EXPERIENCE IN APPLYING EXPERT ANALYSIS AND RANK CORRELATION DURING GEOLOGICAL AND FORECASTING WORK ON HYDROCARBONS

Problem Statement and Purpose. The diverse and extensive geological materials and geophysical data used in forecasting and prospecting operations for oil and gas exploration within the Black Sea’s northwestern shelf are usually interpreted in a complex manner. This complexity, however, is often spontaneous and fragmentary, with no new theoretical concepts taken into account. For example, the theory of global fluidogenesis represents an alternative hypothesis to the organogenic-sedimentary formation of hydrocarbon deposits but is commonly not considered.The authors of this article believe that the complexity should be replaced by a method of directed rank interpretation of geological information, in which ranking distinguishes the maximum information on one of the sections of geological work with regard to its completeness and consistency. The purpose of this study is to apply expert analysis and rank correlation of obtained information to geological and forecasting research on hydrocarbons. To achieve this goal, two basic principles, rank and direction, are considered. The data include seismic, tectonic-geodynamic, lithological, geochemical, mineralogical, and meiobenthic characteristics of the sea bottom scaled according to their sensitivity to the independent variable, i. e., hydrocarbons.Study area, material, and methods. The Pradneprovskaya area has been chosen for analysis. It is a promising location for oil and gas exploration in the shallow part of the inner northwestern shelf of the Black Sea. Geologically, this is the area of junction for the South Ukrainian monocline with the system of blocks included in the Predobrudzha-Crimean riftogenic trough. The area was studied by geophysical, geochemical (hydrocarbon gases, liquid hydrocarbons), mineralogical (X‑ray diffraction of clay minerals, thermobarogeochemistry of inclusions in minerals), and paleontological (foraminifera, nematodes, ostracods) methods, along with the drilling of a deep (2352 m) parametric well (Pradneprovskaya‑2).Results. It is shown that expert analysis of the diverse information within the framework of selected tasks, which are poorly formalized into a single system, provides the basis for an effective method of comparing information databases formed on different principles. This analysis is founded on the principle of directed (vector) ranking, where each rank narrows the search area and becomes more sensitive to the hydrocarbon deposit. The article gives an example of how this method can be used for a specific objective in its entirety, namely: (a) selection of information ranks (parameters), (b) selection of experts, (c) calculation of the rank matrix, (d) determination of hierarchical dependencies of ranks, (e) verification of the matrix, and (e) calculation of the concordance coefficient. The obtained results indicate that the proposed method of expert analysis with rank correlation under modern conditions using diverse geological materials that are not formalized into a single database, provides high reliability in the comprehensive interpretation of data.The method will help optimize geological and forecasting research on hydrocarbons within the Black Sea shelf.