Environmental probabilistic quantitative assessment methodologies

Abstract

Probabilistic methodologies developed originally for one area of application may be applicable in another area. Therefore, it is extremely important to communicate across disciplines. Of course, a physical reinterpretation is necessary and perhaps some modification of the methodology. This seems to be the situation in applying resource assessment methodologies as environmental assessment methodologies. In this paper, four petroleum resource assessment methodologies are presented as possible pollution assessment methodologies, even though petroleum as a resource is desirable, whereas pollution is undesirable. It is ironic that oil as a precious resource in the ground can become a serious pollutant as a spill in the ocean. There are similarities in both situations where the quantity of undiscovered crude oil and natural gas resources, and the quantity of a pollutant or contaminant are to be estimated. Obviously, we are interested in making a quantitative assessment in order to answer the question, “How much material is there?” For situations in which there are a lack of statistical data, risk analysis is used rather than classical statistical analysis. That is, a relatively subjective evaluation is made rather than an evaluation based on random sampling which may be impossible. Hence, probabilistic quantitative assessment methodologies are needed for the risk analysis. A methodology is defined in this paper to consist of a probability model and a probabilistic method, where the method is used to solve the model. The following four basic types of probability models are considered: (1) direct assessment, (2) accumulation size, (3) volumetric yield, and (4) reservoir engineering. Three of the four petroleum resource assessment methodologies were written as microcomputer systems, viz., TRIAGG for direct assessment, APRAS for accumulation size, and FASPU for reservoir engineering. A fourth microcomputer system termed PROBDIST supports the three assessment systems. TRIAGG is the simplest assessment system, and therefore, more general in application. APRAS is more complex, and FASPU is the most complex. TRIAGG and PROBDIST are completely generic, or general probability systems, in that they are not dependent on the physical context, only on the structure. Hence, they would require no modifications for pollution assessment application. APRAS would require some modification, and FASPU even more; however, many of the ideas of both would be applicable. The three assessment systems have different probability models but the same type of probabilistic method. The type of probabilistic method is analytic probability theory instead of Monte Carlo simulation. Among the advantages of the analytic method are its computational speed and flexibility, making it ideal for a microcomputer. All four IBM-PC compatible microcomputer systems are available from the U.S. Geological Survey Books and Open-File Reports Section.