Point statistics may help to get an impression of the nature of your point data prior to for instance a point interpolation, and to find necessary input parameters for a Kriging operation.
For point statistics a point map is required in which:
The output of the Spatial correlation, Cross Variogram and the Pattern analysis operations is a table. By displaying the output table, you can also display graphs, especially semi-variograms. The output of the Variogram surface operation is a plot-like raster map.
Operations to calculate point statistics:
Regarding spatial autocorrelation and spatial variance is the user encouraged to compare his or her data set with a data set consisting of the same point locations, with a set of attribute values, approximately in the same range as the measured variable, but created at random (using one of the RND functions in Table Calculation). If the correlation/variance graphs are very much the same for the measured data and the random data, no autocorrelation exists between the data points. Hence, point interpolation is not useful.
By calculating semi-variogram values, you can display a semi-variogram. By modelling the semi-variogram, you can find the necessary input parameters, such as a model (spherical, exponential, etc.) and sill, range, and nugget values, for a Kriging operation.
Subsequently, you can calculate directional semi-variograms by using the directional method in the Spatial correlation operation. From the output table of Spatial correlation, you can prepare a semi-variogram model and investigate the range of the variable in the semi-variogram model both in the direction of anisotropy as well as in the direction perpendicular to it. Then, you are ready to perform Anisotropic Kriging.
From the output table of the Cross Variogram operation, you can create semi-variogram models for both variables and a cross-variogram model for the combination of the variables. All three models serve as input for the CoKriging operation. CoKriging calculates estimates or predictions for a poorly sampled variable (the predictand) with help of a well-sampled variable (the covariable). The variables should be highly correlated (positive or negative).
By inspecting the graphs of distances against probabilities, you may recognize distribution patterns of your points like random, clustered, regular, paired etc.
See also: