AGH University of Science and Technology, Faculty of Geology Geophysics and Environmental Protection, Department of Geophysics, al. Mickiewicza 30, 30-059 Cracow, Poland
Principal component analysis, cluster analysis, and discriminant analysis were applied to well-logging data from the Miocene clastic formation in the Carpathian Foredeep, Poland. The main goal was to improve the results of interpretation of well logging in terms of determining gas-saturated horizons. The presented examples illustrate how statistical methods help limit the number of log data while preserving sufficient information. In addition, the two cases illustrate the grouping of data into clusters to reveal sets of features attributed to reservoir horizons and sealing layers and construction of discrimination functions to distinguish between gas- or water-saturated beds of sandy-shaly lithology.