Porosity prediction based on conventional and geostatistical image computation, Lower to Middle Eocene siliciclastics in the Tuz Gölü Basin (Central Anatolia, Turkey).

The study area is located at the east of Tuz Gölü Basin, one of the sedimentary basins in the Central Anatolia basins between Ş. Koçhisar and Karapınar town. The main subject of study is based on the Boyalı Formation consisting of braided river deltas and submarine fan deposits with shelf carbonates blocks of the Lower to Middle Eocene. Laboratory analysis showed that porosity and permeability values are between % 1.40–19.40 (average % 9.15) and 0.01–43.95 mD (average 5.07 mD), respectively. The conventional method is compared with the geostatistical image analysis method by rock plug samples parallel to layering for porosity prediction. These samples were saturated with blue resin and made thin sections. The boundaries of the blue areas (mean pores) were lined for pore volume calculation to create solid objects. A proposed empirical mathematical method, geostatistical based, was used to gain porosity values to converge to the laboratory values. Also, in the scope of this study, an architectural, mathematical logic called geostatistical first-order logic architecture (GFOLA) was generated. The geostatistical based programme "GeoST Porosity v.1.0" has been developed for numerical analysis. The empirical approach has been used to transform the areal measurement into volumetric measurement, eliminate the conditions that could not be measured in the 3rd dimension, and positively affects the porosity approach ratio, which is the ratio of empirical to laboratory porosity. In the empirical approach, the porosity is the function of grain size (x), sorting (y) and skewness (z) ∅ = g (x , y , z). To obtain the deviation from the "ideal condition (μ)", the yielded result has been modelled as an attenuation relation " e G f " which is a function of grain texture factor (Gf). A correlation graph of the grain texture factor and porosity approach ratio obtained by the empirical approach has been determined to be applied to siliciclastics. For practical purposes, a geostatistical modelling study proved that an empirical approach could be proposed as a better method instead of conventional laboratory porosity measurement. • Empirical mathematical method and geostatistical first-order logic architecture (GFOLA) was generated. • The conventional method is compared with the empirical geostatistical image analysis method for porosity predictions. • The geostatistical method porosity data were approximated to the measured in the laboratory at an average rate of 99%. • Via the empirical method, too much petrophysical data can be obtained from the different depths in the homogeneous rock. [ABSTRACT FROM AUTHOR]