Abstract
The petrophysical properties of Miocene mudstones and gas bearing-heteroliths were the main scope of the work performed in one of the multihorizon gas fields in the Polish Carpathian Foredeep. Ten boreholes were the subject of petrophysical interpretation. The analyzed interval covered seven gas-bearing Miocene horizons belonging to Sarmatian and Badenian deposits. The water saturation in shaly sand and mudstone intervals was calculated using the Montaron connectivity theory approach and was compared with Simandoux water saturation. Additionally, the Kohonen neural network was used for qualitative interpretation of four PSUs (petrophysically similar units), which represent the deposits of comparable petrophysical parameters. This approach allowed us to identify the sediment group with the highest probability of hydrocarbon saturation. Then, the spatial distribution of PSUs and reservoir parameters was carried out in Petrel. The resolution of the model was selected to reflect the variability of log-derived parameters. The reconstruction of the spatial distribution of shale volume, porosity, and permeability was performed with standard parametric modeling procedures using the Gaussian random function simulation stochastic algorithm, while PSU distribution and hydrocarbon saturation (SH) required a separate approach. The distribution into PSU groups was carried out by facies classification. Predefined ranges of clay volume, effective porosity, and permeability were used as discriminators to achieve spatial distribution of the PSU groups. The spatial distribution of hydrocarbon saturation was performed by creating the meta-attribute of this parameter and then reducing the derived pseudo-saturation model to physical values. Results included the creation of maps of hydrocarbon saturation that show the preferable areas with the highest hydrocarbon saturation for each gas horizon.