Plot the experimental variogram.
Choose several models that appear to have the right shape and fit each in turn by weighted least squares (Cressie, 1985; McBratney and Webster, 1986) in an accredited program.
Plot the fitted models on the graph of the experimental variogram and assess whether the fit looks reasonable.
If all plausible models seem to fit well, choose the one with the smallest residual sum of squares (RSS) or smallest mean square.
Size of sample
Lag interval and bin width
Marginal distribution of the data
Anisotropy
Trend
How to choose an appropriate model function
How to judge fitting quality
Sample size influence
Source: https://scikit-gstat.readthedocs.io/en/latest/auto_examples/tutorial_03_variogram_models.html#sphx-glr-auto-examples-tutorial-03-variogram-models-py
Explore data
Combining knowledge of geologists
Azimuth angle in degrees clockwise from north
Number of lags
Distance between lags
Lag tolerance