinc)~1, ~x+y, model = m, data = meuse, newd = meuse.grid)"
your help is greatly appreciated! yan
If you're familiar with the functions, you can pick the right one after plotting
the sample variogram. If not: try them:
- calculate sample variogram - plot sample variogram - estimate initial model by eye - fit.variogram() to fit the parameters - plot the sample variogram with the fitted model
and do that for all models, or until you're satisfied.
Please use the gstat-info mailing list for such questions. -- Edzer
