Lithological trend analysis using well logging and geological data for improved reservoir rock identification – the Krosno Beds (Oligocene-Miocene), the Outer Polish Carpathians

Abstract

The Integrated Predictive Error Filter Analysis (INPEFA) technique was applied to investigate lithological variations with the aim of enhancing the identification of reservoir intervals within a sandstone-shale succession. The study utilized geological and well logging data from three wells penetrating the Krosno Beds in the Silesian Unit (Jurassic-Neogene) of the Outer Carpathians. Available well logs were reinterpreted to delineate petrophysical lithofacies based on shale content. Effective porosity (PHI) and estimated permeability values (K) were employed to calculate the Flow Zone Index (FZI) and classify Rock Types. Macroscopic core and cuttings descriptions, supplemented by field geological observations, facilitated the identification of geological lithofacies. These geological and petrophysical lithofacies, along with Rock Types, were correlated with INPEFA curves derived from well log parameters including gamma ray (GR), spontaneous potential (SP), bulk density (RHOB), and acoustic transit time (DT), as well as reinterpreted effective porosity (PHI) and shale volume (VCL). Facies classification within the sandstone-shale profiles of the Krosno Beds was refined by analysing INPEFA trends and inflection points. The integration of multiple analytical approaches based on various scale data, i.e. well logs, results of macroscopic analysis of cutting samples and cores together with information from geological field studies enabled a more precise delineation of intervals showing favourable reservoir properties within the stratigraphic columns. The study presented shows that information from various sources and on various scales, combined for a single purpose – lithofacies identification – results in a more accurate lithological recognition, which also affects understanding of reservoir properties. The trends obtained from the Cyclolog program made it possible to generalize very detailed and, by necessity, fragmented information from well logging and, at the same time, to responsibly incorporate independent, multi-scale data from core, cutting sample and field information.