KGS Open-File Report--2002-49--Part 2

KGS Open-file Report 2002-49

Prev Page--Report Start

Example--"D" Sand, Sooner Unit, Colorado

Objectives

Determine "D" sand thickness between well control points.

Identify faults/discontinuities which could be barriers to fluid flow in the reservoir.

Figure 7. "D" Sand Thickness from Wells. A large version of this figure is available.

Sooner 3-D seismic survey

Black horizons delineate seismic reflections corresponding to the top and bottom of the "D" sand. Over most of the 3-D survey area, the "D" sand is below seismic resolution (a "thin bed"). Below seismic resolution, reflections from the top and bottom of the sand maintain a constant temporal separation, which is unrelated to the true sand thickness. Envelope, frequency, and spectral decomposition attributes, which have been used elsewhere to estimate bed thickness, will be examined as potential predictors of "D" sand thickness

Figure 8. Vertical Seismic Section.

D sand is well delineated in seismic section

The envelope attribute has been calculated for the entire Sooner seismic data volume. A cross-sectional slice through the envelope volume gives a 2-D view of how this attribute varies in time. 3-D visualization can be used to display the spatial distribution of high envelope amplitudes (blue).

Figure 9. Cross-sectional view and 3-D visualization.

envelope attribute displayed in 2-D and 3-D

A map view shows how the envelope attribute varies laterally along the horizon corresponding to the top of "D" sand. This view is used to identify spatial patterns in the envelope amplitude which may correspond to changes in sand thickness. In general, higher envelope amplitudes correspond to thicker sands.

Figure 10. Envelope Attribute extracted along top "D" sand horizon. Black contours of "D" sand thickness based on well data. A large version of this figure is available.

A map of instantaneous frequency extracted along the horizon corresponding to the top of "D" sand. In general, lower frequencies correspond to thicker sands.

Figure 11. Instantaneous frequency extracted along top "D" sand horizon. Black contours of "D" sand thickness based on well data. A large version of this figure is available.

Spectral Decomposition--Selected frequency slices from the spectral decomposition of a 50 ms window centered on the "D" sand. Of these slices, the 30 Hz slice appears to be the best at imaging "D" sand thickness.

Figure 12. Spectral Decomposition--50 ms window centered on "D" sand. Black contours of "D" sand thickness based on well data. A large version of this figure is available.

A more detailed analysis of the spectral decomposition shows that 29 Hz is the best frequency for imaging "D" sand thickness.

Figure 13. Spectral Decomposition--29 Hz--50 ms window centered on "D" sand. Black contours of "D" sand thickness based on well data. A large version of this figure is available.

Crossplots of attributes versus "D" sand thickness give a quantitative estimate of the relationship between each attribute and sand thickness. Because the "D" sand is a thin bed, the seismic isochron does not indicate true bed thickness. Instantaneous frequency and envelope both show a gross correlation to "D" sand thickness, but there is a large amount of scatter in the data. The 29 Hz component of the spectral decomposition gives the best fit to "D" sand thickness.

Figure 14. Crossplots of attribute versus "D" sand thickness.

isochron

instantaneous frequency

envelope

spectral decomposition

Instantaneous phase and coherence, extracted along the "D" sand horizon, are examined for discontinuities/faults which may indicate reservoir flow boundaries.

Figure 15. Instantaneous phase. A large version of this figure is available.

NE-trending linear discontinuities in the northern portion of the survey area are visible on the coherence slice.

Figure 16. Coherence. A large version of this figure is available.

Sooner 3-D Summary

The 29 Hz Spectral Decomposition slice is the best single attribute for predicting "D" sand thickness.

A set of linear discontinuities in the northern part of the Sooner 3-D survey is best imaged on the Coherence map. These discontinuities may be faults, and are potential reservoir flow boundaries.

Conclusions

Attributes reveal information which is not readily apparent in the raw seismic data

Dozens of seismic attributes can be calculated using a variety of software packages

Attributes may be interpreted singly or using multi-attribute analysis tools

Different attributes reflect different physical properties of the underlying rock system

Acknowledgments

Access to The KINGDOM Suite software and the Sooner 3-D seismic data set was provided by Seismic Micro-Technology, Inc.

Example--"D" Sand, Sooner Unit, Colorado

Conclusions

Acknowledgments

Suggested Readings