To summarize the steps of the process:

(1) Create a grid across the area of interest.

(2) In the center of each grid cell (zo), roughly estimate, using whatever information is at hand, the probability that a sample taken at that location would result in a measurement exceeding the release criterion.

(3) In the center of each grid cell (z0), roughly estimate the uncertainty of the estimate made in step 2.

(4) These estimates of : and F for the prior distribution of are converted into the parameters " and $ of a Beta distribution for each grid cell z0.

(5) Take hard data samples and convert the results to an indicator variable that is equal to 1 if the measurement exceeds the release criterion and 0 otherwise.

(6) Develop a reasonable exponential correlation model for the transformed values.

(7) Perform ordinary kriging at the center of each grid cell using transformed data and the correlation model developed in step 6. This gives you a kriging estimate of p(z0) at the center of each cell.

(8) Update the parameters " and $ of the Beta distribution.

(9) Calculate updated estimates of : and F.