Seismic Image Analysis Using Local Spectra

Lamoureux, Michael (2005) Seismic Image Analysis Using Local Spectra. Canadian Industrial Problem Solving Workshops > 9th IPSW [Calgary 15/5/2005- 19/52005].

Full text available as:

PDF - Requires Adobe Acrobat Reader or other PDF viewer.

Abstract/Summary

This report considers a problem in seismic imaging, as presented by researchers from Calgary Scientific Inc. The essence of the problem was to understand how the S-transform could be used to create better seismic images, that would be useful in identifying possible hydrocarbon reservoirs in the earth.

The important first step was to understand what aspect of the imaging problem we were being asked to study. However, since we would not be working directly with raw seismic data, traditional seismic techniques would not be required. Rather, we would be working with a two dimensional image, either a migrated image, a common mid-point (CMP) stack, or a common depth point (CDP) stack. In all cases, the images display the subsurface of the earth with geological structures evident in various layers.

For a given image the local spectrum is computed at each point. The various peaks in the spectrum are used to classify each pixel in the original seismic image resulting in an enhanced and hopefully more useful seismic pseudosection. Thus, the objective of this project was to improve the identification of layers and other geological structures apparent in the two dimensional image (a seismic section, or CDP gather) by classifying and coloring image pixels into groups based on their local spectral attributes.

Problem Statement

Calgary Scientific Inc. is currently developing a technique to classify pixels in seismic pseudosections based on their local spectral characteristics. These are obtained from a modified Gabor transform, in which only certain wavevectors are represented in the local spectrum of each pixel. This classification technique involves finding the dominant peak in each local spectrum, and identifying the corresponding pixel with the wavevector and amplitude of the dominant peak.

This method proved ineffective since it is inflexible when one wants to investigate any interesting features not associated with the dominant peak. The identification of secondary peaks is complicated by the fact that dominant peaks typically cover several pixels of its local spectrum; hence, the wavevector with the second largest amplitude is likely to contain a significant contribution from the primary peak. The main goals were to efficiently identify secondary peaks and to identify correlations amongst peaks from pixel to pixel. Ideally, a fully processed seismic pseudosection should contain all the information the
interpreter needs to unambiguously identify potential drilling targets, such as reefs, anticline traps, and so forth. This involves identifying layer boundaries on the pseudosection, a process that can be considered as more or less equivalent to identifying the layers themselves as continuous groups of pixels. The presence of significant amounts of noise in the data, and the accumulation of errors during the numerical steps of seismic processing, can complicate this process by obscuring important features in the pseudosection. A final goal is to improve the signal to noise ratio to facilitate interpretation of the pseudosection.

Archive Staff Only: edit this record