Publication Date

2018-04-27

Availability

Embargoed

Embargo Period

2020-04-26

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PHD)

Department

Marine Geology and Geophysics (Marine)

Date of Defense

2017-08-25

First Committee Member

Christopher Harrison

Second Committee Member

Peter Swart

Third Committee Member

Harry DeFerrari

Fourth Committee Member

Jorge Willemsem

Fifth Committee Member

Shari Vaughan

Abstract

Seismic reflection has been an ongoing and active research discipline since the first experiments in 1921. Seismic reflection research has been overwhelmingly concentrated on the reflection characteristics of the subsurface. These characteristics are derived from instantaneous measurements of the seismic signal. Since the first seismic experiments, to the present day, analysis of the seismic signal stationary values for accurately defining a reflection horizon, and subsequently the morphological changes in the discrete strata, has been an ongoing goal of seismic processing. Initially, the interpretation consisted of imprecise conversion of event times to depth for determination of subsurface geologic structure. The advent of digital data recording in 1963 increased the dynamic range of the recorded seismic data, in essence increasing the difference between the largest and smallest signal that can be recorded, which allowed the investigation into the change in amplitude of the signal. This directly led to the identification of “bright spots” as hydrocarbon indicators. The success of this technique prompted other direct indicators of hydrocarbon deposits to be sought in the seismic signal. The “low frequency shadow” directly followed. This effect is caused by the shifting to lower frequencies as the seismic signal passes through a gas reservoir. Overlaying this derived data set onto the original seismic variable-area-wiggle trace with the initial sonograms, enabled the incorporation of a calculated data set into the established data display. These innovations, again, centered on the instantaneous derived quantities of the seismic signal. As the number of attributes proliferated, the association of these derived quantities with the actual underlying geology became increasingly tenuous, prompting intensive work in coagulating apparently unrelated attributes in the hope that the sum of the parts would be greater than the whole. These included response attributes and pattern recognition developments. However, these techniques were unsuccessful in providing any new geologic insights. The use of 3D seismic acquisition and interpretation revived signal processing research on attribute analysis. The ability to visualize horizons in the 3D cubes gave an added tool for interpreters. Although this advancement from 2D to 3D enhanced the accuracy of subsurface geological models, instantaneous attributes remained the critical component for understanding the subsurface through the reflected signals. The ability to move from the instantaneous, stationary analyses, to the analysis of the temporal and spatial morphological changes to the signal would give a method to describe, without a priori information, contiguous subsurface strata independent of source signal input. This research will demonstrate a methodology that moves to a continuous analysis of the seismic waveform, both temporally and spatially. The modification of the seismic signal as it moves through the surrounding medium is idiosyncratic to that medium group, and as will be demonstrated, it is independent of the source input signal to that material group. This will then enable comparisons of geological strata down line, to other seismic lines, and indeed to separate seismic acquisition data sets.

Keywords

seismic; fractal; m-sequences; cwt

Available for download on Sunday, April 26, 2020

Share

COinS