HANDBOOK OF GEOPHYSICAL EXPLORATION SEISMIC EXPLORATION

VOLUME 1

BASIC THEORY OF EXPLORATION SEISMOLOGY

HANDBOOK OF GEOPHYSICAL EXPLORATION SEISMIC EXPLORATION Editors: Klaus Helbig and Sven Treitel PUBLISHED VOLUMES 1984 - Mathematical Aspects of Seismology. 2 nd Enlarged Edition (M. Båth and A.J. Berkhout)* 1984 - Seismic Instrumentation (M. Pieuchot) ISBN 0-08-036944-8 1984 - Seismic Inversion and Deconvolution (a) Classical Methods (E.A. Robinson)* 1985 - Vertical Seismic Profiling (a) Principles. 2 n d Enlarged Edition (B.A. Hardage)* 1987 - Pattern Recognition & Image Processing (F. Aminzadeh)* 1987 - Seismic Stratigraphy (B.A. Hardage)* 1987 - Production Seismology (J.E. White and R.L. Sengbush)* 1989 - Supercomputers in Seismic Exploration (E. Eisner)* 1994 - Seismic Coal Exploration (b) In-Seam Seismics (L. Dresen and H. Rüter)* 1994 - Foundations of Anisotropy for Exploration Seismics (K. Helbig) ISBN 0-08-037224-4 1998 - Physical Properties of Rocks: Fundamentals and Principles of Petrophysics (J.H. Schön) ISBN 0-08-041008-1 1998 - Shallow High-Resolution Reflection Seismics (J. Brouwer and K. Helbig) ISBN 0-08-043197-6 1999 - Seismic Inversion and Deconvolution (b) Dual-Sensor Technology (E.A. Robinson) ISBN 0-08-043627-7 2000 - Vertical Seismic Profiling: Principles. 3 d Updated and Revised Edition (B.A. Hardage) ISBN 0-08-043518-1 2001 - Seismic Signatures and Analysis of Reflection Data in Anisotropic Media (I. Tsvankin) ISBN 0-08-043649-8 2001 - Computational Neural Networks for Geophysical Data Processing (M.M. Poulton) ISBN 0-08-043986-1 2001 - Wave Fields in Real Media: Wave Propagation in Anisotropic, Anelastic and Porous Media (J.M. Carcione) ISBN 0-08-043929-2 2002 - Multi-Component VSP Analysis for Applied Seismic Anisotropy (C. MacBeth) ISBN 0-08-0424439-2 2002 - Nuclear Magnetic Resonance. Petrophysical and Logging Applications (K.J. Dunn, D.J. Bergman and G.A. LaTorraca) ISBN 0-08-043880-6 2003 - Seismic Amplitude Inversion in Reflection Tomography (Y. Wang) ISBN 0-08-044243-9 2003 - Seismic Waves and Rays in Elastic Media (M.A. Slawinski) ISBN 0-08-043930-6 2004 - Quantitative Borehole Acoustic Methods (X. Tang and A. Cheng) ISBN 0-08-044051-7 2004 - Seismic While Drilling - Fundamentals of Drill-Bit Seismic for Exploration (F. Poletto and F. Miranda) ISBN 0-08-043928-4 * Book out of print.

SEISMIC EXPLORATION

Volume 1

BASIC THEORY OF EXPLORATION SEISMOLOGY by John K. Costain and Cahit Çoruh Department of Geological Sciences Virginia Polytechnic Institute and State University Blacksburg, VA, U.S.A.

2004

ELSEVIER Amsterdam - Boston - Heidelberg - London - New York - Oxford Paris - San Diego - San Francisco - Singapore - Sydney - Tokyo

ELSEVIERB.V. Sara Burgerhartstraat 25 P.O. Box 211, 1000 AE Amsterdam, The Netherlands

ELSEVIER Inc. 525 B Street Suite 1900, San Diego CA 92101-4495, USA

ELSEVIER Ltd. The Boulevard Langford Lane, Kidlington, Oxford OX5 1GB, UK

ELSEVIER Ltd. 84 Theobalds Road London WC1X8RR UK

© 2004 Elsevier Ltd. All rights reserved. This work is protected under copyright by Elsevier Ltd., and the following terms and conditions apply to its use: Photocopying Single photocopies of single chapters may be made for personal use as allowed by national copyright laws. Permission of the Publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying for advertising or promotional purposes, resale, and all forms of document delivery. Special rates are available for educational institutions that wish to make photocopies for non-profit educational classroom use. Permissions may be sought directly from Elsevier's Rights Department in Oxford, UK: phone (+44) 1865 843830, fax (+44) 1865 853333, e-mail: [emailprotected]. Requests may also be completed on-line via the Elsevier homepage (http://www.elsevier.com/ locate/permissions). In the USA, users may clear permissions and make payments through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; phone: (+1) (978) 7508400, fax: (+1) (978) 7504744, and in the UK through the Copyright Licensing Agency Rapid Clearance Service (CLARCS), 90 Tottenham Court Road, London W1P 0LP, UK; phone: (+44) 20 7631 5555; fax: (+44) 20 7631 5500. Other countries may have a local reprographic rights agency for payments. Derivative Works Tables of contents may be reproduced for internal circulation, but permission of the Publisher is required for external resale or distribution of such material. Permission of the Publisher is required for all other derivative works, including compilations and translations. Electronic Storage or Usage Permission of the Publisher is required to store or use electronically any material contained in this work, including any chapter or part of a chapter. Except as outlined above, no part of this work may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the Publisher. Address permissions requests to: Elsevier's Rights Department, at the fax and e-mail addresses noted above. Notice No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made.

First edition 2004 Library of Congress Cataloging in Publication Data A catalog record is available from the Library of Congress. British Library Cataloguing in Publication Data A catalogue record is available from the British Library.

ISBN: ISBN CD-ROM: ISSN:

0-08-037019-5 0-08-044589-6 0950-1401 (Series)

@ The paper used in this publication meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). Printed in The Netherlands.

Working together to grow libraries in developing countries www.elsevier.com

www.bookaid.org

www.sabre.org

CONTENTS

1

Contents 1 Introduction 1.1 2

3

Acknowledgments

15 23

Groundwork 2.1 Complex numbers 2.1.1 Manipulation of Complex Numbers 2.1.2 Real and Complex Exponentials and Trigonometric Functions 2.1.3 Powers and Roots of a Complex Number 2.1.4 Logarithm of a Complex Number 2.1.5 Functions of a Complex Variable 2.1.6 Representation of Signals by Phasors 2.1.7 Linear equations

24 24 25

Fourier Transforms 3.1 Introduction 3.2 Signal Nomenclature 3.2.1 FT—Continuous time, continuous frequency 3.2.2 DFS—Continuous time, discrete frequency 3.2.3 DTFT—Discrete time, continuous frequency 3.2.4 DFT—Discrete time, discrete frequency 3.3 The Fourier Coefficients 3.3.1 Sign Convention 3.3.2 Determination of the Fourier coefficients 3.3.3 Fourier Coefficients from Linear Equations 3.3.4 Numerical Example 3.3.5 Fourier Coefficients from Orthogonality 3.3.6 Numerical Example 3.3.7 Dirichlet conditions 3.3.8 Summary 3.3.9 Sign Convention Revisited 3.3.10 Seismogram from the Atlantic Coastal Plain 3.3.11 Independence of the Fourier coefficients 3.4 From Fourier Series to Fourier Integrals 3.4.1 Complex forms and Fourier Integral

39 39 41 42 43 43 45 48 52 55 55 56 57 64 64 65 74 75 77 83 85

28 30 31 33 34 35

2

CONTENTS 3.4.2 Fast Fourier Transform 3.5 Applications of Fourier Transforms 3.5.1 Time-shifting theorem 3.5.2 Time differentiation of the Fourier transform 3.5.3 Time integration of the Fourier transform 3.5.4 Introduction to the unit-impulse function 3.5.5 The sine function 3.5.6 Definition of a linear System 3.5.7 Impulse Response 3.5.8 The time-convolution theorem 3.5.9 Autocorrelation and Crosscorrelation 3.5.10 The frequency-convolution theorem 3.5.11 Hilbert Transforms 3.6 2-transform 3.6.1 Factors of a finite, discrete function 3.6.2 Phase of minimum and maximum delay couplets 3.6.3 Amplitude and phase of a z-transform 3.6.4 Introduction to filters

89 93 93 97 98 102 107 109 109 Ill 120 123 129 163 165 168 169 175

4 Computational Considerations 4.1 Effect of Analysis Window on Fourier Spectrum 4.2 Aliasing 4.2.1 Sampling in the time domain - aliasing in the frequency domain 4.2.2 Example

185 185 188 189 190

5 Synthetics and Velocity Functions 5.1 Normal-incidence reflection coefficient 5.1.1 Example 5.2 Values of Reflection Coefficients 5.3 The Zoeppritz Equations 5.4 AVO and Zoeppritz equations in T-X domain 5.5 Synthetic Seismograms 5.6 The Reflectivity Function 5.7 Velocity functions 5.7.1 Impulse (thin bed) 5.7.2 Step 5.7.3 Ramp 5.7.4 Wavelet Tuning 5.7.5 Summary of velocity functions 5.8 Seismic trace attributes

195 195 199 201 204 208 214 218 219 221 221 223 231 238 242

6 Traveltime curves and velocity 6.1 Snell's law 6.1.1 The ray parameter p 6.2 Reflection Traveltime Curves

253 254 257 258

CONTENTS

3

6.2.1 The root-mean-square velocity 266 6.2.2 Determination of interval velocities using Dix Equation . 271 6.2.3 Effect of dip on reflection traveltime curves 286 6.2.4 The normal moveout correction 290 6.3 Refraction traveltime curves 293 6.3.1 Refractions from a single horizontal interface 296 6.3.2 Delay time 298 6.3.3 General expression for delay time 302 6.3.4 Two-layer model 302 6.3.5 General expression for multilayer refraction traveltime curves from horizontal layers 303 6.3.6 Reflection and refraction traveltime curves combined— comments 305 6.3.7 Effect of dip on refraction traveltime curves 305 6.3.8 The Principle of Reciprocity 311 6.3.9 Refraction traveltime curves over various geologic models 312 6.3.10 Dipping plane interfaces 317 6.3.11 Linear increase in velocity with depth 322 6.3.12 Turning waves 325 6.4 Composite Refraction-Reflection Stacks 325 7 Seismic Source Wavelets 7.1 Energy Sources 7.1.1 Dynamite 7.1.2 Vibroseis 7.1.3 DinoSeis, Thumper, and others 7.1.4 Marine 7.2 Mathematical Descriptions of Wavelets 7.2.1 Ricker wavelet 7.2.2 Klauder wavelet 7.2.3 Comments 7.3 Wavelet z—transform representation 7.3.1 Physical requirements for real wavelets 7.3.2 A simple 2-point wavelet 7.3.3 Generation of wavelets 7.3.4 Partial energy of a wavelet 7.3.5 Roots plotted in the z-plane 7.3.6 Root on the unit circle

338 342 342 342 352 352 354 355 355 357 358 359 361 363 366 367 368

8 Wavelet Shaping and Deconvolution 8.1 Inverse infinite filters, finite input 8.1.1 Inverse filtering of a minimum-delay 2-term wavelet using 2-transforms 8.1.2 Inverse filtering of a maximum-delay 2-term wavelet using 2-transforms 8.1.3 Inverse filtering of a seismic trace using z-transforms . . .

370 373 379 384 392

4

CONTENTS

8.2 Inverse finite filters, infinite input 8.2.1 Exact filters for wave guides 8.3 Inverse filters and input each of finite length 8.3.1 General shaping and least-squares method 8.3.2 Predictive deconvolution 8.3.3 What does predictive deconvolution do? 8.3.4 Summary guidelines for predictive deconvolution 8.3.5 Predictive deconvolution—Conclusion 8.4 Spectral whitening 8.5 Further Applications of Hilbert Transforms 8.5.1 Relationship between the amplitude and phase spectrum of a causal function 8.5.2 Q References Index

394 394 418 420 454 480 497 499 507 514 514 518 549 564

LIST OF FIGURES

5

List of Figures 1.1 1.2 1.3 1.4

Seismic record section Crustal profile Seismic data cube Seismic record section—Coal

16 18 19 20

2.1 2.2 2.3

Graphical representation of a complex number Polar coordinates Roots of a complex number

25 28 32

3.1

Vibroseis seismogram obtained by the Virginia Tech seismic crew on the Atlantic Coastal Plain near Richmond, VA Fourier series (DFS) application A discrete-time signal Relation between angular frequency and linear frequency . . . . Frequencies required for a periodic series Dirichlet conditions Gibbs phenomena Time-symmetric sine wave Time-symmetric cosine wave Sum of time-symmetric sine and cosine waves Early shot record from Bellshill Lake, Alberta A seismic trace Fourier decomposition and synthesis: Rectangular pulse . . . . Fourier decomposition and synthesis: Triangular pulse Atlantic Coastal Plain seismic data Fourier transform of infinite and windowed sinusoid Fourier harmonics are independent of each other Fourier harmonic interference Effect of width of analysis window Effect of shape of analysis window Frequencies required for a periodic series F F T procedure Effect of time shift on the Fourier spectrum Integrated seismic trace A snapshot in time during wavelet integration Comparison of trace integration

3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 3.11 3.12 3.13 3.14 3.15 3.16 3.17 3.18 3.19 3.20 3.21 3.22 3.23 3.24 3.25 3.26

40 45 45 49 50 66 67 68 68 69 70 71 73 74 76 78 79 80 81 82 84 94 96 99 100 101

6 3.27 3.28 3.29 3.30 3.31 3.32 3.33 3.34 3.35 3.36 3.37 3.38 3.39 3.40 3.41 3.42 3.43 3.44 3.45 3.46 3.47 3.48 3.49 3.50 3.51 3.52 3.53 3.54 3.55 3.56 3.57 3.58 3.59 3.60 3.61 3.62 3.63 3.64 3.65 3.66 3.67 3.68

LIST OF FIGURES Spectra of differentiated and integrated Klauder wavelet . . . . Seismic trace and its derivative Integrated seismic trace Spectra of integrated Trace 18 Temperature log and the geothermal gradient—effect of differentiation Rectangular pulse centered about t = 0 Derivation of frequency content of Dirac delta function A few cosine waves before summing to get Dirac delta function Approximation to Dirac delta function by summation of a few cosine waves Graphic display of derivation of frequency response of Dirac delta function Relation between period of rectangular pulse and spacing of null points in the Fourier transform of the pulse Definition of a linear system Response of linear system to an impulse Impulse response of a filter Convolution caution The reflection process Source wavelet, reflection coefficients, and seismic trace The Coherence Cube Convolution in the frequency domain Recovery of true spectrum is not possible What is a Hilbert transform? Why 90°? What if h(t) is not causal? Phasor rotation convention for Hilbert transform Overview of Hilbert transform symmetry Reconstruction of causal function from even part Hilbert transform of sinusoid (a) Hilbert transform of sinusoid (b) Fourier transform of sign(t) Hilbert transform of a continuous function using —i sign[w . . . Fourier transform of 1 /n t Hilbert transform of sine wave using Hilbert transformer . . . . Hilbert transforms of cosine wave - frequency domain Graphic display of quadrature filter Hilbert transform of continuous data by swapping Hilbert transform of discrete data by swapping He(u) from Ho{u) Continuous phase spectrum Unit circle in the z-plane Frequency response of P(z) = —z\ + z Complex conjugate roots Single root plotted in z-plane

102 102 103 103 104 105 105 107 107 108 108 109 110 110 115 118 119 122 127 128 130 131 133 134 135 137 139 140 143 144 145 149 150 151 154 155 160 162 165 170 171 172

LIST OF FIGURES

7

3.69 3.70 3.71 3.72 3.73 3.74 3.75 3.76

Unwrapped phase of maximum-delay wavelet Vector plot of P{z) = -zi+z Plot of minimum-phase P(z) = —z\ + z Frequency response of an averaging filter Frequency response of a differencing filter Simple notch filter Simple low-pass filter Butterworth low-pass filter

173 173 174 177 177 180 182 183

4.1 4.2 4.3 4.4 4.5 4.6 4.7

Fourier transform of complex exponential Effect of truncation on Fourier spectrum Effect of width of analysis window on Fourier spectrum Effect of shape of analysis window on Fourier spectrum Two sinusoids showing why aliasing occurs Example of an aliased spectrum How to check for aliasing

186 187 187 188 191 192 194

5.1 5.2 5.3 5.4

Velocities and densities of Atlantic Coastal Plain sediments . . Domain of real reflection coefficients from Zoeppritz equations . Angles of reflection and transmission for incident P—wave . . . Zoeppritz equations for various angles of incidence and velocity ratios 5.5 AVO common midpoint geometry and energy partition 5.6 Example 1 of amplitude variation with offset 5.7 Example 2 of amplitude variation with offset 5.8 Peterson Model for synthetic seismograms 5.9 Reflection coefficients obtained directly from impedance contrast versus approximation from derivative of Log(v) 5.10 The only three types of velocity functions needed to model a transversely isotropic earth 5.11 The thin bed velocity function 5.12 Thin-bed detection by integrated energy spectra 5.13 Analytic convolution of source wavelet with step function . . . 5.14 The ramp velocity function 5.15 Integration of a source wavelet by a ramp velocity function . . 5.16 Ratio of Fourier transform of differentiated source wavelet to transform of source wavelet 5.17 Integration of source wavelet by triangular velocity function . . 5.18 Ricker wavelet—50 Hertz 5.19 Wavelet differentiation 5.20 Wavelet interaction as a differentiation 5.21 Effect of wavelet differentiation on amplitude spectrum 5.22 Analytically differentiated Ricker wavelet 5.23 Effect of layer thickness on reflection amplitude 5.24 Wavelet interaction as an integration 5.25 Effect of wavelet integration on amplitude spectrum

202 205 206 207 210 213 214 216 220 222 224 225 227 228 229 230 231 233 234 234 235 235 236 236 237

8

LIST OF FIGURES

5.26 5.27 5.28 5.29 5.30 5.31 5.32 5.33 5.34

Analytically integrated Ricker wavelet Ricker wavelet (80 Hz) Velocity functions w(t) from Sengbush et al. [161] Velocity functions w(t) and their transforms Seismic attribute—quadrature trace Seismic attribute—quadrature trace Seismic attributes—trace, instantaneous phase and frequency Seismic attributes—trace, envelopes Hilbert transform of vibrator sweep

6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 6.10 6.11 6.12

Physical significance of Snell's law 254 Derivation of Snell's law 255 Another application of Snell's law 257 Snell's law extended downward along a raypath 274 Apparent velocity 275 Emergence angle of wavefront and slope of traveltime curve . . 275 Hyperbolic traveltime curve from single interface 276 Reflection traveltime curves from 2-layer model 277 Notation for an n-layer reflection model 278 Traveltime curves from a 5-layer model 279 Traveltime curves from a 5-layer model with low-velocity layer . 279 Wide-angle reflections to 10 km from 4-layers with low-velocity layer 280 Ray parameter versus offset for reflections from multilayer model280 Ray parameter versus offset for reflections from model with uniform velocity change 281 Determination of accurate reflection traveltimes 281 Ficticious straight-line vrms reflection raypath versus actual path282 Comparison of traveltime computations using exact parametric equations, the rms velocity, and the third term in the power series expansion 283 Another comparison of traveltime computations using exact parametric equations, Equation (6.15), and the third term in the power series expansion 284 RMS velocity (ft/sec) distributions in major oil-producing sediments 284 Trace gather for an rms velocity search 285 Velocity spectrum 285 Velocity panel used to determine stacking velocities 286 Geometry for derivation of traveltime curve from a single dipping reflector 287 Effect of dip on the reflection traveltime curve from a single dipping interface 287 Effect of dip on reflection moveout 289 Normal moveout correction applied to reflections from zero-dip reflector 291

6.13 6.14 6.15 6.16 6.17

6.18

6.19 6.20 6.21 6.22 6.23 6.24 6.25 6.26

237 238 239 241 248 249 . 250 251 252

LIST OF FIGURES 6.27 6.28 6.29 6.30 6.31 6.32 6.33 6.34 6.35 6.36 6.37 6.38 6.39 6.40 6.41 6.42 6.43

9

6.49 6.50 6.51 6.52 6.53 6.54 6.55 6.56

Normal moveout correction applied to 20° dip 292 Reflection and refraction arrivals 294 Refraction raypath geometry for single interface 297 Real-data examples of refraction head waves 299 The concept of delay time 300 Delay time for single segment in third layer of n-layer model . . 301 Three-layer refraction model 304 Reflection and refraction traveltime curves combined 306 Down-dip refraction traveltime over single dipping interface . . 307 Up-dip refraction traveltime over single dipping interface . . . . 309 Reversed refraction traveltime curves over single dipping interface311 First arrival data for reversed refraction profile of Figure 6.37 . 313 Reversed refraction profile over the Price Mountain anticline . . 314 Reversed refraction profiles over vertical fault to surface . . . . 315 Reversed refraction profile over fault of large throw 316 Reversed refraction profile over fault of small throw 317 Reversed refraction profiles over buried high-velocity layer and buried vertical fault 318 Adachi's derivation of refraction traveltime curves 319 Raypath geometry for linear change of velocity with depth . . . 322 Geometry of refraction raypath for linear increase of velocity with depth 324 Traveltime curve for linear increase in velocity with depth . . . 324 Turning rays for normal reflection and for reflections from salt overhang 326 Refraction stacks—refraction moveout 1 329 Refraction stacks—refraction moveout 2 329 Refraction stacks—multfold refraction raypaths 330 Refraction stacks—dipping refractor 332 Refraction stacks—composite refraction-reflection stack . . . . 334 Refraction stacks—comparison with reflection 335 Refraction stacks—two refractors, same line 335 Refraction stacks—shallow comparison with reflection 336

7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 7.9 7.10 7.11 7.12

Oscillatory character of a general pulse Minimum-delay wavelet Geophone arrays Vibroseis method Vibroseis method Failing Y-1100A vibrator Shear vibrator Example of processing single-sweep vibrator data Single-vibrator over crystalline terrane Tapered versus untapered vibroseis sweep Single-vibrator over Paleozoic shelf strata Single-vibrator on Atlantic Coastal Plain sediments

6.44 6.45 6.46 6.47 6.48

339 340 341 343 344 345 346 347 349 350 351 353

10 7.13 7.14 7.15 7.16 7.17 7.18 7.19 7.20 8.1 8.2 8.3 8.4 8.5 8.6 8.7

LIST OF FIGURES Marine seismic source Ricker wavelet Fourier transform pair Klauder wavelet Fourier transform pair Minimum-delay wavelet from impulsive source Real seismic wavelet decays to zero Modulus of e~iu is radius of unit circle Single root and its complex conjugate Root of z-transform on unit circle

354 355 357 359 362 363 368 369

Complex 2-term wavelet 383 Positive-imaginary complex minimum-delay 2-term wavelet . . 384 Negative, real, minimum-delay 2-term wavelet 384 Inverse filtering of complex, maximum-delay 2-term wavelet . . 389 Plot of inverse-filter coefficients for 2-term wavelet 390 Complex time-domain equi-delay 2-term wavelet 391 Complex minimum-delay 2-term wavelet with root close to the unit circle 391 8.8 How to generate a synthetic seismic trace 393 8.9 Water-layer wave guide 394 8.10 The "effective source wavelet" 395 8.11 Response of wave guide, thickness 11 m 397 8.12 Response of wave guide, thicknesses of 15, 30, 60 meters . . . . 399 8.13 Effect of seismometer depth on wavelet character 400 8.14 Response of wave guide for water depths of 3, 6, and 9 meters . 401 8.15 Water reverberations—Persian Gulf 401 8.16 Water reverberation spectra—Persian Gulf 402 8.17 Backus 2-point filter for long-period water-confined reverberations404 8.18 Backus 2-point filter for short-period water-confined reverberations 405 8.19 Backus 3-point filter for long-period deep reflection reverberations409 8.20 Backus 3-point filter for short-period deep reflection reverberations 410 8.21 Neidell's 2-layer reverberation model 411 8.22 Neidell's 2-layer reverberation model - Example 1 412 8.23 Neidell's 2-layer reverberation model - Example 2 419 8.24 Neidell's 2-layer model with convolved wavelet 419 8.25 Crosscorelation sign convention 425 8.26 Procedure for obtaining positive-lag crosscorrelation coefficients 426 8.27 Procedure for obtaining negative-lag crosscorrelation coefficients 427 8.28 Minimum-delay wavelet shaping by least-squares filter 429 8.29 Correlations using Figure 6-2 from Robinson and Treitel . . . . 430 8.30 More correlations from Robinson and Treitel Figure 6-2 . . . . 434 8.31 Example (c) of Robinson and Treitel 436 8.32 Example (d) from Robinson and Treitel 437 8.33 Wavelet shaping using least-squares filter and maximum-delay input 438

LIST OF FIGURES 8.34 8.35 8.36 8.37 8.38 8.39 8.40 8.41 8.42 8.43 8.44 8.45 8.46 8.47 8.48 8.49 8.50 8.51 8.52 8.53 8.54 8.55 8.56 8.57 8.58 8.59 8.60 8.61 8.62 8.63 8.64 8.65 8.66 8.67 8.68 8.69 8.70 8.71 8.72 8.73 8.74 8.75

11

Example (a) from Figure 6-2 of Robinson and Treitel 440 Wavelet shaping of mixed-delay input 441 Example from Figure 6-3 of Robinson and Treitel 444 Wavelet and roots of mixed-delay wavelet 445 Figure 6-3 of Robinson and Treitel[153] with correlations added 446 Figure 6-3 of Robinson and Treitel[153] with correlations and shift added 447 Inverse filtering of Vibroseis signal 452 Shaping of vibroseis wavelet 453 Predictive deconvolution of minimum-delay wavelet 456 Autocorrelation of wavelet and seismic trace 459 Comparison of wavelet and trace autocorrelations 460 Input and desired output for predictive deconvolution 466 Prediction filter output— 1 468 Prediction filter output— 2 469 Predictive deconvolution matrix comparison 474 Minimum-delay wavelet shaping by predictive deconvolution . . 476 Spiking deconvolution of a minimum-delay wavelet. Prediction distance a = 1. (PredictiveDeconvolution.nb —> PredictiveDeconvolutionWavelets.cdr —» PDmin2.wmf) 478 Predictive deconvolution of longer minimum-delay wavelet . . . 479 Effect of predictive deconvolution on wavelet 480 Deconvolution to a "hanging wavelet" 483 Second zero-crossing 484 Second zero crossing? 485 Predictive deconvolution applied to non-minimum-delay wavelets486 Predictive deconvolution of seismic trace 487 Trace energy after predictive deconvolution 488 Effect of design window width on predictive deconvolution . . . 490 Observable effects of wavelet truncation 491 An alternative to spiking deconvolution 492 Alpha versus thin beds 494 Effect of noise on predictive deconvolution 501 Removal of 1 s t order long-period multiples by predictive deconvolution 502 Removal of 2 n d order long-period multiples 503 Removal of 1st order short-period multiples 504 Backus filter coefficients from z-transforms and predictive decon 505 Autocorrelation of reverberations 506 Spectral separation by stretching 509 Model data after spectral separation by stretching 510 SAGC with and without amplitude balancing 511 Application of SAGC on real data 512 Spectral whitening of vibroseis data 513 Phase spectra of minimum-, mixed-, and maximum-delay wavelets517 Absorption definition 522

12 8.76 8.77 8.78 8.79 8.80 8.81 8.82 8.83 8.84 8.85 8.86 8.87 8.88

LIST OF FIGURES Experimental data underlying linear-with-frequency assumption 523 Absorption coefficient a versus frequency v 523 Absorption without dispersion 526 Dispersive velocity and Hilbert transform 531 Dispersion curve in Pierre Shale 536 Body wave dispersion curves for various values of Q 537 Discrete versus continuous formulation of dispersion 538 What if absorption is not linear-with-frequency? 540 Absorption impulse response of the earth for Q = 20, a = 1500 m541 Comparison of absorptive impulse response computed from Gladwin and Stacey [78] with that from Equation (8.127) 542 Wavelet spreading for Q = 30 544 Wavelet shape change for Q - 20 548 Klauder wavelet shape change for Q = 20 at distance 100 m . . 548

LIST OF TABLES

13

List of Tables 3.1 3.2 3.3 3.4 3.5 3.6

Notation for the Fourier transform (FT) 42 Notation for discrete Fourier series (DFS) 44 Notation for the Discrete Time Fourier Transform (DTFT) . . . 46 Notation for the Discrete Fourier Transform (DFT) 46 Fourier Transform Pairs 126 Examples of minimum, maximum, and mixed delay functions and their z-transforms 168

5.1 Velocities and densities of Paleozoic shelf strata 203 5.2 Notation for conversion of Zoeppritz equations to time-offset domain 211 5.3 Trigonometric relations for AVO forward modeling 212 7.1 Examples of 2-term wavelets 7.2 Examples of convolution of binomials

362 365

8.1 Summary of equations for absorption-dispersion pairs

539

This page is intentionally left blank

Chapter 1

Introduction The material in this volume provides the basic theory necessary to understand the principles behind imaging the subsurface of the Earth using reflection and refraction seismology. The end products are a "record section" such as that shown in Figure 1.1 and, from a collection of "wiggly traces" that are recorded in the field (see Figure 3.1 on page 40 for an example of a wiggly 42-trace seismogram), derived information about subsurface structure and rock properties. For the most part, the principles of imaging are the same regardless of the depth to the target; i.e., the same mathematical background is necessary for targeting a shallow water table as for investigating the deep Mohorovichic discontinuity (the "Moho"), which marks the bottom of the earth's "crust" at a depth of 30-50 km. It is generally accepted that of all the geophysical techniques reflection seismology provides the best resolution and the most information about the subsurface. It is also the most expensive method, and for this reason has not yet been widely applied to help define the geometry of near-surface aquifer systems; however, the potential is there. This method of exploring the subsurface has been developed and highly refined by the petroleum industry in the search for hydrocarbons onshore and offshore. The images of the subsurface obtained by reflection seismology must generally be refined and sharpened before a geologic interpretation can be made, especially for data acquired on land. In this regard, Fourier theory is important in order to understand the physical meaning of the seismic "wiggly trace", and the collections of these that we refer to as a "record section". Practical details of reflection or refraction data acquisition or processing are not included in this volume nor are field procedures and techniques of proper data acquisition; however, they are all equally important. The reader is referred to the excellent treatments of data acquisition and processing in this Series and elsewhere. Benyamin [16] provided Mathematica programs in his tutorial on key elements of total field design. A good image requires proper acquisition; the image can often be further improved (made more meaningful) by data processing. The basic theory that underlies such techniques is included in this volume. 15

16

CHAPTER 1. INTRODUCTION

Figure 1.1: A 2-dimensional display of multi-trace seismic data called a "record section". Such data are now acquired in such a way as to produce the 3dimensional cube of data shown in Figure 1.3. Seismic data courtesy of Paradigm Geophysical.

We summarize the mathematical concepts required to understand reflection seismology data sets used in exploration geophysics. Because the authors have a combined experience of almost 100 years in reflection seismic multifold data acquisition, processing and interpretation, and because we operated our own field crew at Virginia Tech for ten years, we are able to include some results of our own applications of the theory discussed in this volume. Much of our research has been directed toward imaging the shallow and deep crust down to the Mohorovichic discontinuity with the objective of understanding the internal architecture and structural evolution of the Appalachian orogen in the eastern United States. We include illustrations of seismic images, many of which we generated ourselves, of the major geologic provinces of the Appalachian orogen. The Alleghanian plateau, folded Valley and Ridge, overthrust (allochthonous) Blue Ridge, the metamorphic core of the orogen (the Piedmont, Figure 1.2), and the younger sediments of the Atlantic Coastal Plain that conceal the eastward

17

extension of the exposed Piedmont, yield images of a complex tectonic history that are a necessary complement to the surface boundary conditions provided by geologic maps. We hope that including portions of seismic record sections of the internal geometry of a major orogen in this kind of a book on "basic theory" will help to convey the excitement that accompanies the merger of the disciplines of geology and geophysics. Although geologists and geophysicists work with completely different kinds of basic field data, they share a common interest and enthusiasm in understanding how these data collectively contribute to our understanding of the tectonic evolution of mountain belts. In viewing a two-dimensional seismic "record section" or a three-dimensional volume of seismic data (Figure 1.3) it is desirable to have some appreciation of how different mathematical operations effect the basic data acquired in the field. The fundamental seismic "unit" is the seismic source wavelet that leaves the source point. The wavelet is reflected or refracted from the subsurface geometry and returns to the surface where it is recorded. After much data processing, the arrival times of reflected wavelets and the nature of the distortion of the recorded wavelets yield the seismic record section and the properties of the rocks that the wavelets have traversed. This is not a book on processing techniques. There are many references (Yilmaz, [204]) to this subject including the massive two-volume series by Yilmaz [203]. Seismic data processing is a dynamic and changing field with new processing techniques continually emerging. In the course of interpreting reflection data acquired over the Appalachian orogen in the southeastern United States. Peavy [135] showed the advantages of introducing geology early in the processing sequence by summing reflection points along the direction of tectonic strike. In this volume we wish to illustrate more general topics such as how the shape of the source wavelet can be affected by, for example, "body wave dispersion" caused by intrinsic damping, and how we can use the Hilbert transform as a special application of Fourier theory to examine the effect that body wave dispersion has on changes in wavelet shape. Some appreciation of how certain velocity distributions in common geologic models can behave as mathematical operators to change the shape of the seismic wavelet somewhere between the source and receiver is useful to a geologist as well as a geophysicist. What effects does the geology have on the shape (frequency and phase) and propagation of the seismic wavelet because of the earth materials? In this volume we review reasons why only three simple "velocity functions" are required to generate any synthetic sequence of reflection coefficients, no matter how complicated the geology (page 219).

I—'

00

I I Figure 1.2: Portion of a conventional multifold (24-fold) seismic reflection line across the Appalachian orogen in central Virginia. The seismic signature of the arcuate/eastward-dipping folded and thrust-faulted reflectors is continuous along strike for over 400 km from Virginia to Georgia [50]. Crustal velocity is 6 km/sec. These structures cradle most of the earthquakes in the Central Virginia Seismic Zone. Prom Coruh et al. [45], Pratt et al. [139], and Lampshire et al. [104]. Data acquired by the U.S. Geological Survey and reprocessed by Pratt [138].

o

1

19

Figure 1.3: Modern seismic data acquisition and processing has matured to become a remarkably successful 3-dimensional effort. The 3-dimensional "cube" of data shown in this figure is the end product of 3-D data acquisition and processing. Data courtesy of Exxon.

The material in this volume is appropriate for shallow geophysical applications such as coal exploration (Figure 1.4) as well as for deeper studies. Indeed, the best recent images of ground penetrating radar (GPR) are the result of processing such shallow data in a manner similar to that used for tectonic studies and hydrocarbon applications. It is just a question of scale.

to

o

i

I Figure 1.4: Top: An interpreted record section over shallow coal deposits in the Illinois Basin showing bimodal paleochannel development at about 150 ms two-way traveltime in the shallow Herrin Coal Complex. The Herrin No. 6 seam, the major target in the area, is at a depth of about 160 m and is 2 to 3 m in thickness. Interruptions in seam thickness by paleochannel development have a direct bearing on the economic feasibility of mining operations. Bottom: Uninterpreted section. Field data courtesy of Conoco, Inc. Reprocessed and interpreted by Weisenburger [188].

o

i

21

Most of the material in this book has been offered in courses for geophysics majors at Virginia Tech. A prerequisite to these courses is a course commonly referred to as "Engineering Mathematics" where basic concepts such as complex numbers, vectors, differential equations, and Fourier theory are covered. We begin this volume with a brief summary of these subjects. Fourier theory is used extensively in geophysics to analyze and process seismic and other kinds of data. The examples given in this volume apply the theory developed in the text. The intent of the presentation is to make the reader "computer literate" rather than "icon literate." There is an advantage to having a background that enables one to work and think independently, to know both the theory and how to develop and implement algorithms. You can be sure of what someone else's programs do, and be better positioned to test and challenge canned software. More important, when you get an idea about something theoretical or practical, you will be able to test it yourself. The philosophy behind the intensive use of programming by our students to illustrate and apply the lecture material can be summarized by the following: Tell me and I'll forget. Show me and I may remember. Involve me and I'll understand ... Chinese proverb The elaborate and beautifully efficient commercially available icon-oriented software for production processing of reflection seismic data is often not the simplest way to develop an idea or educate oneself about the inner workings of what happens when you click on an icon. It is hoped that this volume will contribute to an understanding of what lurks beneath some of the icons. The courses taught by the authors at Virginia Tech required a background in Fortran programming, which in the last couple of years migrated to Mathematica [198]. Mathematica, first released in 1988, is a fully integrated environment for technical computing. Individual packages have long existed for specific numerical, algebraic, graphical and other tasks but the concept of Mathematica was to create a single system that could handle all the various aspects of technical computing in a coherent and unified way. Mathematica combines symbolic manipulation, numerical mathematical methods, outstanding graphics output suitable for publication, and a sophisticated programming language. For a user with even a modest background in programming, the transition from Fortran (or from anything else) to Mathematica is not difficult. Even without such a background, however, the learning curve is not that steep. As a versatile scientific and technical computing language Mathematica has been chosen for this volume in order to focus the reader's attention on concepts rather than programming. The programs are provided in the form of Mathematica statements and notebooks, which are easy to follow and adapt to a user's needs. Complete Mathematica notebooks together with plot statements are on the included CD-ROM in the hope that the reader will explore and extend the examples given, and invent new ones. Readers can download a free Mathematica Notebook Reader for the computer platform they use (MathReader at

22

CHAPTER 1.

INTRODUCTION

http://www.wolfram.com/products/mathreader/). In order to change the programs, however, a licensed version is necessary. This volume occasionally uses the Mathematica add-on package called Signals and Systems in order to facilitate some computations and generate a few of the plots. Signals and Systems requires a separate license. The earth changes the shape of the seismic wavelet as it propagates away from the source. For example, wavelet shape is changed by virtue of its encounter with certain kinds of subsurface velocity distributions, which reflect back differentiated or integrated versions of itself. Mathematica movies of these processes are included on the CD-ROM to aid the reader in understanding how this works. There is some advantage to writing Volume 1 in this Handbook of Geophysical Exploration for Seismic Exploration and at the same time being among the last to turn in the manuscript for publication. It gives us the opportunity to refer here to topics that are covered in greater detail in other volumes of this same Seismic Exploration series and we have attempted to point the reader to this more advanced material in this Series as well as elsewhere. The manuscript for this volume was prepared using the WT^/i [103] software system, which is based on Donald Knuth's TgX [98] language. Both were implemented using the PCTeX for Windows [88] software, Version 5.1. John K. Costain Cahit Qoruh Blacksburg, Virginia August 2004

1.1. ACKNOWLEDGMENTS

1.1

23

Acknowledgments

Many funding agencies supported our research over the decades including the National Science Foundation, the Nuclear Regulatory Commission, the U.S. Geological Survey, Virginia Division of Mineral Resources, Maine Geological Survey, Sohio, Chevron, Southeastern Exploration and Production Corporation (SEPCO), and many others. From 1979-88 we operated a full-time Virginia Tech vibroseis data acquisition crew. In 1981-83, Sohio was able to take advantage of any available crew time. Ditto for SEPCO and Chevron. Each industry collaboration gave us a little more insight into the local or regional geology. From 1982-84 we formed the Virginia Tech Vibroseis Consortium (VTVC) and thought up projects that would be of interest to us as well as to industry and governmental subscribers. Our non-vibroseis "Thin-Bed Consortium" focused on the theoretical aspects of thin bed detection and the seismic response expected from sequences of thin beds as well as the potential of high resolution seismic data to examine and reinterpret the depositional environment of coal. Some of the results of this research are included in this volume. Virginia Tech was the first university in the United States to install the then state-of-the-art DISCO (Digicon Interactive Seismic COmputer) processing software. Prior to this all of our seismic data processing was done on the university's IBM mainframe using homemade software written entirely in Fortran. This important turning point toward "icon-based" commercial reflection seismic data processing software has today evolved into the sophisticated processing and interpretation industry software now available at Virginia Tech. In the classroom, however, we also used the mainframe and Fortran so that our geophysics majors would be "computer literate", and not just "icon literate". We gratefully acknowledge the many contributions to geophysics of Enders Robinson and Sven Treitel. Their tutorial and insightful imprint on the literature has hopefully been successfully passed along by us to our students. Their influence should be evident in many parts of this volume. John Costain takes pride in revealing that, in the jungles of Venezuela with a Socony-Mobil of Venezuela seismic crew, he taught Sven Treitel everything he knows about cable splicing and repair. Indeed, John is certain that it was this thought-provoking challenge in the early 50s that planted the seeds of intuition that now blossom as Sven's productive excursions into neural networks and their applications to geophysics. (In the world we live in you don't have to be right, you just have to be certain.) Teaching undergraduate and graduate courses and doing research with students and faculty colleagues is one of the most intellectually stimulating environments around. We thank our students at the University of Utah (JKC: 1960-67), Virginia Tech (JKC: 1967-96 and CC: 1979-2004), Istanbul University (CC: 1976-83), and Istanbul Teknik University (CC: 1984-85) for all the discussions, for their enthusiasm, and for sharing their ideas. We are grateful to the many students who took our courses and it is to them that this volume is dedicated.

Chapter 2

Groundwork Complex numbers are needed to describe and understand many aspects of exploration seismology. The study of forced vibrations due to a dynamic system, the behavior of a linear system, and the processing of data in the time and frequency domains all involve the use of complex numbers. A short summary of complex numbers and complex variables is therefore considered to be groundwork for this volume.

2.1

Complex numbers

A complex number is a quantity given in the form z = x + iy where x and y are real numbers and "i" is defined as i 2

i

= v73! = -1

In this definition x is the real part of the complex number z and y is the imaginary part. Both x and y are real numbers. The notation for the real part Re[ ] and imaginary part Im[ ] of a complex number z is defined as x

=

Re [z]

y

=

Im [z]

respectively. In a Cartesian coordinate system, complex numbers can be plotted with the real part of the complex number on the x-axis (the abscissa) and the imaginary part of the complex number (the ordinate) on the y-axis. In this definition the x-axis is called the axis of reals and the j/-axis is called the axis of imaginaries. Such a two-dimensional space is known as the complex plane. 24

2.1. COMPLEX NUMBERS

25

Figure 2.1: Graphical representation of the complex number, P(x,y) = x + iy. Projection of OP onto x—axis is the real part of the complex number P(x, y). Projection of OP onto the y—axis is the imaginary part of the complex number. Distance OP = yjx2 + y2 is called the modulus of the complex number, i = \/—T. The vector defined by OP is called a phasor. Groundwork.cdr—> GraphOfComplexNumber.wmf

A point P is shown in Figure 2.1 in the complex plane as defined by the coordinates of x and y . The x—axis corresponds to the axis of real numbers. The y—axis corresponds to the axis of imaginary numbers. The value z of the complex number located at the point P is defined as z = x + iy Any number can be considered to be a complex number. For example, a real number may be defined as a complex number for which the imaginary part of the complex number is zero. Correspondingly, an imaginary number is a complex number whose real part is zero. Therefore, the complex number system is a more general definition of a number system than the real number system. There are two fundamental rules for the manipulation of complex numbers: 1. A complex number z = x + iy is equal to zero if x = 0 and y = 0. 2. Complex numbers obey the ordinary rules of algebra. Note that j 2 = — 1.

2.1.1

Manipulation of Complex Numbers

Using the above rules we give the rules for addition and multiplication of complex numbers. If zi = xi + i yi, and z-l.-iV3}}

as shown in Figure 2.3. When the phase 6 of the phasor is equal to zero, then z = r = 1. Therefore, in the general polar form with k an integer, I _ ei(0+2k*) _ ei(2fc7r)

and the root rxln can be given as ll/n

_ ei{2k*/n)

or l 1 / n = cos (2kn/n) + i sin {2kn/n) forfc= 0,1,2, • • • n - 1. From this result we see that the n roots of unity are located around the unit circle in the complex plane.

2.1.4

Logarithm of a Complex Number

In general if z = eu then UJ

= lnz

CHAPTER 2. GROUNDWORK

32

Figure 2.3: The three roots of 8 1 / 3 plotted on a circle of radius \z\ = 8 1/3 = 2 and spaced 120° apart, cuberoot.nb =>• cuberoot.cdr => cuberoot.wmf Using this definition properties of the logarithmic function may be given as follows. In(zi22)

=

ln( —) = n

ln(z )

In2i+lnz2 lnzi — In22

= nlnz

The logarithm of a complex number is another complex number, the real part of which is the logarithm of the modulus of z and the imaginary part is equal to the phase of z. If z = x + iy = r {cos 6 + ism 6) = rtie then lnz = ln{reie) = lnr + l n e i S = lnr + i