ABSTRACT

The objective of this paper is to compare three successive sampling models using nine simulated data sets and a Norwegian petroleum play. The three models are: the Geological Survey of Canada's lognormal discovery process model, Kaufman's anchored discovery process model, and a Norwegian discovery process model. Nine sub-samples were taken from a population with field size mean and variance known. The three

End_Page 159------------------------

values of the "discoverability" parameter were selected to be 0.25, 0.50, and 0.75; and three percentages (25%, 50% and 75%) of discovered fields were used in the sampling process. For example, a sub-sample was taken by applying the value of 0.25 and up to 25 per cent of the number of fields in the population. Similarly, a second sub-sample was taken by applying the value 0.5 up to 25 per cent of the number of fields in the population. This process was repeated seven more times. The three discovery process models were tested using these nine data sets.

The applicability of these three discovery process models and the statistical assumptions required are tested by the discoveries in a Jurassic play in the Norwegian part of the North Sea.

End_of_Record - Last_Page 160-------