It seems only gstat can make a 3D nested variogram.
Recently I find georob is also a useful package to do 3D estimation with data have trend,
But I did not find a way to create a 3D nested variogram in georob,
Does anyone kown about this, or any other package for auto fit a 3D nested variogram?
Thank you in advance!
I would use GSLIB to estimate the empirical variogram followed by either R or Matlab to fit the nested models through constrained non-linear optimization.
That said, automated fitting of a nested model generally results in highly underdetermined sets of equations with multiple combinations of number of nested models and their parameters fitting the empirical estimates equally well. It usually requires some subjective choices. Looking at the physical geological setting usually helps to narrow the sets of plausible nested models and their ranges of parameter estimates.
I would use GSLIB to estimate the empirical variogram followed by either R or Matlab to fit the nested models through constrained non-linear optimization.
That said, automated fitting of a nested model generally results in highly underdetermined sets of equations with multiple combinations of number of nested models and their parameters fitting the empirical estimates equally well. It usually requires some subjective choices. Looking at the physical geological setting usually helps to narrow the sets of plausible nested models and their ranges of parameter estimates.
Dear Dasapta Erwin Irawan,
Thank you very much for your answer, but It seems that "geodata" fotomat in GeoR does not support "3D" data. Do you have any experience using geoR to estimate 3D data?
Dear Jhon and Bart, Thank you very much for the answer, I already tried gstat and get a 3D nested variogram with large/small range values in lateral/vertical respectively.
I aslo tried the new package:georob, "It provides functions for fitting linear models with spatially correlated errors by robust and Gaussian Restricted Maximum Likelihood and for computing robust kriging predictions". But I did not find a way to make and fit nested 3D variogram using it.
So the problem now is , for the vertical trend in different location are various due to ground water or fault. I hope these trends can be robustly nested in several 3D variogram models.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Statistical Software
- R Statistical Package