The paper presents the results of applying an algorithm for creating a stochastic structural 3D geological model using Markov processes to constrain the resulting series of probability models. The analysis of methods for creating multivariate Markov chain models using an approximate Bayesian computational approach is presented to avoid specifying the likelihood function for a digital deposit model. A probabilistic geomodeling approach is justified for capturing uncertainty by automatically constructing a series of geomodels from perturbed input data sampled from probability distributions at the mineral deposit. An algorithm for creating a stochastic structural 3D geological mo del that can infer posterior distributions of the model paramet ers using topological information about the mineral deposit is described in detail. This may lead to unreliable estimates of the resulting digital deposit models, such as underestimation of small classes (i.e. classes with areas smaller than the average) in the simulated realizations.
А. А. Basargin (Wed,) studied this question.