Assessment of geological uncertainties based on probabilistic models has become widespread in the last few decades. One of the main problems of probabilistic modeling is the complexity of justifying ranges of variable parameters. Analogs play an important role in solving this issue. In this paper, the specifics of utilization of analogous information in probabilistic geological modeling are considered, and tools for integrating analogs and actual data in the model at different states of geological exploration are proposed. So, to justify the range of acceptable values of a parameter with analogs, it is rational to use statistical criteria, rather than minimum and maximum values among analogs list. This leads to reduction of sample size influence on estimate reliability. With a small amount of information, distribution type of an analyzed quantity can be proposed analytically. Most geological parameters can be described by normal and lognormal distributions. For complex quantities, the uncertainty of which is formed by several factors, the distribution type depends on their exploration maturity. As a rule, the greater the uncertainty, the stronger the mode of distribution is shifted towards smaller values. The refusal to use analog data in a probabilistic model should be made only if the object studies are sufficient to characterize the entire volume of the accumulation. In other cases it is recommended to use both actual field data and parameters collected from analogs.

You can access this article if you purchase or spend a download.