[SOLVED] Variography with random samples

[niafall]

When the survey samples are randomly distributed in 2D space, can the variogram be calculated and fitted in the usual way, i.e.:

Code: Select all

experimentalVario <- vario.calc(db.data, lag = llag, nlag = nlag)
modelVario <-  model.auto(experimentalVario, struc = c("Spherical"))


Or are there other steps that must be taken due to the spatial distribution of the samples?

Many thanks!

[Fabien Ors]

Dear Nial
Sorry for our very late answer.
The experimental variogram can be calculated whatever the way your data are distributed.
You just have to make the variogram lag and lags number parameters compatible with the extension of your data.
You may need to define some calculation directions in the case where your data are not uniformly distributed.
Please look at our course on variography here : http://rgeostats.free.fr/doc/Courses/04_Variography.Rmd for more details.
Best regards,
Fabien