Hydrocarbon exploration is an economic and a probabilistic enterprise. Especially in mature provinces, where the giant discoveries have mostly already been made, we must incorporate probability and economics into exploration if our efforts are to be successful. Not even the largest companies have the resources to be active in all the exploratory plays possible; we find ourselves concentrating on only a small number of the available plays.

This choice of where to explore should be made by defining the objective of exploration and then concentrating on those plays which have the highest probability of achieving that objective. For a limited partnership drilling fund, the objective was formulated as a 3:1 present worth return on the money risked by the investors. A trend analysis process was developed which combines the probability of making a discovery with the probability distribution of reserves found to determine the probability of obtaining a desired return.

The industry's performance was analyzed for the years 1970 through 1981. For the foreseeable future we will be using the same technology as the 1970s and can reasonably expect the continuation of the trends of physical results (success ratios and the size distribution of discoveries). These physical results were combined with price and cost forecasts for the 1980s to obtain a realistic projection of expected exploratory success.

The steps in the analysis are as follows. (1) Geological Classification--A data base was developed containing the well information on 4,800 new-field and other exploratory wells drilled in south Louisiana. Each well was classified as to the objective formation and as to the producing formation(s) if successful. Twenty-nine separate trends were identified and analyzed. (2) Exploratory Drilling Data Analysis--A computer program was written to sort the well data by trend, project and exploratory success ratios for oil and gas, and prepare depth and cost analysis. (3) Reserves Added Analysis--Reserve estimates were made for 858 (557 gas, 301 oil) discoveries by projecting rate-cumulative production decline curves. Discovery sizes for the trends exhibited the expected log-normal frequency istributions. (4) Economic Projections--Using projected costs and product prices, a profile was developed for each trend of present worth profit as a function of discovery size. As usual, many discoveries can be expected to be "geologic successes but economic failures." (5) Probabilistic Analysis--Utilizing the Monte Carlo technique, a computer program was written to realistically simulate an n-well exploration program. The result for each trend was a cumulative frequency distribution of the return per exploration dollar. Using the same exploratory budget for all trends allowed us to rank trends based upon the probability of achieving the desired present worth return or better. Examples are presented for trends of varying rank.

In summary, computer data banks were used, along with thorough geologic analysis and some common sense, to provide a sound basis for concentrating exploration effort on those trends where we are most likely to achieve our objective.

End_of_Article - Last_Page 491------------