UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Structural characteristics of a subducting oceanic plate off Western Canada Waldron, David Anthony 1982

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1983_A6_7 W35.pdf [ 5.68MB ]
Metadata
JSON: 831-1.0052985.json
JSON-LD: 831-1.0052985-ld.json
RDF/XML (Pretty): 831-1.0052985-rdf.xml
RDF/JSON: 831-1.0052985-rdf.json
Turtle: 831-1.0052985-turtle.txt
N-Triples: 831-1.0052985-rdf-ntriples.txt
Original Record: 831-1.0052985-source.json
Full Text
831-1.0052985-fulltext.txt
Citation
831-1.0052985.ris

Full Text

STRUCTURAL CHARACTERISTICS OF A SUBDUCTING OCEANIC PLATE OFF WESTERN CANADA by DAVID ANTHONY WALDRON B.A. (honours Theoretical Physics), Cantab, 1980 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Department Of Geophysics And Astronomy We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA December 1982 © David Anthony Waldron, 1982 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of Geophysics and Astronomy. The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 DE-6 (3 /81) i i Abstract The plate tectonic regime off the southwest coast of British Columbia is convergent; the oceanic Juan de Fuca and Explorer plates are obliquely subducting beneath the continental America plate. To investigate the structural complexity of this region, the Vancouver Island Seismic Project was conducted in August 1980. The principal component of this project was a reversed refraction profile perpendicular to the continental margin, extending 350 km from the America plate to the deep ocean of the Juan de Fuca plate. This work deals with the marine section of the profile. Data from large explosive sources and an airgun were recorded on three ocean bottom seismographs (in the deep ocean, on the continental rise and on the shelf). Continuous seismic reflection profiles complemented the refraction information. To adequately model the seismic structure of this complex region required the application of ray-tracing procedures and a synthetic seismogram technique, based on asymptotic ray theory, for laterally varying media. Consistency between the seismic interpretation and previously published gravity anomaly variations across the margin was verified with the aid of empirical velocity-density relations. Additional seismic constraints were provided by multichannel reflection sections and sonic log data from a nearby well. The aim of a l l procedures for data modelling was to obtain velocity and density models which had the least structure consistent with a l l available geophysical information. The interpreted structural section indicates that the sediments thicken from 1 km in the deep ocean to 2 km at the base of the continental rise, where the eastward dip of the oceanic basement increases from 1.4° to about 3° and sediment velocities increase landwards. A str a t i f i e d upper crustal velocity sequence has been derived below the deep ocean which is similar to that deduced from other studies on the Juan de Fuca plate. This stratification is discordant with the descending oceanic plate further east; i t is replaced by a block of relatively low velocity material beneath the continental rise. This unit is interpreted to be a highly sheared and compressed melange. Lower crustal structure remains constant over the entire marine profile, suggesting that the melange material may have been formed as upper crustal layers were scraped off the descending oceanic plate. The lower crust has been modelled as a constant velocity gradient region, extending down to approximately 9 km below sea floor in the deep ocean. At this depth there is a strong decrease in velocity gradient, interpreted to be the Mohorovicic discontinuity. Resolution of Moho structure is poor, but no velocity discontinuity at the boundary is required by the data. There is an increase in the dip of the Moho under the continental rise from about 1° to 6 ° . The refraction data are thus explicable in terms of a relatively simple, two-dimensional velocity model, which is consistent with multichannel, well-log, and gravity information. i v Table of Contents Abstract i i List of tables vi List of figures v i i Acknowledgements ix CHAPTER 1 An Introduction 1.1 The Tectonic Regime 1 1.2 Subduction? Evidence for and against 3 1.3 Geology of the Margin 6 CHAPTER 2 On the Interpretation of Seismic Data 2.1 Marine Refraction Data Analysis 10 2.2 Travel Time Inversions 12 2.3 Synthetic Seismograms 16 2.4 Practical Interpretation 18 2.5 The Philosophy of Modelling 24 CHAPTER 3 The VISP-80 Data Set 3.1 The Vancouver Island Seismic Project 25 3.2 Data Description 27 3.3 Inherent Data Errors 35 3.4 Topography Corrections 39 3.5 Comments on Data Trends 41 CHAPTER 4 Data Interpretation - a detailed description 4.1 Summary of the Approach Taken 44 4.2 Travel Time Modelling 45 4.3 Amplitude Interpretation of OBS #1 Data 50 4.4 Practical Use of WKBJ and ART 57 4.5 Amplitude Characteristics for a l l Receivers .... 60 4.6 Construction of Final Models 62 4.7 Final Models - a detailed description 62 4.8 Interpretation Summary 75 CHAPTER 5 Gravity Interpretation 5.1 On the Analysis of Gravity Data 76 5.2 Gravity Data Description 78 5.3 Gravity Data Interpretation 80 CHAPTER 6 Summary and Discussion 86 Bibliography 92 Appendix A WKBJ and ART Algorithms . 99 Appendix B Instrumentation and Data Digitization 104 Appendix C Chevron Multichannel Data 105 Appendix D Shell Cygnet Well 110 Appendix E Structure of the Moho 111 Appendix F Bubble Pulse Oscillations ». 114 Appendix G 2-D Gravity Modelling Algorithm 116 Appendix H U.S.G.S. Multichannel Data 119 vi List of Tables 1.1 History of the continental margin 9 3.1 Inherent travel time errors 36 3.2 Instrument deployment times 37 5.1 Velocity vs. density in marine sediments 78 C.1 NMO velocities for Chevron data 107 E.1 The effects of Moho structure type ". 112 v i i List of Figures 1.1 Tectonic map of the margin 2 1.2 Geology of the margin 7 1.3 Ponded sediments on the rise 8 2.1 Travel time non-uniqueness 14 2.2 Transition zones - travel times 20 2.3 Transition zones - amplitudes 23 3.1 The Vancouver Island Seismic Experiment 26 3.2 CSP data 27 3.3 Raw ocean bottom seismograph data 29 3.4 Airgun data 32 3.5 Explosion data 33 3.6 Seafloor and basement topography 38 3.7 Amplitude variations: OBS #1, #3, and #5 42 4.1 A preliminary travel time model 46 4.2 Amplitude characteristics: OBS #1 52 4.3 The effect of Moho depth on amplitude 56 4.4 Comparison of WKBJ and ART algorithms 58 4.5 Amplitude trends and charge size 61 4.6 Airgun models 63 4.7 OBS #1 Travel time model 66 4.8 OBS #1 Amplitude model 67 4.9 OBS #3 Travel time model 70 4.10 OBS #3 Amplitude model 71 4.11 OBS #5 Travel time model 73 4.12 OBS #5 Amplitude model 74 4.13 Final velocity structure 75 v i i i 5.1 Location map of gravity profile 79 5.2 Gravity model 82 6.1 Final velocity and density structure 87 6.2 Geological structure of the upper crust 89 C.1 Chevron Standard multichannel data 105 C. 2 V-Z structure from NMO velocities 108 D. 1 V-Z structure from sonic log 110 E. 1 Possible velocity variations at the Moho 112 G.%1 Gravitational attraction of a polygon 116 H. 1 Location of U.S.G.S. multichannel section 119 H.2 U.S.G.S. data and geological interpretation 120 i x ACKNOWLEDGEMENTS Many friends and colleagues have helped me in the formulation and completion of this thesis, and in the enjoyment of my two years spent in Canada. Fir s t l y , I thank my parents and teachers in England, together with the Canadian Commonwealth Scholarship Committee, without whose help I would never have been able to come to Canada. Dr. R.M. Clowes, my supervisor, provided friendly advice, encouragement, and support throughout my studies. His help has been invaluable. I also wish to thank Dr. R.M. E l l i s for his help during Dr. Clowes' period of work in Costa Rica. Dr. G.K.C. Clarke and Dr. D.W. Oldenburg, by means of their contagious enthusiasm, have kept very much alive my interest in other areas of geophysics. I have always found the research environment in the Department of Geophysics and Astronomy at the University of British Columbia to be superb. I greatly appreciate the patient, knowledgeable advice of Ken Whittall in a l l matters seismological. I have also had numerous informative discussions with George Spence. Thanks are due to Ed Waddington for, among other things, the use of his drafting equipment; without which most of the diagrams in this thesis would be a mess. My 'army' of drafting help was formed of: Norman Avolino, Kerrie Blakeley, Mark Charnell, Maj DePoorter, Terence Sunderland and Gail Weitzman. Diane Donnelly became a source of amazement in the computer room for the speed and accuracy with which she typed this manuscript; I am very grateful to her for her fast fingers and quick wit. Invaluable editorial comments were provided by Dr. R.M. Clowes and Dr. G.K.C. Clarke; I greatly appreciate their patient and thorough readings of this work. The cooperation of the sci e n t i f i c staff at the Pacific Geoscience Centre on Vancouver Island was invaluable; without the use of their OBSs the data would not have existed. The Canadian Forces through the Defense Research Establishment Pacific provided the CFAV Endeavour, making the marine component of VISP-80 possible. The assistance of the Fleet Diving Unit, Pacific Maritime Command, who detonated the explosive charges at sea is appreciated. Personal financial support was provided by a Canadian Commonwealth Scholarship. Principal support for the study was provided by EMR research agreements (Earth Physics Branch) 214/3/81 and 42/3/82. Additional funding was provided by NSERC operating grant #A7707 and strategic grant #G0738. X " We shall not cease from exploration, And the end of our exploring, Will be to arrive where we started, And know the place for the f i r s t time."1 'T.S. E l l i o t , L i t t l e Gidding. 1 " What we c a l l the beginning is often the end And to make an end is to make a beginning. The end is where we start from1.". CHAPTER J_ AN INTRODUCTION 1.1 Plate Tectonics off the West Coast of British Columbia, Canada. In the Vancouver Island region the plate tectonic regime is dominated by the relative motions of three main lithospheric plates: the Pacific, America and much smaller, intervening, Juan de Fuca plate system (see figure 1.1). A small northern part of this system, the Explorer plate, has been shown to be moving independently (Riddihough, 1977). Based on magnetic anomaly patterns, the current perpendicular convergence rates are about 3 cm yr" 1 for the Juan de Fuca-American and less than 2 cm yr" 1 for the Explorer-American plates, thereby producing l e f t - l a t e r a l strike-slip movement along the Nootka fault zone at a rate of about 2 cm yr" 1 (Hyndman et a l . , 1979). Juan de Fuca ridge system consists of numerous en-echelon spreading axes, offset by short transform segments. Across this boundary the present f u l l spreading rate of motion ranges from about 4 cm yr" 1 to 6 cm y r " 1 . Tectonic evolution of the western Pacific margin has been discussed by Atwater (1970), whilst the geological implications of changes in plate configuration are discussed by Muller (1977). 1T.S. E l l i o t . 2 Figure 1 . 1 A tectonic map of the southern B r i t i s h Columbia continental margin. The main ll t h o s p h e r l c boundaries and plate motions r e l a t i v e to North America held stationary are shown. TWs » Tuzo Wilson knolls; PRfz = Paul Revere fracture zone; Sfz • Sovanco fracture zone. Dk • Oellwood kno l l s . In the past few tens of millions of years the tectonic regime of the western Canadian margin has changed rapidly, leaving a very complex geological record. It has been well established that subduction has occurred along the west coast of British Columbia, Washington and Oregon in the past few million years, but there has been debate as to whether i t continues at present, and i f i t does, whether i t is happening in an aseismic manner (Crosson, 1972; Riddihough and Hyndman, 1976; Keen and Hyndman, 1979; Ando and Balazs, 1979; Savage et a l . , 1981; Hyndman and Weichert, 1982; Riddihough et a l . , 1982). 3 1.2 Subduction ? Evidence for and against The continental margin off Vancouver Island shows a number'of features which are uncharacteristic of subduction zones around the world. There is no deep margin trench, nq deep earthquake suite in a Wadati-Benioff zone, or thrust fault-mechanisms beneath the continental slope and shelf; and there is limited activity in the Cascade volcanic chain. It is generally believed that because of the proposed very slow sinking rate (3 cm yr" 1) and the young, thin nature of the subducting Juan de Fuca plate , many of the phenomena normally associated with subduction will either be absent or not detectable (Keen and Hyndman, 1979). However there are a number of pieces of evidence supporting current subduction beneath Vancouver Island: 1) The existence of old ( >10 Ma) magnetic lineations west of the Juan de Fuca ridge axis, but not to the east. The required symmetry in sea-floor spreading implies that older parts of the plate have disappeared beneath the continent (Vine, 1966; Morgan, 1968). 2) Compressive deformation of sediments on the continental slope west of Vancouver Island postulated to result from underthrusting of the America plate by the Juan de Fuca plate (e.g. T i f f i n et a l . , 1972; Barr, 1974). The continental slope changes in character southwards from steep and narrow off the Queen Charlotte Islands and northern Vancouver Island to a much wider, i r r e g u l a r s l o p e o f f s o u t h e r n V a n c o u v e r I s l a n d , W a s h i n g t o n and O r e g o n ( f i g u r e 1 . 2 ) . T h i s seems t o i n d i c a t e a c h a n g e i n t h e i n t e r a c t i o n r e g i m e a l o n g t h e m a r g i n f r o m t r a n s f o r m f a u l t i n g i n t h e n o r t h t o c o m p r e s s i o n a n d s u b d u c t i o n i n t h e s o u t h . 3) A ' d o w n - t o - t h e - e a s t ' c r u s t a l t i l t r a t e a c r o s s w e s t e r n W a s h i n g t o n i s i m p l i e d by p r e c i s e l e v e l l i n g o v e r a 70 y e a r p e r i o d . Ando a n d B a l a z s ( 1 9 79) s u g g e s t t h a t t h i s t i l t r a t e i n d i c a t e s a s e i s m i c s l i p i n t h e J u a n de F u c a p l a t e s u b d u c t i o n z o n e : w h i c h may h a v e been c o n t i n u o u s f r o m 0.25 Ma (Adams and R e i l i n g e r , 1 9 8 0 ) . The l a c k o f any m a j o r t h r u s t e a r t h q u a k e s d u r i n g t h e p e r i o d o f t i d a l a n d l e v e l l i n g o b s e r v a t i o n s i s u s e d by Ando a n d B a l a z s a s e v i d e n c e f o r a s e i s m i c s u b d u c t i o n . I n t e r p r e t a t i o n o f t h e g e o d e t i c r e l e v e l l i n g d a t a i s s t i l l b e i n g d e b a t e d ; a r e v i e w i s g i v e n by R i d d i h o u g h ( 1 9 8 2 ) . 4) The N o o t k a f a u l t z o n e (Hyndman e t a l . , 1979; A u , 1981) i s a p p r o x i m a t e l y p e r p e n d i c u l a r t o t h e m a r g i n . L e f t - l a t e r a l s t r i k e - s l i p m o t i o n a l o n g t h i s f a u l t r e q u i r e s c o m p r e s s i o n a l o n g t h e m a r g i n t o t h e s o u t h o r s e a f l o o r m o v i n g away f r o m t h e m a r g i n t o t h e n o r t h . The l a t t e r p o s s i b i l i t y i s v e r y u n l i k e l y . 5) The g r a v i t y f i e l d shows t h e c h a r a c t e r i s t i c l o w a n d h i g h g r a v i t y b a n d s o f a c t i v e s u b d u c t i o n z o n e s , a l t h o u g h t h e t r e n c h i s much r e d u c e d by t h e c o m p l e t e s e d i m e n t i n f i l l (Riddihough, 1979). 6) The heat-flow pattern for southwestern British Columbia and for western Washington and Oregon exhibits the characteristic subduction zone pattern: a band of low heat flow extends from the trench to the volcanic arc about 200 km inland, and a much higher than normal heat flow extends for a considerable distance inland (Hyndman, 1976). 7) The oceanic basement is observed to dip under the continental rise, forming a sediment-filled trench; thereby implying subduction. 8) The andesitic volcanism of the Cascade mountains suggests the existence of a down-welling oceanic plate. Relatively recent volcanism implies that subduction has occurred and has continued up to at least 1 Ma ago (Riddihough and Hyndman, 1976). 9) There is a concentration of seismicity in the Puget Sound lowlands with a truncation of seismicity to the north at the 49th parallel (Rogers, 1982)- indicating some form of tectonic activity. The maximum depth of earthquakes in the region is about 60 km. This uncharacteristically shallow depth is thought to be due to the slow rate of convergence and the thin plate involved (Riddihough and Hyndman, 1976). 6 It i s very d i f f i c u l t to detect 'present' subduction. Each source of geophysical information has a d i f f erent time sca le . Offshore magnetic anomalies can give information back several tens of m i l l i o n years, but the ir time reso lut ion i s l i t t l e better than 1 Ma; earthquake data extend back only about 80 years (Rogers, 1982) and represent a very short , recent time i n t e r v a l . In summary, the boundary between the America and Juan de Fuca plates i s thought to be a zone of convergence or subduction, with underthrusting probably s t a r t i n g near the base of the cont inenta l slope with rates from 1 to 3 cm y r " 1 . 1.3 Geology of the Margin The cont inenta l margin off southern Vancouver Island cons i s t s of two d i s t i n c t regions; the Cascadia and Tofino basins , separated by the cont inenta l slope (f igure 1 . 2 ) . Cascadia Basin forms the Abyssal p l a i n , bounded to the east by the cont inenta l r i s e , and to the west by the Juan de Fuca r idge . The oceanic basement here dips to the east and is conformably over la in by a hemiterigeneous 1 sequence formed in l a te Miocene-Pliocene time (Scho l l , 1974). These beds are over la in by a landward-thickening t u r b i d i t e wedge, formed during the la s t 1.0 Ma (Carson, 1973). The cont inenta l slope has an average width of 70 km and 'A hemiterigeneous sequence is composed c h i e f l y of eros ional d e t r i t u s or p y r o c l a s t i c debr i s , derived from a nearby landmass, and there i s a l so a pelagic component. 7 F i g u r e 1.2 N e a i — s u r f a c e s t r u c t u r e s of the c o n t i n e n t a l margin, ( a f t e r T i f f i n e t a l . . 1972) The p o s i t i o n s of T o f i n o and C a s c a d l a B a s i n s a r e shown. Note the I n c r e a s e i n d e f o r m a t i o n southward on the c o n t i n e n t a l s h e l f . is composed of Quaternary and upper Tertiary strata which are deformed into a series of asymmetric folds and northeastward-dipping imbricate thrusts (Snavely and Wagner, 1981). The lower slope is composed of deformed and uplifted Cascadia Basin strata forming discontinuous ridges, in general fault-bounded, behind which sediments are ponded, thereby subduing the relief (Barr, 1974; figure 1.3). Tofino basin is a large Tertiary depositional area lying to the east of the continental slope. There are thick sequences of late Eocene to Pliocene mudstones containing abundant foraminifera ( T i f f i n et a l . , 1972) which indicate a bathyal depositional environment throughout most of the 8 F i g u r e 1.3 Ponded sediments on the c o n t i n e n t a l r i s e . A l i n e - d r a w i n g I n t e r p r e t a t i o n of p a r t of the CSP p r o f i l e ( f i g u r e 3.2) has been made. The dashed l i n e s a r e p r o b a b l e f a u l t s . Note the u n c o n f o r m i t y s u r f a c e between the eastward f l a n k of the r i d g e and the ponded sediments. Tertiary. Subsequent uplift has exposed the deep water sediments on the shelf over much of the area. Eocene-Oligocene sediments occur in a belt along the inner shelf, while Miocene and Pliocene rocks l i e seaward of this. Snavely and Wagner (1981) propose the existence of a mass of highly sheared and compressed melange' lying beneath the sediments of the continental rise. They name this unit the "Hoh Melange" (middle Miocene), which is thought to have been 'Definition of a melange: A heterogeneous medley or mixture of rock materials. A mappable body of deformed rocks consisting of a pervasively sheared, fine-grained, commonly p e l i t i c matrix, thoroughly mixed with angular and poorly sorted inclusions of native and exotic tectonic fragments, blocks or slabs. 9 uplifted prior to deposition of the continental slope Pliocene strata. Comprehensive reviews of the geology of the continental margin may be found in T i f f i n et a l . (1972), or Chase et a l . (1975). Shouldice (1971) has compiled a depositional history of the continental margin, which is summarized below. Geologic Time Depositional environment Recent Pleistocene Pliocene Regression in upper Pliocene Lesser one in lower Pliocene Miocene Major transgression in upper Miocene. Period of regression, crustal deformation, and u p l i f t . Oligocene Transgressions of Oligocene and lower Miocene seas Eocene Widespread lower - middle Eocene submarine volcanic activity. I n i t i a l uplift followed by subsidence'in late Eocene. > Cretaceous Mesozoic sediments. Intrusive and extrusive igneous features. Table 1.1 History of the continental margin. 10 " It requires a very unusual mind to undertake analysis of the obvious 1." CHAPTER 2 ON THE INTERPRETATION OF SEISMIC DATA 2.1 Marine refraction data analysis - general comments. In the analysis of seismic refraction data there are three groups of factors, which by their nature prevent a complete, accurate inversion: 1) Errors associated with the data 2) Real earth violation of simplifying assumptions 3) Limits on the resolution of velocity structures 1) The exact positions of source and receiver are not known in marine work. The magnitude of the distance errors varies with the type of receiver used ( c f . a fixed OBS and a drifting sonobuoy). In general, a l l shot locations and receiver positions are determined from near shore navigation systems supplemented by sa t e l l i t e checks. For studies off the west coast of Canada, the relative position accuracy is then 200 m with greater uncertainties in absolute position (Hyndman et a l . , 1979). There are intrinsic timing errors associated with the receiver clock and shot detonation time, as well as travel time picking errors of f i r s t arrivals. Due to the lengthy nature of the marine seismic source wavelet (Appendix F), the picking of secondary arrivals is usually d i f f i c u l t and often A l f r e d North Whitehead 11 impossible. Thus certain sources of definitive information may be lost entirely, e.g. sub-critical reflections. 2) In a l l modelling of seismic data, simplifying assumptions restrict the range of possible earth structures and enable the use of standard analysis techniques. In most cases the simplification is to assume that velocity varies only with depth, i.e. a one-dimensional structure. The real earth is often highly variable; even at flat-lying, seemingly uncomplicated sites, heterogeneity can be inferred on a scale of kilometres (Spudich and Orcutt, 1980). Shot-to-shot random variations in amplitude give evidence of lateral variations in structure (e.g. Fowler and Keen, 1978). Even when two-dimensional models are considered, these can only be a very simple approximation to the real earth. 3) A l l seismic waves have f i n i t e , non-zero wavelength X; turning rays are therefore unable to resolve structural details which are smaller than about 1X. The acoustic properties of a heterogeneous medium are averaged when a seismic wave passes through and, consequently, different earth structures having the same velocity profile when averaged over X will be indistinguishable when viewed by seismic waves with wavelength X. Typically, the wavelengths used to study the crust range from 0.5 to 2 km for explosive sources, 0.1 to 2 km for airguns and 0.5 m for downhole velocity-logging devices. In the majority of crustal material, velocity (and therefore wavelength) increases with depth. Thus, the maximum possible resolution that may be obtained decreases with depth. 1 2 The other intrinsic limit on the resolving power of the data is the spatial separation of the sources and/or receivers employed. The discrete nature of the information obtained makes identification of cross-over points between two branches in the travel time-distance graph accurate at best to the value of the separation between traces. Subtle changes in amplitude with distance may not be identified with confidence unless a reasonably dense sampling interval is obtained. 2.2 Travel-Time Inversions An important simplifying assumption is usually made: that the velocity c varies only with depth z i.e. one-dimensional structure. With this restriction on possible models, the Herglotz-Wiechert formulae may be derived (Herglotz, 1907; Wiechert, 1904): T(p) = 2 f ^ c C z ) ) - 2 d z o J [ ( c ( z ) ) " 2 - p 2 ] " 2 These expressions are a formulation of the forward problem: for a given model c(z), the data T(X) may be derived. There is a construction method for the inverse problem: Abel's solution. X(p) [ ( c ( z ) ) - 2 - p 2] where P z(p) ( P ray parameter bottoming depth dT/dX ) 13 It may be shown that: (see Aki and Richards, 1980). z(c) = -1 ^T] 1/2 where c 0 = velocity at z = 0 This is an exact solution of the inverse problem: given data X(p) we may derive the velocity-depth structure z(c). It would seem that one-dimensional travel time inversion, using the above formula, is a straightforward, purely mechanical procedure. In practice, the Abel inverse solution may not be used directly as the 'real' data are discrete, noisy, and band-limited. Even given a hypothetical situation where there were 'perfect' data, i.e. continuous, accurate and full-band; i f only f i r s t arrivals are picked there might s t i l l be problems. Regions of rapid velocity increase with depth lead to triplications in the T(X) curve. In order to derive a continuous X(p) curve from the data both the normal and receding branches must be correctly identified, i.e. some later phases must be picked. An example of the non-uniqueness of solutions in this type of structure is given by Healey (1963); the data and possible models are shown in figure 2.1. Depth uncertainties from seismic f i r s t arrival travel times are discussed by Berry (1971). In order to cope with the noisy and discrete nature of the data, travel time inversion techniques employ weighting and interpolation schemes. The' reparameterization of travel time data, T(X), into the delay time function - ray parameter domain, r(p) (where r(p) = T(p) - pX(p)), is frequently F i g u r e 2.1 The non-uniqueness o f f i r s t a r r i v a l t r a v e l t i m e I n t e r p r e t a t i o n s ( a f t e r H e a l y . 1963). a) T r a v e l t i m e o f t h e f i r s t a r r i v a l Is known as shown h e r e , but th o s e of l a t e r a r r i v a l s a r e unknown. b) V e l o c i t y d i s t r i b u t i o n s t h a t g i v e the same f i r s t - a r r i v a l t i m e s as shown i n ( a ) : n i s t h e number of l a y e r s . performed (e.g. Kennett, 1976; Bates and Kanasewich, 1976). Use of this function is advantageous since it is single-valued and a monotonically decreasing function of p (McMechan and Wiggins, 1972). Also, i t has been noted (Johnson and Gilbert, 1972) that f i r s t order errors in p only propagate to second order errors in r, which is important for the construction of r(p) from the data, T(X). Three one-dimensional travel time inversion schemes are summarized below: 1) Nonlinear extremal bounds (Bessonova et a l . , 1974). 'Averaging' operators are passed along the r(p) bounds to obtain upper and lower bounds on possible velocity depth distributions. The derived extremal bounds are very broad (Spudich and Orcutt, 1980; Au, 1981) and consequently are virtually useless for obtaining results of geophysical interest. Greater constraints may be placed on the extremal bounds with the addition of X(p) data (Orcutt et a l . , 1980; and Jurkevics et a l . , 1980), but as the aim in most studies is to obtain a starting model for synthetic seismogram calculations, the small improvement with. the X(p) data addition is not significant. 2) Linearized inversion (Johnson and Gilbert, 1972). In this method an iterative improvement is made to a t r i a l velocity-depth profile until the calculated r(p) values l i e within the error bounds for the observed r. The final model has the tendency to minimize velocity gradients. This approach is summarized by Kennett and Orcutt, 1976; Kennett, 1976; and Cheung, 1978. 3) Linear programming inversion (Garmany et a l . , 1979). For this procedure, classes of extremal solutions are considered for which the depth to a given velocity is maximized or minimized subject to a f i n i t e number of data constraints. A l l models within the derived bounds do not satisfy the data, but any velocity within or on these bounds is realizable by at least one of the infinity of possible solutions, i.e. no v-z points can be discarded as unrealistic. The main advantage of this technique is that computation is very fast and 1 6 economical and a starting model for amplitude studies is easily obtained. This method is summarized by Au (1981). 2.3 Synthetic Seismograms The amplitude of a seismic waveform is a potential source of information regarding earth structure through which the wave has passed. Amplitude interpretation is usually a t r i a l -and-error, forward modelling procedure in which synthetic seismograms are subjectively compared to observed records. Work on the direct inversion of seismograms for velocity structure is currently being undertaken by C.H. Chapman and J. Orcutt. Clayton and McMechan (1981) have developed a method which does not directly u t i l i z e seismic wave amplitudes, but in which the whole data wave f i e l d is linearly transformed from the time - distance domain into the slowness - depth domain, where the velocity profile can be picked directly. In formulating synthetic seismogram algorithms, assumptions are always made to simplify computation. The most common assumption made is that the earth is laterally homogeneous. It is necessary to evaluate an expression of the form: t transform r transform Bessel function transform of the source time function 1 7 There are a number of methods available for computing a synthetic seismogram, given a one-dimensional velocity and density structure for an earth model: a) WKBJ method (Chapman, 1978; Appendix A) b) Reflectivity method (Fuchs and Muller, 1971) c) Generalized Ray Theory (GRT): (Wiggins and Helmberger, 1974) d) Full wave theory (FWT) (Richards, 1973; Choy, 1977) e) Disc-ray theory (DRT) (Wiggins, 1976) f) Asymptotic Ray Theory (ART) (Krebes and Hron,1980) These methods differ primarily in: i) the degree of approximation used in deriving F, i i ) the order in which the integrals over k and s are performed, i i i ) the path of integration, T. There is always a trade-off between accuracy and expense. Extensive reviews of the various methods can be found in Chapman (1978) and Richards (1979), while a comparison of the practical application of some methods is given by Spudich and Orcutt (1980). There are two inherent d i f f i c u l t i e s involved in the current use of synthetic seismograms for forward modelling purposes: instability and non-uniqueness. Instability The relation of the seismograms to the velocity models is non-linear and unstable, since small changes in the t r i a l velocity models sometimes produce large changes in the resulting synthetic seismogram. 18 Instability Model Space Data Space Non-Uniqueness Model Space Data Space Non-uniqueness The trial-and-error reliable error estimates on produces. The solutions are different models may produce appraisal is not possible modelling approach modelling technique yields no the velocity models which i t inherently non-unique: two widely the same theoretical data. True with the trial-and-error forward 2.4 Practical Interpretation 2.4.1 Traveltime interpretation The inability of discrete, noisy, band-limited travel time data to resolve details of the form of a velocity-depth distribution is well known (see for example Spudich and Orcutt, 1980). A clear illustration of the wide range of models which may be found to f i t any given set of real travel time data is the computation of extremal bounds (see section 19 2.2). These limits are always found to be very broad, due to the relatively large errors inherent in the data. The practical resolution that may be obtained with travel time data is illustrated in figure 2.2. The solid lines forming a triplication (a to d) represent the travel times through the mean oceanic crust ( MOC ) model of Raitt (1963, upper l e f t ) . Synthetic travel time data have been calculated from gradient models which provide various rates of transition from the 6.69 to 8.1 km/s velocities of Raitt's model. It can be clearly seen from figure 2.2 that the most diagnostic feature of a rapid transition zone is the presence of secondary arrivals in the tr i p l i c a t i o n region. A discrete boundary is only distinguishable from a 1 km thick transition region by the presence of sub-critical reflections (produced by the boundary, figure 2.2a), and from a 4 km transition zone by the presence of wide-angle reflections (to within 0.1 s error for mantle arrivals, figure 2.2c). It is also clear that if the travel time errors are large ( >0.15 s) at offset distances greater than 50 km (perhaps due to complex structure), then an 8 km transition zone model is consistent with the f i r s t arrival travel time data (figure 2.2d). Thus, the diagnostic characteristics of the travel time data are usually the secondary arrivals which, as has been noted (section 2.1, and appendix F ) are often d i f f i c u l t or impossible to identify. 20 0 20 40 60 80 X (km) X (km) X (km) Figure 2.2 Transition zone c h a r a c t e r i s t i c s and travel times. Upper l e f t : The s o l i d line is the mean oceanic crust model (M0C) of Raitt (1963). A s l i g h t gradient has been introduced into the o r i g i n a l constant v e l o c i t y regions to generate turning rays there. The dashed l i n e s , a, b. c. and d represent Increasingly more gradual t r a n s i t i o n s from 6.69 km/s to 8.1 km/s. In the four diagrams, a. b. c. and d: the crosses represent travel times through the t r a n s i t i o n zone v e l o c i t y model as compared to travel times through the MOC model - shown by the s o l i d l i n e . a) 1 km thick t r a n s i t i o n zone (TZ), b) 2 km TZ. c) 4 km TZ. d) S km TZ. 21 2.4.2 Amplitude Interpretation The observed amplitude of a seismic waveform depends upon three groups of factors: i) characteristics of the source i i ) properties of the earth i i i ) instrument response The influence of earth structure on seismic amplitude must be extracted from other effects in order for interpretation to proceed. In marine refraction seismology a given data set usually consists of many sources recorded on a single receiver; the instrument response is then a constant factor. If a repetitive source is used (such as an airgun), the source function is assumed constant, enabling reliable comparison of relative trace amplitudes for a given profile. The use of explosion sources with variable charge sizes yields much more ambiguous amplitude information. Variations in trace character may then be due either to changes in sampled earth structure or the variable source characteristics. Attempts are usually made to compensate for the variation in charge size before relative amplitude modelling with explosion data. Empirical formulae relating charge size to amplitude of seismic energy generated are generally used (O'Brien, 1960; Kanestr0m and 0vreb0, 1978). The d i f f i c u l t y of reliably estimating these formulae is evident in this work (section 4.3) and has been noted elsewhere (e.g. Clee et a l . , 1974). Relative amplitude modelling of explosion data should ideally be based upon sources of equal charge size, and i f 22 'corrected' amplitudes are used, these should be interpreted with caution. The proposed seismic model is invariably (by the nature of the data) a crude approximation to the real earth. Thus there may be significant unmodelled horizontal variation in structure, producing trace-to-trace amplitude variations. These should not be modelled with detailed vertical velocity structure, but traces of similar character must be grouped together and the general amplitude features modelled. Thus the more spatially dense the data are, the better the resolution which may be obtained with amplitude data. The amplitude characteristics of gradient region variations on the mean oceanic crustal model of Raitt are shown in figure 2.3. The responses have been calculated using the WKBJ synthetic seismogram algorithm due to Chapman (1978). That the secondary arrivals are a very diagnostic feature of the data in determining the form of velocity variation is clear from the figure. If the amplitude characteristics are clear, and the spatial density reasonably high, models which are indistinguishable on the basis of f i r s t arrival travel times become resolvable. The width of the main amplitude group (associated with wide-angle reflections) is greater the sharper the transition. The synthetic seismograms (figure 2.3) illustrate another, potentially misleading, effect -interference. There is relatively l i t t l e amplitude between 40 and 50 km (figure 2.3c) due to the destructive interference between f i r s t and secondary arrivals. This occurs with the 23 X (km) X (km) X (km) F i g u r e 2.3 T r a n s i t i o n zone c h a r a c t e r i s t i c s and a m p l i t u d e s . Upper l e f t : WKBJ s y n t h e t i c set sinogram f o r the mean o c e a n i c c r u s t model (MOC) of R a i t t (1963). S o l i d l i n e l i n k s f i r s t a r r i v a l t r a v e l time p o i n t s , a. b. c. and d a r e WKBJ s y n t h e t i c seismograms f o r the f o u r t r a n s i t i o n zone models d e s c r i b e d In f i g u r e 2.2. M i d d l e l e f t : A comparison of the r e l a t i v e l y s h o r t s y n t h e t i c wavelet used In the WKBJ s i m u l a t i o n s w i t h a t y p i c a l wavelet r e c e i v e d by an OBS (ocean bottom seismograph) from a marine e x p l o s i o n . 24 simple two lobe synthetic wavelet used, which is «/»().35 s in length. True marine data may have -a characteristic wavelet w i.O s in length (figure 2.3; Appendix F), and thus there will be many complex interference effects between the wavelets arriving at any given point but being associated with different events. Anomalously low amplitude traces in the middle of large amplitude groups should generally be ignored as they may be the result of destructive interference, or be due to unmodelled lateral heterogeneities. Traces of this nature could equally well occur at the edges of a large amplitude group, where they would not be considered anomalously low; the amplitude group would merely be regarded as being less spatially extensive. Thus the lateral extent of large amplitude regions has this inherent uncertainty; an uncertainty which reduces the practical resolution that may be obtained between different earth models. 2.5 The Philosophy of Modelling It is relatively easy to produce a complex velocity-depth model which is consistent with the data. Whilst i t is possible that this model may represent the real earth, unless a l l the features are constrained by the data (or additional geophysical information), the complexity is misleading. Geological implications may be drawn from details in the velocity-depth model which could be eliminated and yet s t i l l achieve real data consistency. The aim in a l l seismic modelling should be to produce the v-z model which has the least structure compatible with a l l the data. 25 " The greatest tragedy of science - the slaying of a beautiful hypothesis by an ugly f a c t 1 . " CHAPTER 3 THE VISP-80 DATA SET 3.1 The Vancouver Island Seismic Project In August 1980, CO-CRUST2 conducted the Vancouver Island Seismic Project (VISP-80) which was a series of large scale refraction and reflection seismic experiments, u t i l i z i n g both land-based and ocean bottom seismographs (OBSs). The refraction program consisted of 4 profiles, lines I to IV (figure 3.1). Three of these were along strike, one on Vancouver Island, one on the continental shelf, and one in the deep ocean; and the fourth, across strike, extended from the deep ocean to the inland volcanic arc. The interpretation of the oceanic portion of this onshore-offshore profile is presented here. To provide shallow information, continuous seismic profiling (CSP) was run with a 5 l i t r e airgun. For successively deeper information a 32 l i t r e airgun and explosive sources were recorded on four OBSs (1, 3, 5 and 6-2, figure 3.1). The 37 charges used on the 100 km long explosion profile ranged in size from 50 to 825 kg and were a l l Thomas H. Huxley 2CO-CRUST (Consortium for Crustal Reconnaissance Using Seismic Techniques), in 1980, included participants from Earth Physics Branch (Ottawa), Pacific Geoscience Centre, Atlantic Geoscience Centre, and the Universities of Alberta, British Columbia, Manitoba, Saskatchewan, Toronto, and Western Ontario. F i g u r e 3.1 The Vancouver I s l a n d S e i s m i c P r o j e c t . T h i s l o c a t i o n map shows the r e f r a c t i o n l i n e s ( I - I V ) . the r e f l e c t i o n l i n e ( R L ) . and the a i r g u n and CSP l i n e s . B a t h y m e t r y i s i n meters. I n t e r p r e t a t i o n of the marine p o r t i o n of l i n e I i s p r e s e n t e d i n t h i s work . 27 detonated at a depth near the optimum for maximum seismic energy generation (Shor, 1963). F i g u r e 3.2 Continuous s e i s m i c p r o f i l i n g d a t a a l o n g p r o f i l e I R e f l e c t i o n s from the boundary between the l a t e P l i o c e n e and Q u a t e r n a r y sediments a r e i n d i c a t e d by 'a' and from the basement by 'b'. Note t h a t the upper t i m i n g l i n e i s 2.41 s two way t r a v e l time, and t h a t OBS *3 i s l o c a t e d 9 km o f f the e a s t e r n end of the s e c t i o n . 3.2 Data Description 3.2.1 CSP Record Continuous seismic profiling was carried out along the profile for a distance of almost 70 km west of the continental rise (figure 3.2). The section shows the basementb', to be flat-lying on the westerly 8 km of the profile and then dipping at 1.4° towards the continent. The CSP over OBS1, but 28 perpendicular to line I, shows the basement to be faulted with raised horst structures close to the OBS position. The reflector 'a', ^ 0 . 4 s above the basement in figure 3.2 is thought to be an unconformity between the late Pliocene and Quaternary. The moderately deformed sediments have a dip at depth similar to the basement but become progressively more flat-lying towards the surface, and are terminated abruptly at the continental rise. 3.2.2 OBS Characteristics The ocean bottom seismographs (OBSs) used in the VISP-80 project are described in Appendix B. Signals from two horizontal and one vertical 4.5 Hz geophones, plus an internal clock are recorded on analog magnetic tape. An example of the data from a l l four channels of the instrument is given in figure 3.3. There is cross-talk betveen the three data channels and the 20 Hz time code channel•which is evidenced by the sharp peak in power close to 25 Hz (seen on the power spectrum of the noise for the vertical channel). The presence of cross-talk interference on the time channel (e.g. near the end of trace TC, figure 3.3) led to d i f f i c u l t i e s with the automatic reading of the time code via a computer algorithm. The dominant frequency in the vertical channel seismic energy packet, shown by data in figure 3.2, is approximately 6 Hz. When bandpass f i l t e r i n g was applied to the data the limits used were 3 - 1 5 Hz. The kink in signal waveform at near-zero amplitude level is due to square root signal compression by the amplifiers, a feature which increases the dynamic range of NOISE DATA <U ic o a, o > o Id a o OL, > o id o. °0 25 50 Frequency (Hz) 2s. NOISE V HI Time: °0 25 50 Frequency (Hz) 2s. DATA 1 H2 Figure 3.3 An example of raw ocean bottom seismograph data. The four traces shown were recorded on OBS 05, at a source-receiver distance of 47 km. V - output from the v e r t i c a l seismograph; HI and H2 -output from the two horizontal seismographs; TC - 20 Hz time code. a) Power spectrum for 2 seconds of 'noise' on the v e r t i c a l channel. V. as shown. The large peak close to 25 Hz is due to cross talk from the t ime channel. b) Power spectrum for 2 seconds of 'data' on channel V. as shown. The dominant frequency is 6 Hz. 30 the instrument but gives poor signal reproduction f i d e l i t y when squared after digitization to recover the relative amplitude information. 3.2.3 OBS Data Presentation The six record sections constituting the seismic data set modelled in this study are shown in figures 3.4 and 3.5. The trace amplitudes on a l l sections are scaled with an r 2 spreading factor to enhance weak arrivals at greater distances. A l l the airgun data (figure 3.4) have been bandpassed with a 3 to 15 Hz eight pole, zero phase, Butterworth f i l t e r . The explosion data presented in figure 3.5 are unfiltered, but in order to compensate for the different charge sizes used, an amplitude normalization factor of weight2'3 is applied. This expression is based on empirical results relating the size of the charge to the amplitude of seismic energy which i t generates (O'Brien, 1960; Kanestr0m and 0vreb0, 1978). The validity of this normalization, as applied to the data presented here, is discussed in section 4.3. On a l l sections the OBS clock dr i f t has been removed using the assumption of linearity (section 3.3.1), and a l l source depths have been corrected; placing sources on the sea surface. First arrival travel time picks are indicated by arrowheads. 31 3.2.4 OBS #1: Airgun Data First arrival picks can readily be made to almost 20 km offset on this section (figure 3.4). The velocity indicated along the entire length of the profile is close to 6.0 km/s, and there are no clearly defined major amplitude features. The steady build-up in energy with offset seen in the distant 10 km of the profile is merely an artifact of the r 2 spreading factor. A linear distance enhancement factor would give a more real i s t i c appearance to the section. The large amplitude features seen in the upper left of this and the other airgun sections, are the direct water wave arrivals. 3.2.5 OBS #3: Airgun Data First arrivals are clearly visible out to almost 10 km offset (figure 3.4). The absence of appreciable energy further out is due to the smaller airgun used (16 l i t r e as compared to 32 l i t r e for OBS #1), and also the relatively high attenuation expected in material below OBS #3. The less homogeneous nature of this medium is implied by the trace-to-trace fluctuations in f i r s t arrival travel times and amplitudes. The refraction arrival velocity increases from 2.0 to 3.0 km/s with offset. 3.2.6 OBS #5: Airgun Data First arrivals are visible out to about 12 km offset. High attenuation, and the smaller airgun as mentioned above, are expected to account for the lack of energy at greater offsets. The velocities indicated are similar to those for the OBS #3 airgun section. 32 4 6 8 10 12 14 16 18 20 DISTANCE (km) DISTANCE (km) O H 1 1 1 1 1 1 r 0 2 4 6 8 10 12 14 16 DISTANCE (km) Figure 3.4 Airgun data. Record sections for OBS <M. *3. and *5 are shown. A 32 l i t r e gun was used for OBS #1. whilst a 16 l i t r e source was used for OBSs *3 and #5. A l l the data have been bandpass f i l t e r e d (3-15 Hz) and an r* spreading factor applied to enhance distant a r r i v a l s . F i r s t a r r i v a l travel time picks are indicated by arrowheads. 33 i ^ 1 1 1 1 1 1 1 1 0 10 20 30 40 50 60 70 80 DISTANCE (km) o-i 1 1 -i 1 1 1 i 1 1 "i 0 10 20 30 40 50 60 70 80 90 100 DISTANCE (km) Figure 3.5 Explosion data. Record sections for 03S <H. *3. and *5 are shown. The explosives were detonated on a 100 km l i n e along p r o f i l e I (figure 3.1). A l l the data have been bandpass f i l t e r e d (3-15 Hz) and an r ' spreading factor applied to enhance distant a r r i v a l s . An empirical charge size co r r e c t i o n of W" (where W Is the charge weight in kg), has been used. The f i r s t a r r i v a l travel time picks are indicated by arrowheads. M i s f i r e s account for the two large trace separations on the sections. On the OBS *1 section, times are adjusted to place the shots at the 2.52 km depth of the OBS: for OBS *3 and #5. the travel time e f f e c t of the water column is retained. 34 3.2.7 OBS #1: Explosion Data A l l shots on this section have been corrected to a datum depth of 2.52 km (the depth of OBS #1), by ray tracing through the water column (see figure 3.6). There are clear travel time picks out to about 50 km offset, past which l i t t l e signal energy was received. The apparent refraction arrival velocity increases from 5.7 km/s to 8.0 km/s over the profile. There is a group of traces with increased amplitude between 12 and 18 km offset; the trace at 15 km is comparatively low in amplitude due to the over-compensation of the charge size weighting factor which was applied (section 4.3). The main amplitude group seen between 28 and 45 km may be slightly broader in extent, but is undefined at the edges due to the misfiring of charges at 26 and 48 km offset. The retardation of the main energy group at 51 km, relative to the closer arrivals, suggests the presence of some lateral heterogeneities at the crust-mantle boundary. The oscillatory trace-to-trace variation in apparent noise level is at least partly an artifact of the charge size correction. There is a small d.c. bias on a l l traces, which becomes increasingly more apparent at large offsets, due "to the effect of the r 2 enhancement factor. 3.2.8 OBS #3: Explosion Data There is clear signal energy seen out to 95 km on the section. Reasonably accurate picks (error < 0.6 s) can be made out to 60 km offset. The apparent refraction arrival velocity increases from 2.8 km/s to 8.2 km/s over the profile. The only 35 well defined large amplitude group li e s between 33 and 46 km. The oscillatory noise level and d.c. bias mentioned above are clearly visible on this section also. 3.2.9 OBS #5: Explosion Data First arrival energy can be seen out to 80 km on this profile. Accurate picks (error < 0.06 s) have been made out to 67 km offset, with less accurate picks beyond this distance. The apparent refraction velocity increases from 4.5 km/s to 10 km/s over the profile. There are amplitude variations over the nearer part of the section, but no clearly defined groups. The main amplitude group between 47 and 58 km lacks clear definition due to misfire of the shot at 50 km offset. 3.3 Inherent Data Errors 3.3.1 Travel Time Errors There are four sources of travel time errors. 1) Origin time calculation 2) Digital sampling rate fluctuations 3) OBS internal clock d r i f t 4) Picking Error The estimated magnitude of these errors is summarized in table 3.1. Cause Maximum estimated error origin time calculation ± 0.03 s Digital sampling rate fluctuations < 0.003 s OBS internal clock d r i f t < 0.001 s picking error 0 (0.1) s Table 3.1 Inherent travel time errors. 1) The shot origin time may be estimated from the shot depth d the distance from ship to shot x and the water velocity Vw . The distance x must in turn, be estimated from the ditch time ts and the average ship speed over that time interval Vs. The origin time, t 0 , is then given by: t 0 = t, - 7 d 2 + x 2 Vw where t} - the arrival time of the direct water wave (recorded on a hydrophone towed immediately behind the ship). The shot depth may be estimated in 2 ways: the reflection method or the bubble pulse method (see Horn (1982) for discussion). The maximum probable error is generally ± 30 m. This value, in addition to errors in ts, Vw, Vs, and t 1 f gives a maximum estimated error in the order of ± 0.03 s. 37 2) Sample rate fluctuations in the digitization process (see Appendix B> lead to timing errors of less than 0.003 s. 3) The internal OBS clocks do not maintain their accuracy during the period of deployment; there is clock dri f t relative to the WWVB standard time signal. Comparisons between OBS clock times and WWVB time were made immediately prior to instrument release and after recovery. A linear regression through the differences between the two times provides a d r i f t curve for each OBS clock, which was then used to calculate d r i f t corrections for each shot. The instrument d r i f t rates are given in Table 3.2. Instrument Down Time Total Drift Drift/Hour OBS #1 OBS #3 OBS #5 10 days 11 days 12 days +2.461 s -0.779 s -2.538 s +0.00958 s -0.00275 s -0.00857 s Table 3.2 Instrument deployment times. 4) The picking error varies with signal/noise ratio, but is generally between 0.5 and 0.1 s. 38 DISTANCE (km) Figure 3.6 Seafloor and basement topography. Lower r i g h t : The seafloor topography, obtained from echo-soundings. and the basement topography. Interpreted from the CSP record (figure 3.2). V e r t i c a l exaggeration Is 14:1. Positions of the three OBSs are Indicated by Inverted t r i a n g l e s . Note that there Is no clear Indication of the basement topography east of the continental r i s e . Upper r i g h t : The l a t e r a l extent of the CSP and explosion p r o f i l e s , together with a 1:1 representation of the topography v a r i a t i o n . L e f t : A standard sound v e l o c i t y p r o f i l e for the water column, together with the chosen v e l o c i t i e s for the sub-surface material. 3.3.2 Distance Errors A l l shot locations and OBS positions are determined from Loran C Navigation, supplemented by regular s a t e l l i t e fixes. Relative accuracy of these positions is about 200 m with an 39 absolute accuracy of 300 m. Shot-receiver distances determined in this fashion have higher accuracy than those calculated using direct water wave travel times (Au, 1981). 3.4 Topography Corrections In many marine refraction studies (e.g. Horn, 1982; and Au, 1981), attempts are made to remove the effects of sediment and basement relief in order to simulate one-dimensional structure. Two different approaches are commonly chosen to accomplish these corrections (see discussion in Au, 1981). The degree of seafloor relief along the profile considered in this study is variable (see figure 3.6): high on the continental slope to the east but subdued on the abyssal plain. The form of the sediment-basement interface is determined from CSP records (see figure 3.2) for . the westernmost 50 km of the profile. No profiling was conducted east of the continental rise in the VISP-80 project but other CSP records in the area (Barr, 1974) do not delineate the interface east of the rise. Due to the lateral variation in magnitude of the rel i e f , different approaches to corrections for the topography were used for the individual data sets. The airgun profile recorded on OBS #1 (figure 3.4) was conducted over relatively subdued sediment and basement relief which was accurately determined from 3.5 kHz depth sounding and CSP data. A l l shots were corrected to an equivalent source depth of 2.52 km (the depth of the OBS) by ray tracing through the water column. The sound velocity profile of the water 40 column (figure 3.6) is based upon a compilation of the National Oceanographic Data Center in Washington, D.C. The OBS depth was chosen as an upper sediment datum and an averaged basement depth as the lower datum. Corrections for the variation in depth of the water-sediment and sediment-basement interfaces were then obtained by appropriately replacing one type of material with another above or below the reference datum. The magnitude of the resulting time and distance corrections is relatively insensitive to the chosen ray parameter but extremely sensitive to sediment and basement velocities used (Detrick and Purdy, 1980). This approach was used in the present study as the sediment velocities are constrained by additional information (see Appendices C and D). The near 50 km of the explosion profile recorded on OBS #1 passed over almost flat seafloor and an eastwardly dipping sediment-basement interface (see figure 3.2). For the purposes of one-dimensional velocity inversion (see section 2.2 ), the shots were corrected to the sloping basement, effectively removing the sedimentary wedge. In subsequent two-dimensional modelling (section 4.2.2) the sediments were recombined, as interpreted from the CSP, and shots placed on a datum with the OBS (as for the airgun data above). The airgun data recorded on OBS #3 and #5 (see figure 3.4) were corrected to datum levels at the instrument depths. No basement corrections were made as the data did not sample this deep, and consequently no information on the form of the interface was obtained. 41 The explosion profiles recorded on OBSs #3 and #5 were shot largely -in deep water, whilst the instruments recorded at shallow depths on the continental rise. The data were modelled with the water column present since correcting to datum levels at the OBS depths would involve replacing up to 2 km of water with sediment of uncertain velocity. The structure under OBSs #3 and #5 was known to be laterally variant, indicating a need for two-dimensional modelling techniques; thus the inclusion of seafloor topography in the modelling provided no additional complications for interpretation. 3.5 Comments on Data Trends 3.5.1 Airgun Sections For OBS #1 on the deep ocean floor, the velocities indicate a relatively shallow, high velocity medium (consistent with the basement depth indicated by the CSP record), whilst the clear arrivals suggest a comparatively homogeneous structure. OBSs #3 and #5, on the continental rise and slope respectively, indicate velocities characteristic of compacted, compressed sediments. The slightly erratic, lower energy arrivals suggest some degree of lateral heterogeneity resulting in increased attenuation of the seismic energy. 3.5.2 Explosion Sections The main amplitude group moves out to greater offset from OBS #1 to #5 (see figures 3.5 and 3.7), implying that the lower crustal structure from which the energy returns is deeper to the east. 42 O l 1 1 0 35 70 DISTANCE (km) Figure 3.7 Relative amplitude va r i a t i o n s for OBS #1. 3 and 5. The values shown are the maximum trace amplitudes observed on the record sections (figure 3.J>) during the second following the f i r s t a r r i v a l travel time. The o s c i l l a t o r y nature of the d i s t r i b u t i o n s is due to the inaccurate charge s i z e correction. Note the movement of the main amplitude peak, indicated by the arrowhead, to greater o f f s e t from OBS H\ eastwards to OBS *5. 43 The apparent velocity at large offset on a l l sections li e s between 8 km/s and 10 km/s; increasing from OBS #1 eastwards to OBS #5. This suggests an increase in dip of the Moho, or else a significant lateral increase in mantle velocity eastwards. The latter possibility is very unlikely. 44 " ....that small model of the barren earth, Which serves as paste and cover to our bones1." CHAPTER 4 DATA INTERPRETATION a detailed description 4.1 Summary of the Approach Taken To adequately model the complex seismic structure of the continental margin requires application of ray tracing procedures and synthetic seismogram techniques for inhomogeneous media. For the purpose of one-dimensional velocity inversion, shots recorded on OBS #1 were corrected to the sloping basement, effectively removing a wedge of sediment. The adjusted f i r s t arrival travel time-distance data were then transformed into the r-p domain and a linear programming technique (section 2.2; Garmany et a l . , 1979) applied to find a one-dimensional model which had the maximum velocity at the shallowest depth. Two-dimensional models satisfying the uncorrected travel times were then determined from the i n i t i a l structures by trial-and-error application of the ray tracing technique of Whittall and Clowes (1979). To incorporate the dynamic information, the WKBJ synthetic seismogram algorithm of Chapman (1978) was then used to refine the assumed one-dimensional velocity structure in the descending oceanic crust by subjective comparison of computed and observed records. Combination of the dipping slab model with overlying sediments then gave a two-dimensional model 'Shakespeare: Richard II. 45 which was tested for consistency with the observed data using an asymptotic ray theory (ART) synthetic seismogram algorithm for laterally varying media (McMechan and Mooney, 1980). Eastward extrapolation of the OBS #1 model provided starting models for interpretation of the OBS #3 and #5 explosion data. The final model thus obtained for a l l the data sets was then mutually consistent and compatible with a l l of the information. 4.2 Travel Time Modelling 4.2.1 Airgun data recorded on OBS #1 The raw data were corrected to a source depth equivalent to that of the instrument and topography was removed as described in section 3.4. One-dimensional travel time inversion by linear programming gave a preliminary velocity model which was subsequently modified by ray-tracing. The structure is defined to almost 4 km by these data (to the bottom of region B, inset of figure 4.1) and is very similar to that found by Au and Clowes (1982) approximately 100 km to the northwest on the Juan de Fuca plate. 4.2.2 Explosion data recorded on OBS #1 The near 50 km of the profile was corrected to place a l l shots and the instrument on the sloping basement (section 3.4). The resulting one-dimensional sub-sedimentary structure, deduced from travel time inversion, was then recdmbined with the sedimentary wedge to yield a laterally varying starting model for ray tracing. A model derived by this method is shown 46 Figure 4.1 A preliminary travel time model for OBS *1 Upper: F i l t e r e d (3-15 Hz) record section for OBS #1 explosion data. Times are adjusted to place the shot at the 2.52 km; depth of the OBS. and the amplitudes adjusted with an r* factor. The large amplitude secondary a r r i v a l s which r e p l i c a t e the features of the f i r s t a r r i v a l s are probably a shear phase converted from P-waves at the sediment-basement interface below the OBS. Arrows are f i r s t a r r i v a l picks; s o l i d l i n e links travel times calculated through a preliminary velocity-depth model beneath OBS #1. shown in the inset. The boundaries of A are from the CSP data ( f i g u r e 3.2) with the v e l o c i t i e s assumed; the structure of B from interpretation of the airgun p r o f i l e (figure 3.4); and the structure of C from the explosion data. Lower: A preliminary two-dimensional v e l o c i t y model, from the OBS #1 travel time data, showing ray paths for explosion a r r i v a l s . V e l o c i t i e s (km/s) are given for the top of each layer followed, a f t e r the semi-colon, by the v e l o c i t y gradient (km/s/km). 47 in figure 4.1. The 1.4° eastward dip of the sub-sedimentary layers over the f i r s t 50 km was deduced from the CSP record (figure 3.3). Boundaries below the sediment-basement interface are chosen to dip at the angle of this contact. The margin of the continental rise is approximated by a vertical boundary (50 km east of OBS 1) and the expected lateral change in sediment velocity has been modelled by three additional a r t i f i c i a l vertical boundaries. With these velocities, simultaneous modelling of the data and the extended onshore-offshore profile requires that dip of the sub-sediment layers increase to 4°±1° at the start of the rise. In this preliminary two-dimensional travel time model, the f i r s t arrival energy beyond 50 km was modelled as being due to turning rays in the lower crust. A velocity gradient of 0.1 km/s/km is required to produce these rays and mantle velocities (^8.0km/s) are reached near 12 km, a depth similar to that found along a seismic profile on the northern Juan de Fuca plate by Au and Clowes (1982). When rays were traced through this model into OBS #3 the travel time f i t was reasonable, but subsequent refinement, based on the continental slope sediment velocity information provided by the OBS #3 airgun data, and the Shell Anglo Cygnet well log (Appendix D), improved the agreement. 4.2.3 Airgun data recorded on OBS #3 These data were corrected to a source depth of 1.12 km (position of OBS) and the seafloor topography corrected (section 3.4). The linear programming inversion procedure gave 48 a velocity of 2.0 km/s in the top 1.2 km of sediment, underlain by a gradient region beginning at 2.6 km/s and increasing at 0.3 km/s/km. The data define structure down to almost 2 km. 4.2.4 Explosion data recorded on OBS #3 These data have a higher degree of associated non-uniqueness than the corresponding explosion line recorded on OBS #1. With OBS #1 the dip of the basement for nearer shots is known independently (from CSP records) and sediment velocities may be better estimated from other refraction studies in the area and from the Chevron Standard multichannel data (Appendix C). The basement dip under OBS #3 is unknown, although presumably >1.4°, whilst only shallow sedimentary information is available over a limited distance range. In fi t t i n g the travel times there is thus a trade-off between the magnitude of an eastward sediment velocity increase and the dip of the basement. The closer shots do give additional sediment velocity information, sampling deeper than the airgun, but the sparse shot spacing and expected structural complexity of the continental rise do not lead to well constrained solutions. The a priori modelling assumption used for the OBS #1 data set (i.e. that a l l boundaries below the sediment-basement interface are taken to dip at the angle of that contact) was introduced to a r t i f i c i a l l y constrain the two-dimensional solution, with i t s high degree of non-uniqueness. As noted above, the solution for the OBS #3 data set is even less well 49 constrained; indicating the need to retain this assumption. The closer 50 km of the OBS #1 explosion data are the most diagnostic for the sub-sedimentary structure and thus amplitude modelling of these data was conducted to give velocities and depths for the crust under the deep ocean (section 4.3) before f u l l interpretation of the OBS #3 data 4.2.5 Airgun data recorded on OBS #5 These data were corrected to a source depth of 0.16 km (position of the OBS) and seafloor topography corrections (section 3.4) were applied. The linear programming inversion procedure gave a velocity of 1.6 km/s near the sea floor, with the velocity rapidly increasing to 2.0 km/s at^0.5 km, then increasing at 0.5 km/s/km. The data define structure down to about 2 km depth. The velocity variation is consistent with that derived from interval transit times obtained in the Shell Anglo Cygnet well (Appendix D), which was situated ~~18 km from the OBS #5 site. 4.2.6 Explosion Data Recorded on OBS #5 The easternmost shot of the explosion line was denotated over the continental rise,^ 2 0 km east of the OBS position. The data are unreversed, but constraint is provided by the passage of rays beyond 30 km offset through material sampled by OBS #1 and #3. Linear programming techniques were not applied to the data, due to the expected high degree of lateral heterogeneity. I n i t i a l l y a two-dimensional -model was constructed which incorporated the structure derived for 50 OBS #1 and #3 and continued the dipping layers of the crust beneath the continental rise. This was then used as a starting model for trial-and-error forward modelling using the ray-tracing technique. Various layer dips and upper sediment structures were tried in an attempt to achieve consistency with the observed travel time data. It rapidly became apparent that the assumption of discrete upper crustal layers dipping beneath OBS #5 was incompatible with observations. In order to satisfy the trend of the f i r s t six arrivals (20-32 km on OBS #5 explosion section, figure 3.5), a velocity of ^ 4.5 km/s was required at 4 km depth. Interpretation of a multichannel line (almost co-linear with the line considered here, see appendix H). suggested the presence of a melange at about this depth (Snavely and Wagner, 1981). The upper crustal layers were replaced with a block of material with velocity ^ -4.6 km/s at 4 km, increasing at 0.2 km/s/km. The structure of the lower crust remained as before. 4.3 Amplitude Interpretation of OBS #1 Explosion Data General comments regarding amplitude interpretation are given in section 2.4.2 and a discussion of the computation of synthetic seismograms in Appendix A. For trial-and-error amplitude modelling of this data set, the WKBJ synthetic seismogram algorithm (Chapman, 1978) was used. In order to approximate one-dimensional structure, the data were again corrected to the dipping basement, removing a wedge of sedimentary material. The effect of this material on 51 observed amplitudes was expected to be negligible, due to the near vertical nature of rays passing through the sediment. This expectation was confirmed by the close agreement found in derived relative amplitudes (for a given model) between a WKBJ synthetic seismogram through the sub-sedimentary structure and an asymptotic ray theory two-dimensional synthetic seismogram, for which the model incorporated the sedimentary wedge. Only modelling of the general amplitude features is justified due to the lack of spatially dense data, and the probability of significant lateral heterogeneities in addition to the dipping sub-sedimentary layers. The relative amplitude of the f i r s t 1s of f i r s t arrival energy on the traces recorded on OBS #1 is shown as a function of distance in figure 4.2. Three main features were modelled; the primary large amplitude group (B; 30-40km), the secondary amplitude group CA;„. 12-20 km), and the decay region (C; >45 km). It is clear from the figure that the trace-to-trace amplitude variation is erratic. This is due in part to suspected lateral heterogeneities and focussing effects and to the inaccurate correction for varying charge size. The corrected and uncorrected relative amplitudes are compared in figure 4.2. Ideally the effect of a precise charge size adjustment factor on the amplitudes should be to smooth the trace-to-trace variation. No such smoothing is noticeable (in some cases the traces become more erratic), a feature which has led to concern about the validity of the charge size correction. It is certainly a more reliable procedure to model relative amplitudes from identical charge sizes, although this ~ i r J r 1 r — ' — i 1 i ° 0 10 20 30 40 50 60 70 DISTANCE (km) f— 1 -i I 1 1 0 35 70 0 35 70 DISTANCE (km) DISTANCE (km) F i g u r e 4.2 Amplitude c h a r a c t e r i s t i c s o f OBS *1 e x p l o s i o n d a t a . Upper: R e l a t i v e a m p l i t u d e v a r i a t i o n v s . d i s t a n c e f o r OBS #1. The v a l u e s shown a r e the maximum t r a c e a m p l i t u d e s o b s e r v e d on the r e c o r d s e c t i o n ( f i g u r e 3.5) d u r i n g a 1s i n t e r v a l f o l l o w i n g the f i r s t a r r i v a l t r a v e l time. Each c r o s s c o r r e s p o n d s t o one t r a c e . The d a t a show t h r e e main f e a t u r e s , which were c o n s i d e r e d s i g n i f i c a n t f o r mode 111ng: A) Secondary l a r g e a m p l i t u d e group B) P r i m a r y l a r g e a m p l i t u d e group C) Oecay of a m p l i t u d e p a s t 45 km. Lower: Diagrams to i l l u s t r a t e the e f f e c t o f c h a r g e s i z e v a r i a t i o n , and i t s c o r r e c t i o n , on the r e l a t i v e a m p l i t u d e v s . d i s t a n c e d i s t r i b u t i o n f o r OBS #1. a) U n c o r r e c t e d f o r ch a r g e s i z e v a r i a t i o n b) C o r r e c t e d f o r ch a r g e s i z e v a r i a t i o n c ) O n l y 50 kg c h a r g e s c o n s i d e r e d d) O n l y 2O0 kg c h a r g e s c o n s i d e r e d . 53 method halves the effective spatial data density for the VISP-80 data set. As noted in section 2.5, "the aim in a l l seismic modelling should be to produce the velocity-depth model which has the least structure s t i l l compatible with a l l the data." The preliminary velocity variation beneath OBS #1 (inset figure 4.1), derived from travel time modelling of OBS #1 data has three distinct structural characteristics: i) a velocity discontinuity at 3.75 km i i ) a velocity discontinuity at 7.00 km i i i ) a gradient region in the lower crust with Moho velocities (8.0 km/s) at 12 km. 4.3.1 Boundary at 3.75 km The position of some form of velocity increase is constrained to be centered at 3.75 ± 0.5 km by the travel time data,, with the given errors. Calculated travel times through the model are sensitive to the form of the velocity variation (see synthetic examples; figure 2.2). The presence of a region of rapid velocity increase at 3.75 km is clearly indicated by the near high amplitude group (12-20 km offset). Following the modelling policy of producing a smoothest velocity variation consistent with the data, the boundary was changed to a 1 km thick velocity transition zone (gradient «= 0.8 km/s/km). Further smoothing of the variation removed the observed amplitude decrease around 24 km, and was thus rejected. 54 4.3.2 Boundary at 7.0 km The presence of this boundary, or a region of rapid velocity increase, has the effect of sp l i t t i n g the main amplitude group into two sub-groups. This could be consistent with the observed data, but as noted above, the crude spatial sampling together with uncertainty regarding the validity of the charge size correction means that only a general f i t to the main amplitude group can be made. This feature was removed entirely from the model and replaced with a smooth increase from just below the velocity increase around 3.75 km to the Moho. 4.3.3 Lower crust and Moho structure The decay of- amplitude beyond 45 km offset is a clear feature of the data, although there is an anomalously high amplitude for a shot at 50 km. On the record section (figure 3.5, upper) this energy packet is shifted later in time relative to the corresponding feature for closer shots. The energy is due, at least in part, to some secondary arrival and may be a local focussing effect from the Moho region: a denser sampling in X would be needed to resolve this point. This high amplitude energy packet cannot be satisfactorily modelled by using the methods and approximations employed here as i t is only defined by one point on the relative amplitude-distance plot. A large range of Moho models has been found to be consistent with the observed travel time and amplitude data. The only requirement for synthetic seismogram modelling is 55 that models produce a decay in amplitude after 50 km offset. The i n i t i a l model derived from travel time inversion (inset, figure 4.1) does not satisfy this condition; high amplitudes are produced out to 75 km offset; indicating a Moho significantly shallower than 12 km. The different possible types of velocity transition at the Moho, and their effects on observed amplitudes are discussed in Appendix E. In the OBS #1 explosion data, no clearly identifiable sub-critical reflection energy is observed on the record section (figure 3.5). Thus, the observations favour choice of a smooth, gradational, change in velocity at the Moho, although the possibility of a small step-function velocity boundary is not inconsistent with the data as the resolution of these features is low (due to the long source wavelet and low spatial sampling density). The effect of Moho depth, represented by a change in velocity gradient, on relative amplitude-distance distributions is summarized in figure 4.3. The position of the main amplitude peak is quite sensitive to Moho depth, moving further out as the boundary becomes deeper. By comparison with the real data, the preferred depth for the gradient change is 9.3 km, although as noted above, the position of the peak in amplitude is ambiguous due to the different charge sizes used. With this preferred model the gradient changes from 0.2 km/s/km to 0.01 km/s/km at a velocity of 8.0 km/s. The value for mantle gradient is that used in the modelling of the onshore-offshore profile by Spence (personal communication, 1982). Combination of this velocity structure in the dipping 56 Figure 4.3 The e f f e c t of Moho depth on amplitude d i s t r i b u t i o n s . L e f t : A three-dimensional plot of the Moho depth used in a model against the r e l a t i v e amplitude-distance d i s t r i b u t i o n s produced by the WKBJ synthetic seismogram algorithm used with that model. Note the movement of the main amplitude group to greater o f f s e t with increasing Moho depth. Right: Preferred velocity-depth structure under OBS <M. Moho-depth Is defined as the change In v e l o c i t y gradient at B.O km/s. -p l a t e with the sedimentary s t r u c t u r e produces good agreement with observed data ( f i g u r e 4.7, and 4.8). The a r r i v a l s beyond 40 km are modelled as head waves, produced i n the model by a minute v e l o c i t y d i s c o n t i n u i t y at the Moho. The weak a r r i v a l s seen on the se c t i o n ( f i g u r e 3.5) beyond 50 km o f f s e t may be true head waves, i n t e r f e r e n c e head waves, or tu r n i n g rays from 57 the mantle. With the low mantle velocity gradient assumed here, the . arrival time difference between these different types of rays is negligible. These arrivals do constrain the dip of the Moho under the continental rise to 3°±1°, assuming no lateral change in upper mantle velocity. 4.4 Practical Use of WKBJ and ART Algorithms As noted above (section 4.3), the effect of a sedimentary wedge on the amplitude of f i r s t arrivals is negligible, but a comparison of ART and WKBJ synthetic seismograms through the same model does not show absolute agreement. This is illustrated in figure 4.4. The OBS #1 explosion data, corrected to a sloping basement, are compared to the WKBJ and ART synthetic seismograms, calculated through the preferred velocity-depth structure below OBS #1. Relative trace Overleaf: F i g u r e 4.4 A comparison of the WKBJ and ART s y n t h e t i c seismogram a l g o r i t h m s . a) OBS *'1 e x p l o s i o n d a t a w i t h the t r a v e l time e f f e c t of water column and t h i c k e n i n g of sediments eastwards removed. Sol i d 1ine 1 Inks the f i r s t a r r i v a l t r a v e l times from the model. b) WKBJ s y n t h e t i c seismogram c a l c u l a t e d through the p r e f e r r e d v e l o c i t y - d e p t h model f o r OBS n\ ( d ) . The s o u r c e wavelet used was taken from a c l e a r a r r i v a l energy packet In the r e a l d a t a ( see f i g u r e 2.3). and Is o f a p p r o x i m a t e l y one second d u r a t i o n . Note the p r e s e n c e o f d i f f r a c t i o n s beyond 45 km o f f s e t . c ) ART s y n t h e t i c seismograms. c a l c u l a t e d through the p r e f e r r e d model ( d ) . u s i n g t h e same s o u r c e w a v e l e t as i n ( b ) . Note the absence of d i f f r a c t i o n s p a s t 45 km o f f s e t . d) P r e f e r r e d v e l o c i t y - d e p t h model f o r OBS #1. e) R e l a t i v e a m p l i t u d e d i s t r i b u t i o n s c o r r e s p o n d i n g to the seismograms In a. b. and c. S o l i d l i n e l i n k s the t r a c e a m p l i t u d e s f o r the r e a l d a t a , which have been c o r r e c t e d f o r c h a r g e s i z e v a r i a t i o n . V a l u e s shown a r e the maximum t r a c e a m p l i t u d e s o b s e r v e d on the seismogram d u r i n g the second f o l l o w i n g the f i r s t a r r i v a l t r a v e l time. D o t t e d l i n e l i n k s the t r a c e a m p l i t u d e s f o r WKBJ seismograms. w h i l s t the dashed l i n e c o r r e s p o n d s t o the v a r i a t i o n f o r the ART seismograms. Note t h a t the main a m p l i t u d e peak f o r the ART v a r i a t i o n Is s h i f t e d 7 km r e l a t i v e to t h a t f o r WKBJ and the r e a l d a t a . 58 V (km/s) DISTANCE (km) 0 3 6 9 0 35 70 59 amplitudes for the three sections are compared at the bottom of the figure. The highest amplitude for the WKBJ solution occurs at 34 km, corresponding to that observed for the charge-size-corrected real data. Amplitude then decays smoothly with increasing offset. The ART solution produces a smooth build-up in amplitude, corresponding to rays turning in the velocity gradient of the lower crust. The maximum amplitude then corresponds to the Moho grazing ray. At greater offsets no energy is received. Head waves are not included in the ART approach and the mantle gradient is too low to give turning rays with the ray density used to generate arrivals in the crustal material. As can be clearly seen from figure 4.4, not only is there a difference in the amplitude behaviour for WKBJ and ART beyond the main peak, but also in the position of that peak. The peak in the ART solution is 7 km further out than that for the WKBJ, as calculated through the preferred one-dimensional model. This lack of agreement between the. two methods reflects the different mathematical approaches used. The WKBJ method is a wave-field approach and includes the effects of geometrical diffraction. This leads to the smooth decay of amplitude, seen in a region where classical ray theory would predict no energy, and to the shifting of the main amplitude peak closer in- due to destructive interference in the region of the change in velocity gradient (at the Moho). The zero-order ART method is based, by definition, on ray theory; no diffraction or interference effects are considered, leading to the observed differences from the WKBJ solution. 60 In two-dimensional modelling of the three OBS data sets considered here, relative changes in amplitude characteristics have been matched by corresponding changes in the synthetic seismograms computed using ART. Thus the size of the shift in main amplitude peak from instrument to instrument (figure 3.7) has been modelled, but the peak occurs at larger offset on the synthetic section as compared to the real data. This consistent discrepancy is due to the inherent inaccuracies of the zero-order ART method, discussed above. 4.5 Amplitude Characteristics, for a l l Receivers A comparison of the relative amplitude information obtained from a l l three instruments is shown in figure 4.5. The f i r s t column shows the variation when a l l 35 charges are considered. The distributions are very spiky, due to application of an empirical charge-size correction (section 4.3). The second column gives the distributions for 18 50 kg charges; here the shift of main amplitude peak to greater offset, with more easterly placement of the instrument, is clearly seen. The third column represents 13 200 kg charges; the trend is less clear when fewer shots are being considered. As noted above (section 3.5.2), the shift in main amplitude peak with OBS position is due to the eastwardly deepening crustal structure. The Moho dip is determined from the shift in relative amplitude peak from instrument to instrument. There is considerable ambiguity associated with position of the peak due to the spatially discrete data and inaccurate charge size correction. The peak for OBS #1 occurs 61 > o i—i DISTANCE (km) F i g u r e 4.5 R e l a t i v e a m p l i t u d e t r e n d s and ch a r g e s i z e . These diagrams I l l u s t r a t e the e f f e c t of c o n s i d e r i n g d i f f e r e n t c o m b i n a t i o n s of ch a r g e s on the r e l a t i v e a m p l i t u d e d i s t r i b u t i o n s f o r OBS * 1. H3. and #5. The upper diagrams a r e the v a r i a t i o n s f o r OBS #1. L e f t : C o n s i d e r i n g a l l c h a r g e s and c o r r e c t i n g f o r the v a r i a t i o n i n s i z e . M i d d l e : c o n s i d e r i n g o n l y 50 kg c h a r g e s R i g h t : c o n s i d e r i n g o n l y 200 kg c h a r g e s . The diagrams f o r OBS #3 and 15 f o l l o w the same p a t t e r n : OBS 13. middle; OBS *5. bottom. Note the l a r g e change i n d i s t r i b u t i o n shape w i t h d i f f e r e n t charge c o m b i n a t i o n s . The movement of the main a m p l i t u d e peak t o g r e a t e r o f f s e t from OBS * 1 t o OBS #5 i s p r e s e n t i n almost a l l c o m b i n a t i o n s . ' at approximately 34 km offset, for OBS #3 at 40 km, and for OBS #5 at 54 km. The Moho depth is only sampled once for each data set and thus Moho dip refers to the trend of the line connecting sampled points. Some constraint is provided by head waves or mantle turning rays, although the travel time picks involved generally have large error. 62 4.6 Construction of Final Models Fitting a l l the Data The descending slab structure beneath OBS #1 and #3 was constructed from the one-dimensional structure for OBS# 1 which was derived using the WKBJ algorithm. Under OBS #5 the upper crustal structure was replaced with a block of constant velocity gradient material, required to f i t the travel time data. Forward modelling was then performed to produce final models for each of the data sets which were mutually consistent. 4.7 Final Models - a detailed description One-dimensional seismic structures satisfying the airgun travel time data are shown in figure 4.6, together with a comparison of these data with travel times derived from ray-tracing. Travel time modelling of the explosion data for the three OBSs is illustrated in figures 4.7, 4.9, and 4.11; whilst amplitude modelling with the ART''algorithm is shown in figures 4.8, 4.10, and 4.12. Values for the preferred velocities and velocity gradients for the entire structural section are given in figure 4.13. In a l l models a group of rays turning in the lower crust produces the observed main amplitude group. The sharp cut-off produced by the ART algorithm is not physical, as noted in section 4.4. 63 0 6 H W Q VELOCITY (km/s) 1 2 3 4 5 6 1 1 1 1 — 1 \ >* » • \ 5 N V \ ^ 8 10 12 14 16 18 20 DISTANCE (km) 8 10 12 14 16 DISTANCE (km) 0 2 4 6 8 DISTANCE (km) 10 12 14 16 64 4.7.1 Airgun Models (Figure 4.6) OBS #1: The derived structure exhibits features characteristic of the upper crust in this region (e.g. Au, 1981; Cheung, 1978). The data constrain sub-sedimentary material velocities and indicate a region of rapid velocity increase (1.4 km/s/km) below the sediment-basement interface. At approximately 2.6 km a velocity of 6.0 km/s is attained and there is a decrease in velocity gradient. Between 3.25 and 4.25 km depth there is a region of relatively rapid velocity increase (gradient = 0.76 km/s/km). The structure is constrained down to almost 4.5 km depth by the airgun information. OBS #3: The data do not directly constrain the upper few 100 m of sediment, but in order to preserve travel times the velocities must be low: close to 1.6 km/s. At about 200 m depth there is a rapid increase in velocity (gradient = 1.7 km/s/km) to a depth of 0.7 km where the velocity is 2.5 km/s. Here there is a decrease in gradient to 0.3 km/s/km. The structure is constrained down to about 1.8 km depth by the airgun data. Overleaf: F i g u r e 4.6 A i r g u n models and d a t a comparison. Upper: p r e f e r r e d , o n e - d i m e n s i o n a l . s h a l l o w v e l o c i t y - d e p t h s t r u c t u r e s beneath the t h r e e OBSs. S o l i d l i n e : under OBS #1 Dashed l i n e : under OBS *3 Do t t e d l i n e : under OBS #5. Note t h a t the sediment v e l o c i t i e s (<4.0 km/s) f o r OBS *1 were d e r i v e d from CSP and m u l t i c h a n n e l i n f o r m a t i o n , not from d i r e c t i n t e r p r e t a t i o n of the a i r g u n d a t a . Lower: comparisons of r e a l t r a v e l times w i t h those c a l c u l a t e d through the models above. The s o l i d l i n e l i n k s the a c t u a l f i r s t a r r i v a l t r a v e l time p i c k s , w h i l s t the c r o s s e s a r e computed v a l u e s . 65 OBS #5: The derived structure indicates a low velocity 1.6 km/s in the top 200 - 300 m, below which there is a region of rapid velocity increase (gradient = 1.0 km/s/km). At 0.7 km depth, and a velocity of 2.0 km/s, there is a lowering of the gradient to 0.55 km/s/km. The structure is constrained down to almost 3 km depth by the airgun data. The top 2 km of structure are compared to sonic log data in figure D.I; the agreement is good. 4.7.2 OBS #1 Explosion Profile Model (Figures 4.7 and 4.8) The preferred velocity-depth variation below OBS #1 is illustrated in figures 4.3, 4.4, and 6.1. Derived structure beneath the instrument is typical of marine crust in this region (e.g. Au, 1981). Upper crustal velocities, defined by the airgun data are described in section 4.7.1. The simplest lower crustal structure compatible with the data is a constant velocity gradient of approximately 0.2 km/s/km, changing to 0.01 km/s/km at the Moho where the velocity is 8.0 km/s. This lower crustal gradient region leads to a group of turning rays (figure 4.7b and 4.8a) which produce the observed large amplitudes between 25 and 45 km (figure 4.8 c). OBS 1 66 OH ' 1 1 1 1 1 1 1 0 10 20 30 40 50 60 70 DISTANCE (km) Figure 4.7 OBS #\ t ravel time model and data comparison. (a) 1:1 representat ion of the model boundaries (dashed l ines ) which separate regions of d i f f e r e n t v e l o c i t y and/or v e l o c i t y gradient, the values of which are shown In f i gure 4.13. Note that OBS <M l i e s at the western end of the model. The water column is not included in the model. (b) Ray t rac ing through the model, fo l lowing the method of Whltta l l and Clowes (1979). Rays c lose to the o r i g i n have been a r t i f i c i a l l y blanked o f f . Regularly spaced a r r i v a l s emerging from boundaries 2 and 6 represent head waves. The a r r i v a l time d i f fe rence between head waves produced at the lower boundary (the Moho), and turning rays In the upper mantle Is n e g l i g i b l e . Head waves are represented here for the purpose of 11 lustrat ion. (c) Comparison of rea l and synthet ic t rave l time data. The s o l i d l i ne l inks real f i r s t a r r i v a l travel time p icks , whilst the crosses are a r r i v a l times computed from the ray trace in (b). Ve r t i c a l crosses are turning rays: diagonal ones, head waves. 67 0BS1 DISTANCE (km) 0 10 20 30 40 50 60 70 C O X o -I E - - ^ ^ ^ ^ 0 10 20 30 40 50 60 70 DISTANCE (km) 0 10 20 30 40 50 60 70 DISTANCE (km) 68 The dip of the subducting crust increases from 1° to almost 3° beneath the continental rise. Crustal layers a l l dip at the same angle as discussed in section 4.2.2. The a r t i f i c i a l vertical boundaries in the continental rise region model the lateral increase in velocity from 1.9 to 2.6 km/s. Overleaf: F i g u r e 4.8 OBS M\ Amplitude model and d a t a comparison. (a) R a y - t r a c e through a c u b i c - s p l I n e d g r i d of v e l o c i t y v a l u e s d e f i n i n g the p r e f e r r e d model. f o l l o w i n g the ART method o f McMechan and Mooney (19SO). Due to the r e q u i r e m e n t s of the a l g o r i t h m . t h e r e a r e no v e r t i c a l b o u n d a r i e s : l a t e r a l changes i n v e l o c i t y o c c u r smoothly. The l a c k o f r a y s a t l a r g e o f f s e t ( p a s t 45 km) i s . In r e a l i t y , a s e v e r e d e c r e a s e due to the s m a l l v e l o c i t y g r a d i e n t In the upper mantle. The water column Is not I n c l u d e d i n t h i s model. (b) S y n t h e t i c seismograms c a l c u l a t e d from the r a y - t r a c e above, u s i n g a s y m p t o t i c r a y t h e o r y (ART). The s o l i d l i n e l i n k s f i r s t a r r i v a l t r a v e l t Imes. In s e t shows the r e l a t i v e a m p l i t u d e v a r i a t i o n c o r r e s p o n d i n g to the seismograms. The s o u r c e wavelet used was o b t a i n e d from a r e a l d a t a t r a c e , and Is I l l u s t r a t e d In f i g u r e 2.3 Note the s h a r p c u t - o f f i n a m p l i t u d e p a s t 45 km due t o the absence of r a y s t r a c e d by the program i n ( a ) . ( c ) F i l t e r e d (3-15 Hz) r e c o r d s e c t i o n f o r OBS #1 e x p l o s i o n d a t a . A c h a r g e s i z e c o r r e c t i o n and an r' enhancement f a c t o r have been a p p l i e d . Times a r e a d j u s t e d to p l a c e the s h o t s a t 2.52 km depth, thus removing the water column e f f e c t . S o l i d l i n e l i n k s f i r s t a r r i v a l t r a v e l times through the model. Inset shows the r e l a t i v e a m p l i t u d e v a r i a t i o n c o r r e s p o n d i n g to the seismograms. 69 4.7.3 OBS #3 Explosion Profile Model (Figures 4.9 and 4.10) The velocity-depth distribution in this model is essentially that for OBS #1, with the distance scale reversed. The sediments under the continental rise have higher velocities than those under OBS #1 as described in section 4.7.1. The triangular wedge at the easterly end is the block of melange material described in section 4.2.6. Note that the models for OBS #3 (and #5) include the water column, whereas those for OBS #1 do not. A secondary large amplitude group has been produced by the ART model for OBS #3 but this is not seen in the real section (figure 4.10c). The discrepancy is thought to be due to the high degree of attenuation expected in the continental rise region. F i g u r e 4.9 OBS *3 T r a v e l - t i m e model and d a t a comparison. ( a ) 1:1 r e p r e s e n t a t i o n of the model b o u n d a r i e s (dashed l i n e s ) . V a l u e s f o r v e l o c i t i e s and v e l o c i t y g r a d i e n t s a r e g i v e n In f i g u r e 4.13. Note that OBS *3 l i e s at the e a s t e r n end o f the model. The water column has been i n c l u d e d In t h i s model. (b) Ray t r a c e p e r formed through the model. Rays c l o s e t o the o r i g i n have been a r t i f i c i a l l y b l a n k e d o f f . R e g u l a r l y spaced a r r i v a l s emerging from the lower boundary (Moho) r e p r e s e n t head waves. ( c ) Comparison o f r e a l and s y n t h e t i c t r a v e l time d a t a . S o l i d l i n e l i n k s r e a l f i r s t a r r i v a l t r a v e l time p i c k s , w h i l s t c r o s s e s a r e a r r i v a l times computed from the r a y t r a c e i n ( b ) . V e r t i c a l c r o s s e s a r e t u r n i n g r a y s : d i a g o n a l ones, head waves. 71 OBS 3 E DISTANCE (km) W 0 10 20 30 40 50 60 70 F i g u r e 4.10 OBS 13 Amplitude model and d a t a comparison. (a) Ray t r a c e p e r formed through a c u b i c - s p l i n e d g r i d o f v e l o c i t y v a l u e s d e f i n i n g the p r e f e r r e d model. The water column and s e a f l o o r topography i s i n c l u d e d . (b) S y n t h e t i c d a t e c a l c u l a t e d from the r a y t r a c e above, u s i n g ART. Ins e t shows the r e l a t i v e a m p l i t u d e v a r i a t i o n c o r r e s p o n d i n g t o the seismograms. ( c ) F i l t e r e d (3-15 Hz) s e c t i o n f o r OBS #3 e x p l o s i o n d a t a . D i s t a n c e and ch a r g e s i z e c o r r e c t i o n s as f o r OBS *1 ( f i g u r e 4.8). Times i n c l u d e the e f f e c t of the water column. Inset shows the c o r r e s p o n d i n g r e l a t i v e a m p l i t u d e v a r i a t i o n . 72 4.7.4 OBS #5 Explosion Profile Model (Figures 4.11 and 4.12) The westernmost 70 km of this model is just that used to satisfy the OBS #1 and #3 data. Shallow sediments under the continental slope have relatively high velocities as described in section 4.7.1. At approximately 4 km there is a block of high velocity material (V >4.6 km/s) which has been correlated with a postulated Miocene melange unit (section 4.2.6). The velocity gradient in this block is about 0.2 km/s/km. The vertical boundary separating this unit from the layered upper crust to the west is not thought to be real, but crudely approximates the expected smooth lateral velocity change. Turning rays in the block of material beneath the rise produce an amplitude group not seen in the real data (figure 4.12). This lack of agreement is thought to be due to attenuation in the highly sheared medium and the very simplified model which must be used for the calculations. The lower crustal structure remains constant over the ( whole model, being a velocity gradient region of approximately 0.2 km/s/km. The dip of the Moho increases under the continental slope to approximately 6°. 73 O B S 5 DISTANCE (km) W N i 1 1 1 1 1 1 1 1 1 1 0 10 20 30 40 50 60 70 80 90 100 DISTANCE (km) Figure 4.11 DBS #5 Travel-tImemodel and data comparison. (a) 1:1 representation of model boundaries (dashed l i n e s ) . Values for v e l o c i t i e s and v e l o c i t y gradients are given In figure 4.13. Note that OBS <C5 l i e s at the eastern end of the model. The water column and seafloor topography are included. (b) Ray trace performed through the model. Regularly spaced a r r i v a l s emerging from boundaries 3 and S represent head waves. Note that the v e r t i c a l boundary at 35 km is not thought to be r e a l , but crudely approximates a smooth l a t e r a l v e l o c i t y change. (c) Comparison of real and synthetic travel time data. S o l i d l i n e and crosses as for figures 4.7 and 4.9. O B S 5 c DISTANCE (km) 0 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 DISTANCE (km) F i g u r e 4.12 OBS #5 Amplitude model and d a t a comparison. (a) Ray t r a c e performed through a c u b l c - s p l I n e d g r i d of v e l o c i t y v a l u e s d e f i n i n g the p r e f e r r e d model. The water column and s e a f l o o r topography a r e i n c l u d e d . ( b ) S y n t h e t i c d a t a c a l c u l a t e d from r a y t r a c e above, u s i n g ART. I n s e t shows the r e l a t i v e a m p l i t u d e v a r i a t i o n c o r r e s p o n d i n g to the seismograms. ( c ) F i l t e r e d (3-15 Hz) s e c t i o n f o r OBS *5 e x p l o s i o n d a t a . C o r r e c t i o n s as f o r OBS *3 ( f i g u r e 4.10). Inset shows the c o r r e s p o n d i n g r e l a t i v e a m p l i t u d e v a r i a t i o n . 75 W DISTANCE (km) E 0 20 40 60 80 100 CV3 F i g u r e 4.13 F i n a l v e l o c i t y s t r u c t u r e . P r e f e r r e d v e l o c i t i e s f o r the c o n t i n e n t a l margin a r e shown here. The f i r s t number g i v e s v e l o c i t y ( I n km/s) a t the top of a r e g i o n , f o l l o w e d a f t e r the c o l o n by the v e l o c i t y g r a d i e n t ( I n km/s/km) If one was used. 4.8 Interpretation Summary A two-dimensional seismic structural section has been derived which is consistent with the travel time and amplitude information contained in the separate record sections. Preferred values for velocities and velocity gradients are 'shown in figure 4.13. Features of the model are discussed in Chapter 6. Any derived seismic structure should be consistent with observed gravity variations over that structure. Consistency between gravity data and the preferred seismic model has been checked by conversion of compressional-wave velocities to densities, and is discussed in the next chapter. 76 " Gravity and lightness are only attraction and flight . Nothing is naturally heavy or l i g h t 1 . " CHAPTER 5 Gravity Interpretation 5.1 On the Analysis of Gravity Data The interpretation of gravity information has a higher degree of associated non-uniqueness than that involving seismic travel time and amplitude data. A given anomaly may be produced by an infinite number of bodies of varying density, shape, and depth. The most frequently used method of gravity anomaly interpretation is by two-dimensional forward modelling (e.g. Molnar, 1977). Modifications to an i n i t i a l model are based upon a subjective "goodness of f i t " criterion when comparing the calculated and observed gravity variations. Oldenburg (1974) has formulated an iterative inversion scheme involving Fourier transforms of the gravitational anomaly. The method is fast enough to be practical for an iterative approach, but is limited by the assumptions that an observed anomaly is caused by a single surface between two constant density media and that the perturbing body is two-dimensional. In a l l gravity interpretations additional geophysical information is- used to constrain the models and in general, only large scale features are modelled. Detailed models containing many small blocks of material of different density, 1Giordano Bruno 77 such as that due to Couch and Braman (1979), should be viewed with extreme caution. Whilst f i t t i n g the observed gravity variation almost exactly, the degree of model complexity is misleading; many different combinations of small blocks with differing densities and depths could f i t the data equally well. The aim behind gravity modelling should be the same as that behind refraction data analysis, i.e. to produce the model which has the least structure compatible with a l l the data. For the purposes of crustal studies, i t is generally not necessary to model small variations in the gravity anomaly which may be due to local, near-surface variations in density. In many interpretations, discontinuities in structure derived from seismic data analysis are used to delineate blocks of material of differing density. Whilst this procedure may be valid in many situations, i t should be treated with caution as seismic velocity variations do not depend solely on density variations. In order to constrain a seismic velocity model with a gravity data interpretation, regions of different wave velocity must be assigned densities. Empirical relations between density and velocity are derived from laboratory measurements on rock samples. In marine work, samples are often obtained from the Deep Sea D r i l l i n g Project (DSDP) and are tested under pressure and water saturation conditions closely resembling those experienced by material in situ. The most used velocity-density curve has been that of Nafe and Drake; f i r s t published in 1957, and updated several 78 times (Nafe and Drake, 1963; Talwani et a l . , 1959b). Hamilton (1978) presents separate v-p relationships for principal sediment and rock types in the sea floor; some v-p values for sediments are given in Table 5.1. •V(km/s) p(g cm"3) 1 .9 2.0 2.0 2.1 2.25 2.2 2.50 2.3 2.8 2.4 3.3 2.5 Table 5.1 Compressional-wave velocity vs. density in marine sediments. p=0.917+0.744V-0.08V2 (after Hamilton, 1978)-5.2 Gravity Data Description. A gravity anomaly map (free air over the ocean, Bouguer on land) has been compiled for the region including southwestern British Columbia and northwestern United States (Riddihough, 1979). A number of profiles perpendicular to the coast were interpreted to study the change in subduction regime from northern Vancouver Island to Oregon. In particular, profile 2 (figure 5.1) was nearly coincident with VISP-80 line I. The large scale structure of the continental margin along this profile was modelled by Riddihough (1979), using the gravity data and additional seismic constraints. As noted in section 1.2, the broad , features of the gravity f i e l d , the parallel 'low-high' system corresponding to the trench and the 79 F i g u r e 5.1 L o c a t i o n map o f g r a v i t y p r o f i l e . S o l i d l i n e I n d i c a t e s p o s i t i o n of the p r o f i l e , superimposed on a s i m p l i f i e d t e c t o n i c map of the r e g i o n . arc-trench gap, are features which elsewhere are characteristic of an active margin. In figure 5.2 (lower) the free air gravity anomaly over the marine portion of the profile in figure 5.1 is shown. Over Cascadia Basin on the west, the average gravity anomaly is near zero, indicating that the region is essentially in isostatic equilibrium. The anomaly becomes negative over the continental slope, reflecting the sediment-filled margin trench. The positive anomaly over the outer shelf is due , to shallowing water depth and the greater sediment densities here. Rapidly decreasing gravitational attraction over the inner shelf reflects low density Tofino Basin sediments. 80 5.3 Gravity Data Interpretation 5.3.1 Method Forward modelling of the gravity information was performed with the aid of an algorithm which computes the gravitational attraction due to arb i t r a r i l y shaped polygons in two dimensions, as described in Appendix G. The continental margin off Vancouver Island is thought to be an approximately linear feature over 50 km to the north and south. Thus, the assumption of two-dimensionality of the layers is expected to introduce relatively small errors. An i n i t i a l input model for the algorithm was constructed by forming regions of fixed density with boundaries in the same positions as those of the preferred seismic model (figure 4.13). The densities of the sedimentary blocks were chosen from the average seismic velocities for the regions by use of the relationships shown in table 5.1. Upper crustal density was taken as 2.62 g cm"3 after Couch and Braman (1979), whilst a lower crustal density of 2.92 g cm'3 was used, following Riddihough (1979). A velocity-density relation for the block of highly sheared melange proposed by Snavely and Wagner (1981), and indicated by the VISP-80 data, may only be guessed at, as the true nature of the material is unknown. In forward modelling, the melange block density was l e f t as a free parameter. The allocation of densities to compressional-wave velocity regions is inherently inaccurate for two reasons. Fi r s t l y , the relationships between v and p are empirical and one velocity is related to a range of densities; secondly, 81 many regions in the seismic model have velocity gradients, but only one density is used to model the region. It is therefore possible to perturb the densities of a given gravity model whilst s t i l l maintaining consistency with a corresponding seismic structure. In order to model observed gravity variations over the 100 km long profile between OBS #1 and #5, a density structure must be assumed well beyond the limits of this profile, and must extend much deeper than the maximum depth sampled by the seismic data to eliminate end effects. The central region of the Juan de Fuca plate is thought to be approximately laterally homogeneous; thus structure west of the seismic profile was modelled as plane layers with the same density stratification as the units below OBS #1. East of the profile the earth is known to be highly laterally variable. The structure proposed by Riddihough (1979) was employed to model the descending crust and transition to continental material beneath Vancouver Island. Densities down to 200 km depth were incorporated into the model; values for lithospheric mantle (p=3.34 g cm"3, thickness 13.1 km) and asthenospheric mantle (p=3.295 g cm'3, thickness 176.5 km) were taken from Riddihough (1979). Synthetic gravity variations were computed at stations 10 km apart; a separation similar to that between the real data points obtained from the gravity anomaly map. 5.3.2 Model Description The preferred density structure for the continental 82 DISTANCE (km) F i g u r e 5.2 G r a v i t y model and d a t a comparison. Upper: P r e f e r r e d g r a v i t y s t r u c t u r e . The b o u n d a r i e s c o r r e s p o n d to those i n the s e i s m i c model ( f i g u r e 4.13). D e n s i t i e s a r e shown i n g cm-'. Lower: Computed anomaly ( s o l i d l i n e ) , compared t o r e a l d a t a (dashed l i n e ) ; t h e r e i s c l o s e agreement. S t a t i o n s p a c i n g f o r b o t h v a r i a t i o n s i s IO km. 83 margin is shown in figure 5.2 (upper). The seismic profile extends from approximately '25 to 125 km on the distance scale shown. A l l sub-sedimentary boundaries correspond exactly to those in the preferred seismic model (figure 6.1); however, the upper crustal structure between OBS #1 and the continental rise is given one density (2.62 g cm"3), whereas the seismic model includes three different velocity gradient regions. The difference in calculated anomaly between this procedure and allocation of densities to the individual layers of the upper crust is negligible. Derived densities and seismic velocities (figure 6.1) are consistent (given the inherent uncertainties involved in the allocation of p to v) with the empirical relationships shown in table 5.1. Major boundaries between blocks of sediment with differing density are not thought to be real. Relatively smooth variations in the physical properties of the material are expected, although some shallow fault bounded blocks are indicated by the CSP data (figure 1.3), and deeper faults have been interpreted from a multichannel reflection section (Appendix H). The upper sedimentary block lying between 124 km and the eastern edge of the model (figure 5.2) represents the Tofino Basin sediments. This structure l i e s to the east of OBS #5, and is thus unconstrained by the VISP-80 seismic data. As noted in section 5.2, the presence of relatively low density Tofino Basin sediments is required to produce a rapid decrease in gravitational attraction over the inner shelf. The preferred density for the block of melange material 84 is 2.8 g cm"3. This is undoubtedly an average value. The boundaries of this region are only crudely defined by seismic refraction data; both velocities and densities are expected to be highly laterally and vertically variable within the block. A density of 2.8 g cm"3 and seismic velocities ranging from 4.6 to 6.0 km/s are too high for standard marine sedimentary material (Hamilton, 1978); these values are more characteristic of basalts than of sediment. V-p relations in basalts from the boreholes of the Deep Sea D r i l l i n g Project (Christensen and Salisbury, 1975) indicate a velocity of *5.7 km/s for a density of 2.8 g cm"3. The lower average seismic velocity derived for the melange (~5.3 km/s) may be due to the highly sheared nature of the basaltic material composing the block. Seismic velocity may be further reduced by incorporation of sedimentary material into fissures formed in the upper section of the block. Thus the preferred average velocity and density values are consistent with the notion that a melange has been formed by deformation of upper oceanic crustal material. The computed gravity anomaly from the preferred model is compared with real data in figure 5.2 (lower). Broad features of the anomaly have been matched i.e. the amplitude and wavelength of the sine function shaped variation. The greatest discrepancies between observed and calculated values occur in the continental rise region. This disagreement is expected, due to the modelling of this complex area with simple, block structures. The real gravity data have been modelled by using simple, 85 relatively large scale regions of constant density. Errors wi l l be introduced as described above from violations of the two-dimensionality assumption, and inaccurate allocation of densities to seismic velocities. When reasonable f i t s to observed gravity anomalies can be produced with very simple sub-sedimentary structures, the lack of constraint provided by gravity data becomes apparent. Although i t is encouraging to find consistency between the gravity and seismic models, i t should be remembered that the gravity data only constrain possible density structures very loosely, due to the inherent non-uniqueness of the data inversion. For example, gravity information provides no additional details regarding position of the relatively low density contrast Moho boundary. 86 " The end proveth everything 1." CHAPTER 6 Summary and Discussion A structural model for the subducting Juan de Fuca plate at the continental margin off British Columbia has been interpreted from a seismic refraction profile recorded on three ocean bottom seismographs. Derived seismic models were constrained by continuous seismic profiles, multichannel seismic reflection sections, well-log data, and gravity information. The modelling procedure adopted throughout this study was to produce the simplest velocity and density structures possible, whilst retaining consistency with a l l geophysical information available. The term 'simplest' as applied here is a subjective quality. No purely mathematical definition can be expressed for two-dimensional seismic structures. The derived structural model for the continental margin is shown in figure 6.1. Each of the blocks has a fixed density, compressional-wave velocity at the upper boundary, and velocity gradient. Vertical boundaries in the model are not thought to be real, but crudely approximate smooth lateral changes in velocity and density. Velocities for the sediments are constrained by travel times from a CSP, normal moveout velocity determinations, travel times from multichannel reflection sections, sonic logs 'John Gower 87 F i g u r e 6.1 F i n a l v e l o c i t y and d e n s i t y s t r u c t u r e . Upper: 1:1 r e p r e s e n t a t i o n o f model b o u n d a r i e s . Lower: P r e f e r r e d v e l o c i t i e s and d e n s i t i e s f o r the c o n t i n e n t a l margin a r e shown h e r e . The f i r s t number g i v e s v e l o c i t y ( I n km/s) a t the top of a r e g i o n . f o l l o w e d a f t e r the c o l o n by the v e l o c i t y g r a d i e n t ( i n km/s/km:if one was u s e d ) . The number f o l l o w i n g . In b r a c k e t s . 1s the d e n s i t y ( i n g c m " ). from a nearby well, and the interpretation of VISP-80 airgun data. These velocities increase eastwards in the continental rise region; a concomitant increase in density is also required. Such increases are due to greater compression and compaction of material at the margin. These compressive structures (asymmetric folds and northeastward dipping imbricate thrusts; figure 6.2) result from the relative 88 northeastward underthrusting of the Juan de Fuca beneath the North America plate. East of OBS #5 where there is no seismic control, gravity modelling requires that the density of upper sedimentary material must decrease; an accompanying decrease in seismic velocities is expected. This is consistent with seismic reflection data which shows the presence of undeformed sediments in Tofino Basin on the continental shelf. The strati f i e d upper crustal velocity sequence which has been derived beneath the deep ocean is similar to that deduced from other studies on the Juan de Fuca plate. In terms of the conventional layered model of the oceanic crust, layer 1, the 1 km thick sequence of marine sediments, ranges in velocity from 1.8 km/s to.2.3 km/s. Layer 2, probably containing pillow basalts and sheeted basalt dikes (Christensen, 1978), ranges in velocity from 4.0 km/s, at the sediment-basement interface, to almost 7.0 km/s a t ~ 7 km depth. A relatively high velocity gradient region (dv/dz=0.76 km/s/km) marks the transition from layer 2 to layer 3, the lower crust. The sediment-basement interface dips eastward at 1.4°; sub-basement layers are assumed to be dipping at the same angle. These layers, separated by changes in velocity gradient, increase in dip to about 3° under the continental rise. Beneath the outer edge of the continental shelf there is a structural change from the relatively high velocity layers of the upper oceanic crust to a lower average velocity block (v»5.3 km/s) of constant gradient, extending down more than 5km from a depth of 4 km. There is no evidence for a sharp transition from one region to the next; the transition is 89 thought to be a comparatively gradual one. The position of this block agrees well with that of a middle Miocene melange unit proposed by Snavely and Wagner (1981), (figure 6.2). F i g u r e 6.2 G e o l o g i c a l i n t e r p r e t a t i o n of the upper c r u s t a l s t r u c t u r e . Upper: A g e o l o g i c a l i n t e r p r e t a t i o n based on U.S.G.S. m u l t i c h a n n e l d a t a (appendix H) a f t e r S n a v e l y and Wagner (1981). The s t r u c t u r a l h o r i z o n s which were p i c k e d on the t r a v e l time s e c t i o n have been r e s c a l e d t o depth by u s i n g t h e r e f r a c t i o n v e l o c i t i e s d e r i v e d i n the VISP-80 d a t a I n t e r p r e t a t i o n ( C h a p t e r 4 ) . Lower: P r e f e r r e d model f o r upper c r u s t a l s t r u c t u r e . Numbers I n d i c a t e s e i s m i c v e l o c i t i e s o f the b l o c k s ( i n km/s); f o l l o w e d , a f t e r the semi-c o l o n , by the v e l o c i t y g r a d i e n t ( i n km/s/km). i f a g r a d i e n t was used. Note t h a t the b l o c k o f 4.6 km/s m a t e r i a l a g r e e s w e l l w i t h the p o s i t i o n of the mass o f melange p r o p o s e d by the U.S.G.S. s t u d y . The lower crust has been modelled as a constant velocity gradient region, extending down to 9 km below sea floor in the deep ocean, and may consist of metadolerite sheeted dikes underlain by metagabbro and olivine-pyroxene gabbros (Salisbury and Christensen, 1978). This velocity structure 90 remains constant over the entire marine profile, suggesting that the overlying melange material may have been formed as upper layers were scraped off the descending oceanic plate. Absolute plate motion studies (Minster and Jordan, 1978) indicate that the Juan de Fuca plate is subducting beneath, and also being overriden by the North American plate. The melange unit may represent a downward continuation of the "accretionary wedge" envisaged by the imbricate thrust model for subduction zones, (Beck, 1972; Seely et a l . 1974). However, analysis of d r i l l i n g data at convergent margins (e.g. Moore et a l . 1982) does not indicate incorporation of basaltic basement into an accretionary wedge, either by offscraping. or underplating 1 of material. It is conceivable that the young age of the Juan de Fuca plate together with the overriding motion of the North America plate produce a unique tectonic regime where there is accretion of sub-sedimentary material at the margin. Complete detachment of the upper crust from the descending plate below, if i t occurs, must be a relatively recent process; 3 km of crustal material would produce a mass 900 km2 in cross-section in 1 Ma. It is more probable that the melange material merely indicates a resistance to subduction by the continental crust, the upper oceanic crustal layers being compressed and deformed whilst s t i l l remaining attached to the Juan de Fuca plate. 'In the process of underplating, sedimentary material is added to the accretionary wedge from below, in addition to being scraped off in the continental rise region. 91 The Moho has been modelled as a change in velocity gradient from 0.2 km/s/km in the lower crust to 0.01 km/s/km in the upper mantle. Resolution of Moho structure is poor, but no velocity discontinuity at the boundary is required by the data, although a small change of the order of 0.2 km/s is s t i l l compatible with the data. There is an increase in Moho dip under the continental rise from about 1° to 6°. The convergent margin structure between Juan de Fuca and America plates is anything but simple. It would have been possible to satisfy a l l of the seismic, gravity, and additional information more closely by introducing, sufficient structural complexity into the model. This approach was rejected. Obtaining features common to a l l physically plausible structures is considered to be more valuable than exactly f i t t i n g data with large associated errors. Consequently, the interpreted model (figure 6.1) is not a representation of the real earth, but rather an illustration of certain basic structural features associated with this convergent tectonic environment. 92 BIBLIOGRAPHY Adams,J, and R. Reilinger, Time behaviour of vertical crustal movements measured by relevelling in North America: a geologic perspective. In Proc. 2nd Int. Symp. on problems related to redefinition of North American Vertical Geodetic Networks (NAD). Can. Inst. Surveying, Ottawa, Ont., pp. 327-339, 1980. Aki,K., and P.G. Richards, Quantitative Seismology - Theory and Methods. W.H. Freeman and Company, San Francisco. 2 vols. 1980. Ando,M., and E. Balazs, Geodetic evidence for aseismic subduction of the Juan de Fuca plate. J. Geophys. Res., 84, 3023-3028, 1979. Atwater,T., Implications of plate tectonics for the Cenozoic tectonic evolution of western North America. Geol. Soc. Am. Bull., 81, 3513-3536, 1970. Au,C.Y.D., Crustal structure from an ocean bottom seismometer survey of the Nootka fault zone. Ph.D. Thesis, Univ. of British Columbia, Vancouver, 131 pp. 1981. Au,C.Y.D., and R.M. Clowes, Crustal structure from an OBS survey of the Nootka fault zone off western Canada. Geophys. J. R. Astron. Soc, 168, 27-48, 1982. Barr,S.M., Structure and tectonics of the continental slope west of southern Vancouver Island. Can. J. Earth. Sci., 11, 1187-1199, 1974. Bates,A.C, and E.R. Kanasewich, Inversion of seismic travel times using the tau method. Geophys. J. R. Astron. Soc, 47, 59-72, 1976. Beck,R.H., The oceans, the new frontier in exploration. Aust. Pet. Ex. Assoc J., 12, 1-21, 1972. Berry,M.J., Depth uncertainties from seismic f i r s t - a r r i v a l refraction studies. J. Geophys. Res., 76, 6464-6468, 1971. Bessonova,E.N., V.M. Fishman, V.Z. Ryaboyi, and G.A. Sitnikova, The tau method for inversion of travel times -I. Deep seismic sounding data. Geophys. J. R. Astron. Soc, 36, 377-398, 1974. Braile,L.W., and R.B. Smith, Guide to the interpretation of crustal refraction profiles. Geophys. J. R. Astron. Soc, 40, 145-176, 1975. Carson,B., Acoustic stratigraphy, structure, and history of Quaternary deposition in Cascadia Basin. Deep Sea Res., 20, 387-396, 1973. Cerveny,V., I.A. Molotkov, and J. Psencik, Ray method in Seismology, Universita Karlova, Praha, 214 pp, 1977. Cerveny,V., Ray theoretical seisograms for laterally inhomogenous structures. J. Geophys. Res., 46, 335-342, 1979. Chapman,C.H., A new method for computing synthetic seismograms. Geophys. J. R. Astron. Soc, 54, 481-518, 1978. Chase,R.L., D.L. T i f f i n , and J.wl Murray, The western Canadian continental margin. In Canada's Continental Margins and Offshore Petroleum Exloration, edited by C. Yorath, E. Parker and D. Glass. Can. Soc. Pet. Geol., Memoir 4, 701-721, 1975. Cheung,H.P.Y., Crustal structure near Explorer Ridge: ocean bottom seismometer results parallel to Revere-Dellwood fracture zone. M.Sc. Thesis, Univ. of British Columbia, Vancouver, 80 pp, 1978. Choy,G.L., Theoretical seismograms of core phases calculated by frequency-dependent f u l l wave theory, and their interpretation. Geophys. J. R. Astron. Soc, 51 , 225-312, 1977. Christensen,N.I., Ophiolites, seismic velocities, and oceanic structure. Tectonophysics, 47, 131-157, 1978. Christensen,N.I., and M.H. Salisbury, Structure and constitution of the lower oceanic crust. Rev. Geophys. Space Phys., 13, 57-86, 1975. Clayton,R.W., and G.A. McMechan, Inversion of refraction data by wave f i e l d continuation. Geophysics, 46, 860-868, 1981. Clee,T.E., K.G. Barr, and M.J. Berry, Fine structure of the crust near Yellowknife. Can. J. Earth. Sci., 11, 1534-1549, 1974. Clowes,R.M., and S. Knize, Crustal structure from a marine seismic survey off the west coast of Canada. Can. J. Earth. Sci., 16, 1265-1280, 1979. Couch,R., and D. Braman, Geology of the continental margin near Florence, Oregon. Oregon Geology, 41, No. 11, 1979. Crosson,R.S., Small earthquakes, structure and tectonics of the Puget Sound region. Bull. Seismol. Soc. Am., 62, 1133-1172, 1972. 94 Davydova,N.I., Possibilities of the DSS technique in studying properties of deep-seated seismic interfaces, in Seismic Properties of the Mohorovicic Discontinuity, edited by N.I. Davydova, National Technical Information Service, U.S. Dept. of Commerce, Springfield, Va, 4-22, 1975. Detrick,R.S., and G.M. Purdy, The crustal structure of the Kane fracture zone from seismic refraction studies. J. Geophys. Res., 85, 3759-3777, 1980. Dix,C.H., Seismic velocities from surface measurements. Geophysics, 20, 68-86, 1955. Fowler,C.M.R., and CE. Keen, Oceanic crustal structure - Mid-Atlantic Ridge at 45° N. Geophys. J. R. Astron. Soc, 56, 219-226, 1978. Fuchs,K., On the properties of deep crustal reflectors. Geophysics, 35, 133-149, 1969. Fuchs,K., and G. Muller, Computation of synthetic seismograms with the reflectivity method and comparison with observations. Geophys. J. R. Astron. Soc, 23, 417-433, 1971 . Garmany,J., J. Orcutt, and R. Parker, Travel time inversion: a geometrical approach. J. Geophys. Res., 84, 3615-3622, 1979. Hale,L.D., and G.A. Thompson, The seismic reflection character of the continental Mohorovicic discontinuity. J. Geophys. Res., 87, 4625-4635, 1982. Hamilton,E.L., Sound velocity-density relations in sea-floor sediments and rocks. J. Acoust. Soc. Am., 63, 366-377, 1978. Healy,J.H., Crustal structure along the coast of California from seismic refraction measurements. J. Geophys. Res., 68, 5777-5787, 1963. Herglotz,G., Uber das benndorfsche problem der fortpflanzungsgeschwindigkeit der erdbebenstrahlen. Phys. Z., 8, 145-147, 1907. Horn,J.R., A snapshot of the Queen Charlotte Fault zone obtained from P-wave refraction data. M.Sc. Thesis. Univ. of British Columbia,Vancouver, 79 pp, 1982". Hubbert,M., A line integral method of computing the gravimetric effects of 2-D masses. Geophysics, 13, 215-225, 1948. Hyndman,R.D., Heat flow measurements in the inlets of southwestern British Columbia. J. Geophys. Res., 81, 337-349, 1976. 95 Hyndman,R.D., R.P. Riddihough, and R. Herzen, The Nootka Fault zone - A new plate boundary off western Canada. Geophys. J. R. Astron. Soc, 58, 667-683, 1979. Hyndman,R.D., and D.H. Weichert, Seismicity and rates of relative motion on the plate boundaries of western North America. Geophys. J. R. Astron. Soc, in press. 1982. Johnson,L.E., and F. Gilbert, Inversion and inference for teleseismic ray data, in Methods in Computational Physics, edited by B.A. Bolt, Academic, New York, 11, 231-266, 1972. Johnson,R.V.II, C.R.B. Lister, and B.T.R. Lewis, A direct recording ocean bottom seismometer, Mar. Geophys. Res., 3, 65-85, 1977. Jurkevics,A., R. Wiggins, and L. Canales, Body-wave inversion using travel time and amplitude data. Geophys. J. R. Astron. Soc, 63, 75-93, 1980. Keen,C.E., and R.D. Hyndman, Geophysical review of the continental margins of eastern and western Canada. Can. J. Earth. Sci., 16, 712-744, 1979. Kanestr0m,R., 0. 0vreb0, Seismic measurements of large underwater shots, Geophysia, Geophys. Soc. of Finland, . 15, 215-229, 1978. Kennett,B.L.N., A Comparison of travel-time inversions. Geophys. J. R. Astron. Soc, 44, 517-536, 1976. Kennett,B.L.N., and J.A. Orcutt, A comparison of travel time inversions - marine refraction profiles. J. Geophys. Res., -81 , 4061-4070, 1976. Krebes,E.S., and F. Hron, Comparison of synthetic seismograms for an elastic media by asymptotic ray theory and the Thomson-Haskell Method. Bull. Seismol. Soc. Am., 71, 1463-1468, 1980. Levy,S., and R.M. Clowes, Debubbling: a generalized linear inverse approach. Geophys. Prosp., 28, 840-858. 1980. Lister,C.R.B., and B.T.R. Lewis, An ocean-bottom seismometer suitable for arrays. Deep Sea Res., 23, 113-124, 1976. McMechan,G.A., and W.D. Mooney, Asymptotic ray theory and synthetic seismograms for laterally varying structures. Theory and application to the Imperial Valley, California. Bull. Seismol. Soc. Am., 70, 2021-2035, 1980. McMechan,G.A., and R.A. Wiggins, Depth limits in body wave inversion. Geophys. J. R. Astron. Soc, 28, 459-473, 1972. 96 Minster,J.B., and T.H. Jordan, Present day plate motions. J. Geophys. Res., 83, 5331-5354, 1978. Molnar,P., Gravity anomalies and the origin of the Puerto Rico Trench. Geophys. J. R. Astron. Soc, 51, 701-708, 1977. Moore,J.C, J.S. Watkins, T.H. Shipley, K.J. McMillen, S.B. Bachman, and N. Lundberg, Geology and tectonic evolution of a juvenile accretionary terrane along a truncated convergent margin: Synthesis of results from Leg 66 of the Deep Sea D r i l l i n g Project, southern Mexico. Geol. Soc. Am. Bull., 93, 847-861, 1982. Morgan,W., Rises, trenches, great faults and crustal blocks. J. Geophys. Res., 73, 1959-1982, 1968. Muller,J.E., Evolution of the Pacific Margin, Vancouver Island, and adjacent regions. Can. J. Earth. Sci., 14, 2062-2085, 1977. Nafe,J.E., and C L . Drake, Physical properties of marine sediments, in The Sea, edited by M.N. H i l l , Wiley, New York, 3, 794-815, 1963. O'Brien,P.N.S., The efficient use of large charges, in Seismic Refraction Prospecting, edited by A.W. Musgrave. Soc. Ex. Geophys., 152-170, 1967. Oldenburg,D.W., The inversion and interpretation of gravity anomalies. Geophysics, 39, 527-536, 1974. Orcutt,J., K. MacKenzie, and J.S. McClain, The role of X(p) constraints in linear, extremal inversion of explosion profile data. Bull. Seismol. Soc. Am., 70, 2103-2116, 1980. Raitt,R.W., The crustal rocks, in The Sea, edited by M.N. H i l l , Wiley, New York , 85-102, 1963. Riddihough,R.P., and R.D. Hyndman, Canada's active western margin - the case for subduction. Geoscience Canada, 3, 269-278, 1976. Riddihough,R.P., A model for recent plate interactions off Canada's west coast. Can. J. Earth. Sci., 14, 384-396, 1977. Riddihough,R.P., Gravity and structure of an active margin -British Columbia and Washington. Can. J. Earth. Sci., 16, 350-363, 1979. Riddihough,R.P., Contemporary movements and tectonics on Canada's west coast: a discussion. Tectonophysics, 86, 319-341, 1982. Riddihough,R.P., M.E. Beck, R.L. Chase, E.E. Davis, R.D. 97 Hyndman, S.H. Johnson, and G.C. Rogers. Geodynamics of the Juan de Fuca plate in Geodynamics of the Eastern Pacific, Caribbean and Scotia Arcs, edited by R. Cabre, editor, Geodynamic series of the AGU (in press), 1982. Richards,P.G., Calculation of body waves for caustics and tunnelling core phases. Geophys. J. R. Astron. Soc, 35, 243-264, 1973. Richards,P.G., Theoretical seismic wave propagation. Rev. Geophys. Space Phys., 17, 312-328, 1979. Rogers,G.C, Seismotectonics of British Columbia. Ph.D. Thesis, Univ. of British Columbia, Vancouver, 224 pp., 1982 (submitted). Salisbury,M.H., and N.I. Christensen, The seismic velocity structure of a traverse through the Bay of Islands ophiolite complex, Newfoundland. J. Geophys. Res., 83, 805-817, 1978. Savage,J.C, M. Lisowski, and W.H. Prescott, Geodetic strain measurements in Washington. J. Geophys. Res., 86, 4929-4940, 1981. Seely,D.R., P.R. Vail, and G.G. Walton, Trench slope model, in The Geology of Continental Margins, edited by C.A. Burk and C.L. Drake, Springer-Verlag, New York, 249-260, 1974. Shor.G.G.,, Refraction and reflection techniques and procedure, in The Sea, edited by M.N. H i l l , Wiley (Interscience), New York, 30-38, 1963. Shouldice,D.H., Geology of the western Canadian continental shelf. Bull. Can. Pet. Geol., 19, No. 2, 405-436, 1971. Scholl,D.W., Sedimentary sequences in north Pacific trenches. In the Geology of Continental Margins, edited by C.A. Burk, and CL. Drake, Springer-Verlag, New York, 493-504, 1974. Snavely,P.D., Geophysical data collected on continental margin midway between Washington State and Vancouver, B.C. along Line 19, USGS, S.P. Lee Cruise 3-76: U.S. Geol. Surv. Open File Rep., 81, 895-902, 1981. Snavely,P.D., and H.C Wagner, Geologic cross-section across the continental margin off Cape Flattery, Washington, and Vancouver Island, British Columbia. U.S. Geol. Surv. Open File Rep., 81, 978-984, 1981. Spudich,P., and J. Orcutt, A new look at the seismic velocity structure of the oceanic crust. Rev. Geophys. Space Phys., 18, 627-645, 1980. Talwani,M., L. Worzel, and M. Landisman, Rapid gravity 98 computations for 2-dimensional bodies with application to the Mendocino submarine fracture zone. J. Geophys. Res., 64, 49^59, 1959a. Talwani,M., G.H. Sutton, and J.L. Worzel, A crustal section across the Puerto Rico Trench. J. Geophys. Res., 64, 1545-1555, 1959b. Tiffin,D.L., B.E. Cameron, and J.W. Murray, Tectonics and depositional history of the continental margin off Vancouver Island, British Columbia. Can. J. Earth. Sci., 9, 280-296, 1972. Treitel,S., and E.A. Robinson, The design of high resolution digital f i l t e r s , IEEE Transactions in Geoscience and Electronics GE-4, 25-38, 1966. Tucker,P.M., and H.J. Yorston, P i t f a l l s in Seismic Interpretation. Soc. Ex. Geophys., Monograph Series, No. 2, 1973. Vine,F.J., Spreading of the ocean floor - new evidence. Science 154 (3755), 1405-1411, 1966. Whittall,K.P., and R.M. Clowes, A simple, efficient method for the calculation of travel times and ray paths in laterally inhomogeneous media. J. Can. Soc. Ex. Geophysi, 15, 21-29, 1979. Wiechert,E., Ein astatische pendel hoher empfindlichkeit zur mechanischen registrierung von erdbeben. Beitr. Geophys. 6, 435-450, 1904. Wiggins,R.A., and D.V. Helmberger, Synthetic seismogram computation by expansion in generalized rays. Geophys. J. R. Astron. Soc, 37, 73-90, 1974. Wiggins,R.A., Body wave amplitude calculations, II. Geophys. J. R. Astron. Soc, 46, 1-10, 1976. Willis,H.F., Underwater explosions, time interval between successive explosions, British Report WA-47-21. 1941. Wood,L.C, R.C. Heiser, S. Treit e l , and P.L. Riley, The debubbling of marine source signatures. Geophysics, 43, 715-729, 1978., 99 Appendix A WKBJ and ART Algorithms A.1 WKBJ The WKBJ synthetic seismogram method (Chapman,1 1978) is based upon an approximate solution to the wave equation. The computation of the seismogram reduces to the solution of an ordinary differential equation plus the evaluation of inverse transforms, a frequency and a wave number integral. The innovation in Chapman's method is the technique of evaluating the frequency integral f i r s t and keeping the wave number real. This method is relatively inexpensive. The WKBJ approximation (named after Wentzel, Kramers, Brillouin and Jeffreys) An approximate solution to the second order equation d2tf + u2s2<t> =0 (1 ) dx 2 where u is large and positive, and s = s(x) is required. ±i(JT (x) Try solutions of the form <p = e iwr" - u2(T')2 + w 2s 2 = 0 Approximation 1: neglect urn implies: T " ' < V ± S ( X ) and, r(x)~± |s(x)dx but, J this implies: T " « ~ ' ± S ' Approximation 2: that (T' ) 2 = s 2 ± is'/w i.e. T' = ±s + is'/2soj 100 then, r(x) = ± s(x)dx + (i/2co)lns The corresponding solution for <p i s : A B <f>(x) ~ exp( icj fsdx) + exp( -iu> fsdx) s 1 / 2(x) J s" 2(x) J where: A and B are constants This is valid i f |s'/a>| << |s 2|, Which amounts to validity i f | s' x X | << 2ir | s | , and the change in s(x) in one X must be less than s i t s e l f . This method enables an analytical approximation to the wavenumber integral to be made, avoiding the use of slow and costly numerical solutions in cases where no exact analytical solution exists. A.2 ART The asymptotic ray theory synthetic seismogram algorithm (McMechan and Mooney, 1980) employs, as the name implies, asymptotic ray theory (ART). (see Cerven£ et a l . , 1977; Cerveny, 1979). The method is approximate, but has the advantage of being valid for laterally varying structures. 101 Asymptotic ray theory: Assuming that a time-harmonic solution to the linearised equation of motion: S2W p = (X + n) V(V.W) + jzV2W + VX(V.W) + 8t 2 Vu x(VxW) + 2 ( V M . V ) W (1 ) can be expressed in inverse powers of frequency, a, we may write an expression for the displacement vector: W = exp[icj(t-r) ] Z (ia>)"kW. (2) k=o where T and are independent of o> and t. = kth amplitude coefficient of the ray series. It is assumed that the ray series (2) exists and is asymptotic to the exact solution of the equation ( l ) . The function must be analytic and thus the ray expansion (l) is not valid in the vicinity of those points where the T-X curve associated with the propagating wave has end points, cusps, tangent points with the T-X curve of another wave, or, generally, discontinuous derivatives. Due to the fact that the solution (2) is expressed as an asymptotic expansion in inverse powers of frequency, i t yields best results for high frequencies. Except in the close neighbourhood of singular points, the error which arises in keeping only the f i r s t few terms in the series tends to zero as u) increases. In many cases i t is sufficient to consider only the leading term in (2). Then: W = exp[iw(t-r)]W 0 1 0 2 This is the zero-order solution, which is the solution obtained according to the principles of geometrical optics. The zero-order ART algorithm may be expressed by: A = AoL-'IJ k; J 1 where A = the total complex amplitude associated with a ray A 0 = the i n i t i a l amplitude L = the geometrical spreading, and II = the serial product over a l l the j complex plane-wave transmission and reflection coefficients (k) along the ray path. The synthetic arrival is constructed by performing a ray trace along which travel time, distance and D k: are summed. J J The amplitude computation proceeds by calculating the area of an element of wave-front surface by determining the distances between points of equal travel time on adjacent rays. Limitations of WKBJ Algorithm: 1) Not s t r i c t l y correct for waves turning in the vicinity of a low velocity zone or a velocity discontinuity. 2) Diffractions beyond triplications or precritical reflections from velocity gradient zones are not correctly calculated. When the geometrical amplitude on a wavefront suddenly changes as in a true shadow, or at a rapid change in velocity gradient, there are two diffraction effects (Chapman, personal communication 1982); 1 03 (a) The geometrical e f f e c t of having an incomplete wavefrbnt; completely analagous to c l a s s i c a l Fresnel d i f f r a c t i o n and accurately described by the WKBJ seismogram method. (b) The non-geometrical d i f f r a c t i o n e f f e c t depending on the frequency dependent boundary conditions; not described by WKBJ method. 3) There i s an end-point error from the bottom of the model, but this can be made negligible by making, the lowest model boundary s u f f i c i e n t l y deep. Limitations of ART Algorithm 1) The zero-order approximation, by d e f i n i t i o n , excludes a r r i v a l s that originate in the higher order terms, such as head waves. 2) The method i s based on ray theory and thus does not produce wave e f f e c t s , such as d i f f r a c t i o n s . There are two sources of differences between the results obtained using ART and WKBJ. 1) The d i f f e r e n t methods of vel o c i t y interpolation employed: ART uses cubic splines whereas WKBJ employs a line a r interpolation scheme. 2) Different approaches to the physics of energy propagation: WKBJ i s an approximate wave-type solution, ART i s an approximate ray-type solution. A comparison of the e f f e c t s produced by the d i f f e r e n t algorithms i s given in section 4.4. 104 Appendix B Instrumentation and data digitization The ocean bottom seismographs used were of the free f a l l , pop-up type similar to those described by Lister and Lewis (1976) and Johnson et a l . , (1977). The spherical OBS package descends at 3ms" 1 with the 12 cm thick, 300 kg, concrete anchor attached and is later (usually less than 15 days) released by programmed or acoustic command. The instrument records signals at a tape speed of 1 mm/s in continuous direct AM mode, from one vertical and two horizontal 4.5 Hz geophones with a 20 Hz time code in the f i r s t channel. The overall frequency response of the system is bandlimited between 2 and 100 Hz. Square root signal compression increases the dynamic range, but gives poor signal reproduction f i d e l i t y and only approximate amplitudes. The OBS data tapes are then transcribed from AM to FM at a playback speed of 15/16 inch per second (ips) and a recording speed of 15 ips. The FM tape is then played back at a speed of 15/16 ips for digitization at a sampling rate of 312.5 sps. The overall effect of this process is to give a real-time data sampling rate of approximately 200 sps. The actual sampling rate varies with time and the different instruments due to minor variations in tape drive motor speed and a small degree of tape stretching. In practice, the sample rate fluctuations lead to timing errors of less than 0.003 s, i.e. insignificant in comparison with origin time and picking errors. 1 0 5 Appendix C Chevron Multichannel Data: In 1972, Chevron Standard Limited ran a number of multichannel, seismic reflection profiles perpendicular to the continental slope from the deep water to part of the way up the slope. One of these profiles (113-H) crosses the VISP-80 profile-I at an oblique angle, diverging by a maximum of 17 km laterally at the eastern end. Figure C.1 shows the stacked F i g u r e C.1 Chevron S t a n d a r d m u l t i c h a n n e l d a t a . The s t a c k e d r e c o r d s e c t i o n Is c o m p i l e d from 2400X. 24 c h a n n e l , common r e f l e c t i o n p o i n t d a t a . Automatic g a i n c o n t r o l , bandpass f i l t e r i n g and a s t a n d a r d d e c o n v o l u t i o n p r o c e s s have been a p p l i e d . H o r i z o n A Is the boundary between the 1.8 km/s and 2.3 km/s r e g i o n s used i n the s e i s m i c m o d e l l i n g . H o r i z o n B i s the r e f l e c t i o n c o r r e s p o n d i n g t o the basement. H o r i z o n C i s the c o n t i n u a t i o n of the basement beneath the c o n t i n e n t a l r i s e ; d i s p l a c e d upwards due t o v e l o c i t y p u l l - u p . X. V. and Z a r e the p o s i t i o n s at which NMO. v e l o c i t i e s were c a l c u l a t e d by Chevron S t a n d a r d L t d . 106 record section compiled from 2400% common reflection point data. As has been noted elsewhere (Clowes and Knize, 1979), the data quality is excellent. To the west there is an upper sequence of plane layered, high amplitude, reflections to a time of approximately 4.8 s two way travel time (T.W.T.T.), (horizon A, figure C.1). This corresponds to the 1.8 km/s velocity region used in the seismic modelling. Below these layers the character of the section changes, the layering is less distinct and the amplitudes are not as pronounced, implying a greater degree of compaction of the material; this is the 2.3 km/s velocity region used in the modelling. A strong, irregular reflection which is clearly identified with the basaltic basement is seen at approximately 5.4 s (horizon B, figure C.1). The appearance of reflectors below this horizon, at approximately 7 s, is due to the presence of water bottom and sedimentary layer multiples. The oceanic basement is clearly seen dipping to the east, whilst the overlying sediments here have a fan structure (the upper layers being parallel to the sea floor, the lower ones being parallel to the dipping basement). At the western edge of the continental rise there is a radical change in the character of the section. This is due to the presence of fault-bounded ridges, behind which sediments are ponded (see figure 1.3). The arching of reflecting horizons below the surface expression of the ridges is partly a real effect and partly a result of velocity pull-up (see Tucker and Yorston, 1973), due to replacement of part of the 107 water column with sub-bottom material. The continuation of basement beneath the continental rise can be seen (horizon C, figure C.1 ), displaced upwards due to the velocity pull-up. If the basement dip is assumed to be constant along the whole section then a value for the average sediment velocity under the rise may be obtained from the degree of pull-up ( 0.3 s T.W.T.T. in this case). The derived velocity is 2.0 km/s; consistent with airgun data obtained over the rise (see section 4.7.1). TIME X Y Z 2.8 4900 3.0 4900 5000 3.2 4950 5100 3.4 4900 5000 5200 3.6 4950 5000 5200 3.8 5000 5200 5400 4.0 505.0 5250 5500 4.2 5150 5350 5700 4.4 5200 5500 5800 4.6 5300 5600 5900 5.0 5400 5700 6100 5.2 5600 6000 6500 5.4 5750 6200 6600 5.6 5900 6400 6800 5.8 6100 6800 7000 6.0 6400 6700 7200 6.5 7200 7500 7600 7.0 8000 8000 8000 8.0 9000 9000 9000 Table C.1 NMO velocities at positions X, Y, and Z (figure C.1). Time is T.W.T.T.(s); velocities in ft/sec. 108 The normal move-out velocities employed (table C.1 ) in the data processing were used to generate a v-z curve by use of the Dix (1955) equation: V 2 = ( V 2 t - V 2 t )/( t - t ) n r.m.s,n n r.m.s,n-1 n-1 n n-1 where: V = interval velocity for nth layer n V = r.m.s. velocity for nth reflection, and r.m.s,n t = travel time to nth reflection n V E L O C I T Y 0 1 2 3 ( k m / s ) 4 5 6 F i g u r e C.2 V-Z s t r u c t u r e d e r i v e d from NMO v e l o c i t i e s . S o l i d l i n e s r e p r e s e n t the v e l o c i t y - d e p t h c u r v e s o b t a i n e d from NMO v e l o c i t i e s a t p o s i t i o n s X. Y. and Z ( f i g u r e C.1). The dashed l i n e Is the p r e f e r r e d v e l o c i t y s t r u c t u r e J u s t west of the c o n t i n e n t a l r i s e . Note the d i v e r g e n c e o f v a l u e s below the basement l e v e l (2 km), which r e f l e c t s the company's I n t e r e s t In sediment l a y e r I n f o r m a t i o n . 109 The v-z curves (figure C.2.) were generated by adding the terms Vt down to a given reflector, n. As can be seen from the figure, the agreement between the model velocities and the interval velocities is good. Although use of the Dix formula to determine the v-z structure is an unstable process, the consistency between used and derived values is encouraging. The divergence of values below the basement reflects the lack of resolution of NMO curves at greater depths and the company's interest in sediment layer informat ion. 1 10 Appendix D Shell Cygnet Well In 1968 Shell Canada limited d r i l l e d an exploratory well in the Tofino basin: The Shell Anglo Cygnet Well (Shouldice, 1971). At 48* 19' 40" N, 125* 43' 57" W (see figure H.l) the site is 18 km south-east of OBS #5. In the spring of 1969 gamma ray and sonic logs were taken down the 2 km deep well. The velocity-depth curve calculated from the interval transit times is given in figure D.1. The velocity variation agrees well with that derived from the OBS #5 airgun data. Geological horizons proposed by Snavely and Wagner (1981) have been roughly correlated with features in the sonic log. VELOCITY (km/s ) 1 2 3 4 C\2 F i g u r e 0.1 S o n i c l o g d e r i v e d v e l o c i t y s t r u c t u r e and g e o l o g i c a l i n t e r p r e t a t i o n . S o l i d l i n e i s the v e l o c i t y - d e p t h c u r v e c a l c u l a t e d from i n t e r v a l t r a n s i t times o b t a i n e d down the S h e l l Anglo Cygnet w e l l . Dashed l i n e i s the s t r u c t u r e d e r i v e d from the OBS *5 a i r g u n d a t a . The g e o l o g i c a l i n t e r p r e t a t i o n on the r i g h t i s t h a t p r o p o s e d by S n a v e l y and Wagner (1981). 1 1 1 Appendix E Structure of the Moho The Mohorovicic discontinuity is most often interpreted as a step-function velocity boundary, although a change in velocity gradient (Au, 1981) and a laminated character (Fuchs, 1969) have also been considered. Near vertical deep crustal reflection profiling is the most diagnostic method of examining the fine structure of the crust-mantle transition zone (Hale and Thompson, 1982) due to the large data density and high input frequencies used. Analysis of refraction data, involving a turning ray interpretation of f i r s t arrivals, is very poor at resolving features (e.g. discontinuities) in the v-z structure (see section 2.2), but clear sub-critical reflections are diagnostic in this regard. The variation in character of a record section depending upon the form of the crust-mantle transition has been discussed by Braile and Smith (1974) and Davydova (1975). The various types of Moho considered are illustrated in figure E.1., whilst the effects of these structures on synthetic data is summarised in table E.1. 1 1 2 S t r u c t u r e Type E f f e c t on s e i s m i c d a t a S t e p - f u n c t i o n b o u n d a r y P r e - c r i t i c a l r e f l e c t i o n a m p l i t u d e s i n c r e a s e w i t h v e l o c i t y c o n t r a s t , w h i l e t h e c r i t i c a l p o i n t moves c l o s e r t o t h e s o u r c e . G r a d i e n t zone P r e - c r i t i c a l r e f l e c t i o n a m p l i t u d e s v e r y low. S t e p w i t h g r a d i e n t L a r g e a m p l i t u d e h e a d waves from below t h e b o u n d a r y . S t e p w i t h d e c r e a s e below A m p l i t u d e o f h e a d waves f r o m below t h e b o u n d a r y v e r y low. L a m i n a t e d t r a n s i t i o n zone L a r g e a m p l i t u d e s u b -c r i t i c a l r e f l e c t i o n s f o r some f r e q u e n c i e s , due t o c o n s t r u c t i v e i n t e r f e r e n c e . T a b l e E . 1 . The e f f e c t s o f s e i s m i c d a t a . Moho s t r u c t u r e t y p e on Figure E.I Poss ib le v e l o c i t y va r i a t i ons at the Moho. (1) Step-functIon v e l o c i t y boundary (2) High ve loc i ty - g rad ient t r a n s i t i o n zone (3) Ve loc i t y contrast with gradient region below (4) Ve loc i t y contrast with v e l o c i t y decrease below (5) Thin laminated layer t r a n s i t i o n zones. 1 1 3 The resolution of Moho structure that may be obtained with marine refraction data is very poor. This is due to the inherent data errors and the characteristically long source signature (see appendix F). It should also be remembered that if the Moho is interpreted as a change in velocity gradient ( from high to low ) , enabling only turning rays above the boundary to be interpreted, then each data set samples the Moho at only one place. Thus, as noted (section 4.5), 3 OBSs will weakly constrain the Moho depth at only 3 places. Lateral heterogeneities in Moho depth of the type which have been inferred from high density reflection data (e.g. Clee et a l . , 1974) may well be present but are not detectable with refraction data. 1 14 Appendix F Bubble Pulse Oscillations The long source wavelet characteristic of marine seismic work is a product of bubble pulse oscillations. The problem is caused by successive oscillations of the gas bubble generated by the energy source. Each cycle of the oscillating bubble corresponds to a signal propagating outward. The source wavelet recorded is then a train of wavelets generated by the individual cycles. The number of pulses and their periods are primarily a function of detonation depth and of the energy released during creation of the bubble. The length of the compound wavelet depends on the number of expand-collapse cycles of substantial energy performed by the bubble before i t collapses completely or vents to the atmosphere. The elimination of bubble pulse effects by compression and simplification of the compound wavelet signature has been attempted by using various methods. Wood, Heiser, Treitel and Riley (1978) employ Wiener shaping f i l t e r s (Treitel and Robinson, 1966) in their algorithm. This method has the disadvantage that the choice of f i l t e r length (an unknown parameter) is c r i t i c a l . Levy and Clowes (1980) have used a generalized linear inverse approach which has the advantage that the solution accuracy may be made consistent with observational errors. The disadvantage of this procedure is that i t is computationally expensive. The period, T, of the f i r s t oscillation of the gas bubble 1 15 generated by the explosion may be related to the shot depth, d, and the charge weight, W, by the Rayleigh-Willis bubble formula (Willis, 1941): 2.13 W,/3 (d + 10) 5' 6 where W = energy of the explosion expressed as explosive weight in TNT equivalents. The maximum possible source signal strength is obtained at optimum depth: where the water reverberation frequency is equal to the bubble oscillation frequency. This is the criterion used to compute the most desirable detonation depths for the various shot sizes. 1 16 Appendix G 2-D Gravity Modelling Algorithm This algorithm computes the gravitational attraction due to individual polygons, of given density, at different depths and calculates the total force per unit mass which would be measured at any point on the surface. Any two-dimensional structure may be modelled to any required degree of accuracy, by considering a sufficiently large number of polygons. Analytic expressions for the vertical and horizontal components of gravitational attraction due to an arbitrary polygon are derived below: I 2 Figure G.I Geometrical elements Involved In the gravitational a t t r a c t i o n of an n-sided polygon (after Talwanl et a l . . 1959). ABCDEF (figure G.I) is a polygon with n sides and P is 1 17 the point at which the g r a v i t a t i o n a l a t t r a c t i o n i s to be computed. Hubbert (1948) shows that the v e r t i c a l component of the a t t r a c t i o n due to such a two-dimensional body i s equal, at the o r i g i n , to 2GpJzd0 where the l i n e integral i s taken along the periphery of the body, G i s the universal constant of gr a v i t a t i o n , and p i s the volume density of the body. The corresponding expression for the horizontal component of g r a v i t a t i o n a l a t t r a c t i o n i s given by 2GpJxdc? (Talwani et a l . , 1959a). Computation of the contribution to J*z 66 from BC: z = xtant? for any ar b i t r a r y point R on BC, also: z = (x-a ) tan# from (1) and (2) I a tandtan<> z = - i i -tan0 - tanc9 i or: n a tandtan^ z66 = j ' i i 66 - z tan0 - tan0 B C s i m i l a r l y B r c a tan# j tan<p - tar xd(9 = I — i - i dB = x t nc? B C ~B i The v e r t i c a l component of g r a v i t a t i o n a l a t t r a c t i o n , V, and the 1 18 horizontal component, H, due to the whole polygon, are then given respectively by: n V = 2Gp L Zi Ul n H = 2Gp Z Xi i=l - the summations being made over the n sides of the polygon. The integrals involved in the expressions for Zi and Xi may be easily solved analytically; for details see Talwani et a l . , (1959a). 1 19 Appendix H U.S.G.S. Multichannel Data A 24-channel seismic-reflection profile was collected aboard the U.S. Geological Survey R/V S.P. Lee (Snavely, 1981), and a geological cross section has been constructed, based on the section (Snavely and Wagner, 1981). The profile extends northeastward (figure H.1) from the abyssal plain near latitude 48' 00' N and longitude 126' 43' W, passes 1 km south of the Shell Anglo Cygnet exploratory well (Appendix D), and terminates at a point on the coast of Vancouver Island just southeast of Nitinat Lake. F i g u r e H.1 L o c a t i o n map o f the U.S.G.S. m u l t i c h a n n e l s e c t i o n . The 24 channel s e i s m i c r e f l e c t i o n p r o f i l e e x tends n o r t h e a s t w a r d from A t o A". The p o s i t i o n s o f the t h r e e OBSs, #\, 13. and 15 a r e i n d i c a t e d by the. I n v e r t e d t r i a n g l e s . Note the p o s i t i o n o f the S h e l l Anglo Cygnet w e l l ( s e e Appendix 0 f o r d e t a i l s ) . 120 The section (figure H.2) of five airguns totalling recording system consisted of Global Universal Science instrument. The lower diagram was obtained using a tuned array 1326 i n 3 as a sound source. The a 24 - channel streamer and a model 4300 d i g i t a l recording in figure H.2 is the geological F i g u r e H.2 U.S.G.S. m u l t i c h a n n e l d a t a and g e o l o g i c a l I n t e r p r e t a t i o n . Upper: 24 channel s e i s m i c r e f l e c t i o n d a t a o b t a i n e d a l o n g the p r o f i l e e x t e n d i n g from 'A' t o the p o s i t i o n o f the S h e l l Anglo Cygnet w e l l ( s e e f i g u r e H . I ) . The d a t a a r e u n m l g r a t e d . Lower: A g e o l o g i c a l i n t e r p r e t a t i o n b a s e d on the m u l t i c h a n n e l d a t a ( a f t e r S n a v e l y and Wagner. 1981). The s t r u c t u r a l h o r i z o n s which were p i c k e d on the t r a v e l time s e c t i o n have been r e s e a t e d t o d e p t h by u s i n g the r e f r a c t i o n v e l o c i t i e s d e r i v e d In the VISP-80 d a t a I n t e r p r e t a t i o n ( C h a p t e r 4 ) . Note the p o s i t i o n of the S h e l l A n g l o Cygnet w e l l a t the e a s t e r n end o f the s e c t i o n . interpretation of the multichannel profi l e . The structural horizons which were picked on the travel time section have been rescaled to depth by using the refraction velocities derived in the interpretation of the VISP-80 data set (see Chapter 4). The cross section indicates Quaternary and upper Tertiary strata which have been deformed into a series of asymmetric folds and northeastward dipping imbricate thrusts. 121 These compressive structures are thought to result from the relative northeastward underthrusting of the Juan de Fuca plate beneath the North America plate. The interpretation of Snavely and Wagner proposes the existance of a mass of highly sheared and compressed melange lying beneath the sediments of the continental rise. This middle Miocene unit is thought to have been uplifted prior to deposition of the continental slope Pliocene strata. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0052985/manifest

Comment

Related Items