Seismic impedances calculated from logs show very good correlation to log porosities in wells penetrating the chalk reservoir in the Dan Field, Danish North Sea. This is the basis for an attempt to use seismic impedances derived from inversion as soft data for geostatistical reservoir characterization. The study focuses on porosity description of the Maastrichtian chalk reservoir unit, laterally restricted to an area covered by a subset of a 3D seismic survey. This seismic volume was inverted using the ISIS software producing a volume of seismic impedances.

Spatial porosity realisations are produced using a gaussian collocated cosimulation algorithm, where well log porosities constitute the hard data input and seismic impedances are the soft data input. The simulated volume measures 1400-1525 102 m and is oriented parallel to lines and crosslines in the seismic data set. Simulated blocks measures 25 25 6 m equivalent to twice the line and trace spacing, and approximating the seismic sample rate. The correlation coefficient between log porosities and impedances calculated from logs alone are shown to be misleading since they suggest unrealistic high coefficients. However, the inversion derived impedances and log porosities, still show correlation coefficients in the order of -0.45; sufficient to make the inversion result useful. It is remarkable that the calculated correlation coefficient is based on 15 wells, and the inversion is based on only one well. The impedances show inverse correlation to porosities. The impedance data indicate the level of average porosities at locations between wells. The fine structure is produced by the geostatistic process, with averages constrained by seismic impedances. The inversion derived seismic impedances are thus shown to constitute useful primary data to constrain reservoir characterization between wells.

This content is only available via PDF.
You can access this article if you purchase or spend a download.