There may be something in the RandomFields or georob packages.
--
Don MacQueen
Lawrence Livermore National Laboratory
7000 East Ave., L-627
Livermore, CA 94550
925-423-1062
On 4/13/15, 6:00 AM, "Carlo Innocenti" <carlo.innocenti at isprambiente.it>
wrote:
Hello,
anybody know a function to fit a theoretical 3D variogram model to the
experimental data?
I work with pollution data from boreholes samples of marine sediment.
Generally the model that best fits to the data is a nested variogram with
both geometric and zonal anisotropy, composed by:
- nugget effect
- spherical model with short range (< 1 m) along the z axis and very long
range (1000 km) along x and y axis
- spherical model with very long range (10^6 m) along the z axis and
medium range (< 1 km) along x and y axis
In the past I used Isatis to visually adapt the range of the model to the
experimental variogram and automatically fit the sill.
Now I'm moving on open source software and I'm looking for a way to fit at
least the sill of the 3D variogram to the data.
I found that gstat allows to use the 3D variogram, but doesn't fit the
anisotropic ones, and geoR fit the anisotropic variogram but not the 3D
ones.
Anybody knows a way in R to build and fit a variogram as that described
above?
Thanks,
Carlo
--
Carlo Innocenti, PhD
ISPRA
Istituto Superiore per la Protezione e la Ricerca Ambientale
High Institute for Environmental Protection and Research
Via Vitaliano Brancati, 60
00144 Roma
Italy
phone: +39 06 5007 4639
email: carlo.innocenti at isprambiente.it
[[alternative HTML version deleted]]