STEP #2: Statistical Differences in Data

One key aspect of Zonal Kriging is to confirm that there are statistically different regions in the data set. A quick way to check if there are differences is to visually compare histograms of the data distributions (there are more robust statistical techniques).
Full Data Set: By itself, it's not possible to tell if there are sub-populations in this data set. Ideally, if the histogram is representative of the full data set, subsets should have similar distributions. A quick look at the distributions of the Fountain and Lyons Formations subsets suggests otherwise.

Fountain Formation: Data in this subset tends to favor indicators #4 though #7.

Lyons Formation: Data in this subset tends to favor indicators #1 though #4. Indicators #5 and #6 are very poorly represented, and indicator #8 occurs almost exclusively in this zone.

Based on this limited analysis, it appears that there are distinct zones.

June 1996