DEFINITION OF PETROLEUM RESERVES USING PROBABILITY ANALYSIS

Abstract

The realistic estimation of reserves and resources is important to many diverse groups including explorers, producers, auditors, taxmen, bankers, shareholders and governments. Reserves data are used in different ways for a variety of reasons and often the figures are used without adequate definition and/or recognition of the uncertainties associated with them. Any calculation method which fails to consider the uncertainties involved, cannot portray a realistic assessment of reserves.

Esso Australia Ltd. uses a relatively simple method to generate probability distribution curves in order to allow a more perceptive definition of the range of reserves for the offshore oil and gas fields in the Gippsland Basin and Esso is advocating wider petroleum and mineral industry acceptance of this approach.

The method involves defining data distributions for each of the reservoir properties (volume, porosity, water saturation, compressibility and recovery factor) which are multiplied using Monte Carlo Simulation to generate the distribution of reserves. Actual input consists of data from:

A high confidence area immediately surrounding well control, where the rock volume is relatively closely defined and the distributions of the other parameters, with the exception of recovery factors, reflect the observed variations.

Other areas which are only seismically controlled, where the data ranges reflect both observed and interpreted variations in volume (gross and net), porosity, water saturation, compressibility and recovery factor.

The curves generated for each area are then added by Monte Carlo Summation to yield the probability distribution of reserves for the whole field. In this method all available data are used and fewer subjective decisions are necessary. The computer generated distribution curves plot cumulative probability on the y-axis versus reserves on the x-axis. The curves allow the evaluation of the entire range of potential reserves, are valuable in economic and risk assessments and allow for more consistency in defining reserves for reporting purposes. The different categories of reserves, viz. "proved", "probable" or "possible", can be specified from the total field curves at defined probabilities. Moreover, the slope of the cumulative curve provides a direct indication of the level of knowledge of the field or parts of it.