UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A re-evaluation of the seismic structure across the active subduction zone of Western Canada Drew, Jeffrey John 1987

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

Item Metadata

Download

Media
831-UBC_1987_A6_7 D73.pdf [ 9.6MB ]
Metadata
JSON: 831-1.0085605.json
JSON-LD: 831-1.0085605-ld.json
RDF/XML (Pretty): 831-1.0085605-rdf.xml
RDF/JSON: 831-1.0085605-rdf.json
Turtle: 831-1.0085605-turtle.txt
N-Triples: 831-1.0085605-rdf-ntriples.txt
Original Record: 831-1.0085605-source.json
Full Text
831-1.0085605-fulltext.txt
Citation
831-1.0085605.ris

Full Text

A RE-EVALUATION OF THE SEISMIC STRUCTURE ACROSS THE ACTIVE SUBDUCTION ZONE OF WESTERN CANADA by J E F F R E Y J O H N D R E W B.S . (Geophysics), The University of Cal i forn ia, 1984 A T H E S I S S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R O F S C I E N C E in T H E F A C U L T Y O F G R A D U A T E S T U D I E S Department of Geophysics and Astronomy We accept this thesis as conforming to the required standard T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A March 1987 © Jeffrey John Drew, 1987 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date r^ W e W . tf&l DE-6(3/81i ii ABSTRACT The 1980 Vancouver Island Seismic Project (VISP) was conducted to investigate lithospheric structure associated with the underthrusting oceanic Juan de Fuca plate and the overriding continental America plate. The principal components of the survey were: (l) an onshore-offshore refraction line, which was approximately perpendicular to the continental margin (line I), and (2) a refraction line which ran along the length of Vancouver Island approximately parallel with the continental margin (line IV). Lines I and IV were originally interpreted by Spence el a.1. (1985) and McMechan and Spence (1983), respectively. However since the original interpretations of these lines, deep multichannel seismic reflection data have been obtained on southern Vancouver Island as part of the 1984 L I T H O P R O B E project and off the west coast of the island during a marine survey in 1985. This study was undertaken to resolve differences between the subsurface structures proposed in the original interpretations of lines I and IV and those suggested by the more recently acquired deep reflection data. The vertical two-way traveltimes to promi-nent reflectors, observed in the onshore-offshore deep reflection data, were used as a constraint in constructing velocity models which are consistent with both the reflection and refraction data. The traveltimes and amplitudes observed in the VISP refraction data were modeled using a two-dimensional raytracing and asymptotic ray theory syn-thetic seismogram routine. The principal difference between the model originally interpreted for line I and the revised model involves the introduction of a twice repeated sequence of a low velocity zone (« 6.4 km/s) above a thicker high velocity zone (w 7.1 km/s) for the underplated region directly above the subducting Juan de Fuca plate in place of the single high velocity block underlain by a thick low velocity zone. The revised model for line IV is iii significantly different from the originally interpreted model. The two low-high velocity zones of line ] are continued along the length of the island at depths between 10 and 35 km. Below this, the structure of the subducted plate is included to maintain consistency with the revised model developed for line I. Additional features of the revised onshore-offshore model corresponding to line I in-clude an oceanic lithosphere that dips approximately 3° beneath the continental slope, then 14° to 16° beneath the continental shelf and Vancouver Island, and an average velocity for the upper oceanic mantle of 8.22 km/s. Two separate two-dimensional models were needed to explain the data collected along line IV as a result of consider-able azimuthal coverage due to a 30° change in profile direction. The revised models developed for line IV are consistent with the revised model developed for line I. The velocity in the upper 10 km ranges from 5.5 km/s to approximately 6.7 km/s. Below 10 km the velocity structure is consistent with that interpreted for line 1 and shows some variations along strike of the subduction zone. Several possible interpretations can be made for the origin of the sequence of layers directly above the subducting plate beneath Vancouver Island. The two favored inter-pretations are: (]) a. three stage tectonic process consisting of: stage 1 — offscraping of sediment from the top of the subducting plate forms the uppermost low velocity layer in the sequence; stage 2 — an imbricated package of mafic rocks derived by continuous accretion from the top of the subducting oceanic crust forms the first high velocity layer; and stage 3 — stages 1 and 2 repeat themselves with stage 2 currently occurring; or (2) remnant, pieces of oceanic lithosphere left stranded above the current subduct-ing plate during two previous episodes of subduction in which the subduction thrust jumped further westward isolating the remnant. The revised model along line IV indi-cates that this process of subduction underplating could have been a pervasive feature of this convergent margin. I V TABLE OF CONTENTS ABSTRACT ii LIST OF FIGURES vi ACKNOWLEDGEMENTS viii CHAPTER 1 INTRODUCTION 1 1.1 Tectonics And Geology Of The Vancouver Island Region 1 1.2 Previous Seismic Studies 7 1.3 Experimental Objectives 9 CHAPTER 2 DATA ACQUISITION AND PROCESSING 11 2.1 The Vancouver Island Seismic Project (1980) 11 2.2 Vancouver Island L1THOPROBE Experiment (1984) 13 2.3 FGP Marine Experiment (1985) 14 2.4 Instrumentation 15 2.5 Data Processing 15 2.5.1 V1SP 1980 Data 15 2.5.2 The 1984 L1THOPROBE and 1985 FGP Marine Surveys . . . 24 CHAPTER 3 DATA ANALYSIS 25 3.1 General Characteristics Of The VISP 1980 Data 25 3.1.1 Onshore-Offshore Data From VISP Line 1 29 3.1.1.1 The P-series Sections 29 3.1.1.2 The X-series Sections 31 3.1.2 Data. Collected Along VISP Line IV 31 3.1.2.1 Sections Pertaining To The Two Reversed Segments Of Line IV 33 3.1.2.2 The Two Offline Sections, F-North And N-South . . . . 34 3.2 The Deep Reflection Data Set 36 3.2.1 Data From The 1984 LITHOPROBE Survey 37 3.2.2 Data From The 1985 FGP Marine Survey 39 3.3 Two Dimensional Modeling 39 3.3.1 The Modeling Algorithm 39 3.3.2 Modeling Procedure 41 CHAPTER 4 INTERPRETATION OF SEISMIC REFRACTION DATA FROM LINE 1 OF VISP SURVEY 44 4.1 Initial Constraints 44 4.2 General Characteristics Of The Final Model 47 4.3 Interpretation Of The P-series Sections 49 4.3.1 Section P19 50 4.3.2 Section P13 53 4.3.3 Section P8 55 4.4 Interpretation Of X-series Sections 57 V 4.4.1 Sections X6 And X22 58 4.4.2 Sections X35 And X45 62 4.5 Final Velocity Model 65 CHAPTER 5 INTERPRETATION OF SEISMIC REFRACTION DATA FROM LINE IV OF VISP SURVEY 68 5.1 Initial Constraints 68 5.2 General Characteristics Of The Final Model 69 5.3 Interpretation Of The Hinged Model 71 5.3.1 Reverse Profile Recorded Along Segment N-A . . . . . . . . 72 5.3.2 Reversed Profile Recorded Along Segment A-F 77 5.4 Interpretation Of The Offline Model . . 84 5.5 Final Velocity Models 90 5.5.1 Hinged Velocity Model . 90 5.5.2 Offline Velocity Model 92 CHAPTER 6 DISCUSSION AND CONCLUSIONS 94 6.1 The Onshore-Offshore Profile 94 6.2 N-S Profile Along Vancouver Island . 105 6.2.1 The Hinged Model 105 6.2.2 The Offline Model 109 6.3 Conclusions I l l REFERENCES 114 APPENDIX Additional E-W Record Sections 118 A.l Common Shot Record Sections 118 A.2 Selected Common Receiver Record Sections 135 VI L I S T O F F I G U R E S 1.1 Tectonic and location map 2 1.2 Generalized geologic map of southern Vancouver Island 6 2.1 Tectonic and seismic profile map 12 2.2 Noise before arrival, on a trace from section X35, and power spectrum . 17 2.3 Trace with seismic arrival from section X35, and power spectrum . . . . 18 2.4 Noise before arrival, on a trace from section P13, and power spectrum . 19 2.5 Trace with, seismic arrival from section P13, and power spectrum . . . . 20 2.6 Noise before arrival, on a trace from section A-North, and power spectrum 21 2.7 Trace with seismic arrival from section A-North, and power spectrum . . 22 2.8 Section P8 before and after filtering 23 3.1 Region sampled along onshore-offshore line 26 3.2 Region sampled along N-S line, along Vancouver Island 27 3.3 Section Pi3 30 3.4 Sections X22 and X35 32 3.5 Section A-South 34 3.6 Sections F-North and N-South 35 3.7 Reflection section recorded along line 84-0] 38 3.8 Reflection section recorded along line 85-01 40 3.9 Receiver X35, raytracing, data, and traveltime picks with calculated traveltimes superimposed 42 4.1 Velocity model corresponding to refraction line I 46 4.2 Velocity depth profiles along line J 48 4.3 Shot P19, raytracing, data, traveltimes, and synthetics 51 4.4 Shot Pi3, raytracing, data, traveltimes, and synthetics 54 vii 4.5 Shot P8, raytracing, data, traveltimes, and synthetics 56 4.6 Receiver X6, raytracing, data, traveltimes, and synthetics 59 4.7 Receiver X22, raytracing, data, traveltimes, and synthetics 60 4.8 Receiver X35, raytracing, data, traveltimes, and synthetics 63 4.9 Receiver X45, raytracing, data, traveltimes, and synthetics 64 5.1 Hinged velocity model 70 5.2 Offline velocity model 71 5.3 Section N-North, raytracing, data, traveltimes, and synthetics 73 5.4 Section A-North, raytracing, data, traveltimes, and synthetics 74 5.5 Section N-North without Moho, raytracing, traveltimes, and synthetics . 78 5.6 Section A-South, raytracing, data, traveltimes, and synthetics 79 5.7 Section F-South, raytracing, data, traveltimes, and synthetics 80 5.8 Sections N-South and F-North with traveltimes superimposed 85 5.9 Section N(N-S), raytracing, data, traveltimes, and synthetics 86 5.10 Section F(S-N), raytracing, data, traveltimes, and synthetics 87 6.1 Generalized version of revised, and original onshore-offshore velocity model 95 6.2 Hypocenter cross section superimposed on the onshore-offshore model . . 98 6.3 Alternate onshore-offshore model, raytracing, data (section P13), traveltimes, and synthetics 100 6.4 Sequence of events that may have formed the series of layers directly above the subducting plate in the Vancouver Island region 103 6.5 Generalized versions of two models developed for the N-S profile . . . . 106 6.6 Velocity model proposed by McMechan and Spence (1983) for the N-S profile 107 V l l l ACKNOWLEDGEMENTS To my grandparents: Christina Drew, Ester Lundell, and especially Bert Lundell I would like to thank my supervisor Dr . Ron Clowes for his guidance throughout this project and for the many helpful discussions which helped keep me on track. I would also like to thank Dr . B i l l Slawson for cri t ical ly reviewing this thesis. To my friends in the Department of Geophysics and Astronomy, you know who you arc, and you are marvelousl Thanks especially to Sonya Dehler for the many invaluable discussions we had concerning the interpretation and modeling of the data. I would also like to thank Connie Cudrak and Sonya for their help in proof reeding (oops!) the final draft of this thesis. Thanks also to Ron and Lorraine MacLeod for all the times they had me over for dinner and for the money they let me win in assorted card games. A very special thank you is in order for Co l in , Barry , and Mike Zelt and their parents M r . and Mrs . Zee for making my stay in Vancouver an enjoyable one. Last but certainly not least 1 would like to thank my parents for their past and continuing encouragement and unwavering support. F inanc ia l support for this research came from the L I T H O P R O B E Phase 1 Col lab-orative Special Project grant and from operating grant A7707, both from N S E R C . 1 C H A P T E R 1 I N T R O D U C T I O N The southwest coast of Canada is a zone of convergent interaction between litho-spheric plates, with the oceanic Juan de Fuca and Explorer plates subducting beneath the continental America plate. This particular study concerns the re-evaluation of two seismic refraction models, which are consistent with a subduction zone-type structure for the Vancouver Island region, in light of deep reflection results acquired since their original development. 1.1 T e c t o n i c s A n d G e o l o g y O f T h e V a n c o u v e r I s l a n d R e g i o n The Vancouver Island plate tectonic regime is dominated by the relative motion of four lithospheric plates: (l) the Pacific plate, (2) the Juan de Fuca plate, (3) the Explorer plate, and (4) the America plate (figure l . l ) . The oceanic Juan de Fuca and Explorer plates are undergoing subduction beneath the continental America plate at convergence rates of approximately 4 cm/yr for the Juan de Fuca plate and less than 2 cm/yr for the Explorer plate (Riddihough, 1977). The difference in relative motion is accommodated by the Nootka fault zone which marks the boundary between the two plates (Hyndman et al., 1979) and which itself is being subducted beneath the continental America plate (Cassidy, 1986). In a hot spot frame of reference, the Explorer plate is almost stationary and thus may have stopped subducting in an absolute sense; subduction under northern Vancouver Island can be accounted for by the fact that the America plate is overriding the Explorer plate (Riddihough, 1984). The tectonic history of the convergent margin off the west coast of Canada has been studied and discussed by many authors from Atwater (1970) to Riddihough (1984), while the geologic significance of changes in plate configurations are discussed by Muller (1977). 2 Figure 1.1 Tectonic and location map. Boundaries of the Insular Belt, Coast Plutonic Complex, and Intermontane Belt are indicated by dashed lines (Riddi-hough, personal communication, 1986). 3 The primary crustal structure of Vancouver Island is believed to be comprised of an underthrusting oceanic plate overlain by an amalgam of accreted terranes, plutonic complexes, and successor basins. Most of Vancouver Island is part of Wrangellia, a ter-rane known to be allochthonous on the basis of paleomagnetic studies (Yole and Irving, 1980). The way in which this and other adjacent terranes have been accreted to the continental margin is largely unknown. However, Monger et a/. (1985) suggested that the various terranes of the Canadian Cordillera may have underthrust one another dur-ing accretion so that they are now vertically juxtaposed. It is thought that Wrangellia may have accreted to the continental margin as the oceanic plate on which it was riding began to subduct beneath the America plate. Wrangellia is one of the best recognized exotic terranes, pieces of which are found in southeastern Alaska, the Queen Charlotte Islands, Vancouver Island, and eastern Oregon (Jones et a/., 1977). It is a submarine-arc complex composed of Middle to Upper Triassic tholeiitic basalts, calcareous sedimentary rocks, and early to middle Jurassic terrigenous elastics and volcanics (Coney et ai., 1980). The Wrangellia terrane as defined by Jones et, al. (1977) is equivalent to the westernmost tectonic province (the Insular Belt terrane) in the Canadian Cordillera and comprises most of Vancouver Island. The Strait of Georgia is considered to be the boundary between the two westernmost tectonic provinces: the Insular Belt to the west, and the Coast Plutonic Complex to the east (figure 1.1). Geologically, the strait separates the late Mesozoic-Tertiary crystalline rocks of the mainland from the middle Jurassic volcanic, plutonic, and sedimentary rocks of Vancouver Island. The Insular Belt terrane extends from southern Vancouver Island to south-central Alaska. The geology of the southern portion of the belt, including Vancouver Island, is described by Muller (1977). On Vancouver Island, middle to late Paleozoic rocks (the Sicker Group) consist of breccia, tuff, and flows of basaltic to rhyolitic composition, 4 underlying a sedimentary sequence of argillite, siltstone, and limestone. The Sicker Group is overlain by up to 6000m of pillow lavas, pillow breccias, and minor sediments of the upper Triassic Karmutsen Formation (Muller, 1977). Paleomagnetic measurements made on the Karmutsen basalts (Yole and Irving, 1980) suggest that the Wrangellia rocks of Vancouver Island have moved northward relative to the North American craton by at least 1300 km, and possibly by as much as 4900 km, since the late Triassic. It is thought that Wrangellia (or the Insular Belt terrane) was accreted to the continental margin by mid-Cretaceous time, and since then has been fragmented by faulting (Coney et ai., 1980). The lower Jurassic Bonanza volcanics overlie the Karmutsen Formation. The Island Intrusions form plutons and batholiths of quartz diorite, granodiorite, and quartz mon-zonite. Jura-Cretaceous greywacke, chert, argillite, and greenstone make up the Pacific Rim Complex on the extreme west coast of Vancouver Island (Muller, 1977). The Ozette and Hoh melanges make up the outer continental shelf and slope deposits (Snavely and Wagner, 1981), and a fragment of Crescent volcanics (identified as a detached slab of Eocene oceanic crust) lies on top of the Ozette melange and to the west of the Pacific Rim Complex (Yorath, 1980). On the east coast of the island, the upper Cretaceous Nanaimo Group consisting of sandstone, siltstone, argillite, and conglomerate overlie the pre-Cretaceous rocks. The rocks associated with the Coastal Mountain Range on the mainland side of Georgia Strait are part of the Coast Plutonic Complex which extends from northern Washington state to the Yukon. The Coast Plutonic Complex consists of a. matrix of gneisses, migmatites, and foliated plutonic rocks (Roddick and Hutchinson, 1974). The predominant rock types of the Coast Plutonic Complex include granodiorite, quartz diorite, diorite migmatite, gabbro, gneiss, and migmatite, as well as small amounts of metasedimentary and metavolcanic rocks. Paleolatitude measurements on Cretaceous 5 plutons within the Coast Plutonic Complex suggest that it was 10° to 20° south of its present location when it was emplaced (Symons, 1973; Beck and Noson, 1972). Figure 1.2 is a generalized geologic map of southern Vancouver Island and the Olympic Peninsula of northwestern Washington. This map divides the geology of south-ern Vancouver Island and surrounding areas into four major tectonic units (Clowes et a/., 1987a): (1) A pre-Tertiary continental framework, which on Vancouver Island is equivalent to the Wrangellia terrane (Jones et a/., 1977). (2) The Leech River schist, which as used here also includes some similar Mesozoic rocks that were juxtaposed with Wrangellia at about the same time as the Leech River complex. On southern Vancouver Island these are the Pandora Peak unit; on west-ern Vancouver Island, the Pacific Rim Complex. All of these rocks are considered to comprise the Pacific Rim terrane. (3) Lower Eocene basalts, which are called the Metchosin Formation on southern Vancouver Island and the Crescent Formation in northwestern Washington state and are known to exist offshore as a result of exploratory drilling (Shouldice, 1973). These Eocene basalts form the Crescent terrane. (4) Core rocks, a Cenozoic accretionary complex exposed in the Olympic Penin-sula beneath the Eocene basalts and which underlies the continental margin west of Vancouver Island and Washington. The core rocks combined with the Ozette and Hoh melanges, which comprised the outer continental shelf and slope deposits (Suavely and Wagner, 1981), are equivalent to the modern subduction complex. The modern sub-duction complex lies below terranes (l) — (3) and consists of younger material which has been thrust below the slope and shelf deposits and Vancouver Island. 6 0 20 40 60 km i i i i i | 122° SEATTLE CORE ROCKS (LIGHTER PATTERN WHERE INFERRED) EOCENE BASALTS (LIGHTER PATTERN WHERE INFERRED) HH PRE-TERTIARY CONTINENTAL FRAMEWORK LEECH RIVER SCHIST Figure 1.2 Generalized geologic map of southern Vancouver Island and the Olympic Peninsula of northwestern Washington. The circled numbers identify the L I T H O P R O B E reflection lines 1 to 4. The line labeled A - A ' identifies the location of a geologic cross section examined by Clowes et al. (1987a) and not considered in this study. The refraction line corresponds to line I of the VISP survey (from Clowes et al, 1987a). The different geological character of the Insular Belt rocks and the Coast Plutonic Complex rocks suggest different origins which may be reflecting different deep-seated structures. Both magnetotelluric and heat flow data suggest that a significant change in physical properties exist at depth between the Insular Belt and the Coast Plutonic Complex. There are several theories as to the origin of the Insular Belt and the Coast Plutonic Complex (White, 1983; White and Clowes, 1984). One possible scenario is presented by Dickinson (1976) in which the Insular Belt terrane accreted to the con-tinental margin prior to the Middle to Late Jurrassic. The Coast Plutonic Complex evolved later (primarily during Cretaceous time) as a result of arc magmatism related to subduction. This suggests that the Coast Plutonic Complex was superimposed on the Insular Belt some time after the Insular Belt was accreted to the continental margin. From the viewpoint of this study, a question arises as to whether differences between the Insular Belt and the Coast Plutonic Complex are manifest in the seismic structure. 1.2 P r e v i o u s S e i s m i c S t u d i e s It is generally accepted that subduction has been occurring along Canada's west coast over the past several million years, although the timing of subduction activity is poorly constrained (Riddihough and Hyndman, 1976). The oldest dateable mag-netic anomalies east of the Juan de Fuca ridge (now at the margin and parallel to the Washington-Oregon coast) were estimated to be approximately 9 Ma. Since a complete record older than this lies to the west of the ridge system, subduction must have taken place at least up to the last 9 Ma (Riddihough and Hyndman, 1976). Geological and geophysical data supporting the case for subduction are presented by Riddihough and Hyndman (1976), and a comprehensive geophysical review of Canada's western conti-nental margin is presented by Keen and Hyndman (1979). Results of several geophysical surveys conducted in the Vancouver Island region are discussed by Spence (1984) and Spence et al. (1985). The studies discussed in this section are from recent seismic experiments, the results of which were directly used in this study. 8 In 1980, a large-scale onshore-offshore crustal refraction experiment was conducted in the Vancouver Island region (Ellis et a/., 1983). Specific details of this survey along with a complete examination of some of the data collected are presented in forthcoming chapters. Waldron (1982) interpreted ocean bottom seismograph (OBS) data collected along the offshore part of a refraction line run in the direction of plate convergence. He developed a two-dimensional velocity model which provides information on veloc-ities and thicknesses of layers down to depths less than 15 km for the offshore area. McMechan and Spence (1983) present a two-dimensional model for a refraction line recorded along the length of Vancouver Island. Their model was considered well con-strained to a. depth of approximately 15 km although arrivals were recorded from what was believed to be the top of the continental Moho, which was modeled at a depth of 37 km. Spence (1984) used the model proposed by McMechan and Spence (1983) and that developed by Waldron (1982) to help constrain a two-dimensional velocity model that he interpreted from the onshore-offshore refraction data collected during the same experiment. The model proposed by Spence (1984) and Spence et al. (1985) shows the oceanic lithosphere subducting beneath Vancouver Island at an angle of between 14° and 16°. This range differs from the 9° dip for the subduction of the Juan de Fuca plate as determined by Taber and Lewis (1986) from seismicity studies and from an onshore-offshore refraction profile off the coast of Washington state. Such results sug-gest possible significant lateral variations along strike of the subducting Juan de Fuca plate. Possibly the most interesting feature of the model developed by Spence (1984) is a wedge of material beneath Vancouver Island with a velocity (7.7 km/s) close to what might be expected for the lower crust or upper mantle of the subducting oceanic lithosphere. One of the explanations given for the origin of this high velocity wedge was that it was a piece of the oceanic crust and upper mantle which was emplaced during a previous phase of subduction followed by a westward jump of the onset of subduction. 9 Since the 1980 refraction experiment on Vancouver Island, two large-scale deep reflection experiments have been conducted in the region. The first was a Vibroseis survey conducted in 1984 on Vancouver Island and the second was a marine survey conducted in 1985. Both of these surveys provide deep reflection data which exhibit coherent reflectors that can be correlated from one reflection section to another. A reflector which probably corresponds to the top of the subducting Juan de Fuca plate can be followed from the deep ocean to the west coast of Vancouver Island. Beneath Vancouver Island, there are a series of reflectors which are in approximately the same position as the anomalous feature with a mantle-like velocity required in the model proposed by Spence (1984). However, the subsurface structure for the area beneath Vancouver Island, as suggested by the deep reflection data, does not agree with that proposed by Spence (1984) and Spence et a/., (1985). Clowes et al. (1986) discuss some of these differences. A more thorough discussion of these surveys as well as samples of the data collected are presented in the forthcoming chapters. 1.3 Experimental Objectives As indicated in the previous discussion, the subsurface structure in the velocity models developed by McMechan and Spence (1983) and by Spence (1984), based on the onshore-offshore refraction data collected in the Vancouver Island region, does not agree with the subsurface structure suggested by the deep reflection data recorded in the same region. The objective of this study has been to develop velocity structure models, both for the onshore-offshore line and the line along Vancouver Island, which agree with the refraction and reflection data sets and are mutually consistent. The steps involved in re-evaluating these models were: (l) to re-process the refraction data by applying appropriate filters to optimize the quality of the sections recorded, (2) to constrain 10 the velocity structure model by using the vertical two-way traveltimes to prominent reflectors observed in the deep reflection data as constraints on the subsurface model structure, and (3) to develop velocity models which not only agree wi th the traveltirne and ampli tude constraints imposed by the refraction data, but which also agree with each other at their line of intersection. The velocity structure models were developed using a two-dimensional raytracing and asymptotic ray theory synthetic seismogram routine described by Spence et a/. (1984). 11 CHAPTER 2 DATA ACQUISITION AND PROCESSING 2.1 The Vancouver Island Seismic Project. (1980) The Vancouver Island Seismic Project (VISP) was a large-scale marine/land seismic refraction/reflection survey conducted during the summer of 1980. The objective of the survey was to explore the crust and upper mantle in the transition zone from the Pacific Ocean across Vancouver Island to the mainland of British Columbia. A complete description of the refraction and reflection programs may be found in Ellis and Clowes (1981) and Ellis et al. (1983). A brief description of the elements of the project directly relevant to this thesis are presented below. The seismic refraction program consisted of four profiles; two of them (1 and IV) are shown in figure 2.1. Line I extended from shotpoint P19 in the deep ocean off the west coast of Vancouver Island, across the island to shotpoint J on the British Columbia mainland. Line IV ran along the length of Vancouver Island, approximately parallel to the continental margin. Lines II and III (not shown) were also shot parallel to the continental margin, line II in the deep ocean and line III along the mid-continental shelf. Data from lines II and 111 were not examined as part of this thesis. For line 1, as many as 32 land seismographs were deployed on Vancouver Island, islands in the Strait of Georgia, and on the British Columbia mainland. Four OBS's were deployed along the marine portion of the profile, but only data from the three OBS's shown in figure 2.1 have been interpreted. Two 825 kg shots at shotpoint J near the eastern end of line 1, and a series of 17 shots ranging in size from 200 to 825 kg, designated as the P-series, were fired over the continental slope and ocean basin into these land and marine detectors. Shots P i to P6 were fired on the continental shelf where the water depth ranged from 500m to 2100m. Water depth ranged from 2425m 12 Figure 2.1 Tectonic and seismic profile location map. Arrows show the direc-tion and magnitude of plate motion relative to North America. Solid triangles represent the approximate location of the inland volcanic belt. Lines 1 and IV (dashed) refer to seismic refraction lines recorded as part of the 1980 VISP ex-periment. The solid line segment RL refers to a short reflection survey also recorded during the VISP experiment. Solid lines 84-01 to 84-04 and lines 85-0] to 85-05 refer to crustal reflection lines recorded as part of the 1984 LIT110-PR.OBE survey and the 1985 Frontier Geoscience Program (FGP) survey of the Geological Survey of Canada. Stars on Vancouver Island and the mainland show the locations of individual shots; offshore, stars near P and P' identify the ends of a line of explosive charges. Solid circles are ocean bottom seismograph (OBS) locations. USGS refers to a seismic reflection line recorded by the U.S. Geological Survey; the well symbol on the shelf (near P') is the location of the Shell-Anglo Cygnet drill hole. Bathymetric contours are in meters (after Clowes et a/., 1986). 13 to 2625m for the remaining shots fired over the ocean basin (P8 to P19). An additional eighteen 50 kg charges detonated between P' and P19 and a series of shots from a 16 litre airgun were recorded on the OBS's; a continuous seismic profile (CSP) using a 5 litre airgun was also recorded along the marine portion of line I. The data obtained from the OBS's and the CSP were originally interpreted by Waldron (1982) and have since been reinterpreted by D.J. White (personal communication, 1986). For line IV, 38 land seismographs were originally deployed along the line segment N-A in figure 2.1 and shots of 900, 900 and 1800 kg were detonated at N, A and F, respectively. The seismographs were then redeployed along the line segment A-F and shots of 1800, 900, and 900 kg were detonated at N, A and F, respectively. In addition to the four seismic refraction profiles recorded, there was also a small-scale reflection program conducted to explore the feasibility of obtaining coherent reflections to upper mantle depths. Line RL in figure 2.1 was part of this reflection experiment. Details and results of the reflection program have been reported by Clowes et al. (1983). The favorable results obtained from this survey and the existing refraction interpretation were significant factors in the selection of Vancouver Island as the site of the 1984 LITHOPROBE deep reflection program. 2.2 Vancouver Island L I T H O P R O B E Experiment (1984) The LITHOPROBE experiment conducted on Vancouver Island was designed to study the geometry and structural characteristics associated with the subducting Juan de Fuca plate, and to determine the large-scale structure of several accreted terranes exposed on the island. A total of 206 km of deep seismic reflection data (16s record length) were recorded along four profiles, line 84-01 to line 84-04 (figure 2.1). Line 84-01 crosses the island approximately coincident with line I of the onshore-offshore refraction profile recorded during the VISP experiment. Some three-dimensional control 14 on the interpretation of line 84-01 is provided by line RL and by line 84-03 located approximately 20 km east of line 84-01 (figure 2.1). Two additional lines, 84-02 and 84-04, were recorded in the southern part of the island to help determine the significance and attitude of the Leech River, San Juan and Survey Mountain faults. These four profiles, particularly line 84-01, provide most of the additional constraints on Vancouver Island that were used to reinterpret the seismic refraction data recorded during the 1980 VISP experiment. The principal results of the LITHOPROBE reflection program are discussed in Clowes et al. (1984), Yorath et al. (1985a), Yorath et al. (1985b), Green et al. (1986a), and Clowes et al. (1987a). 2.3 F G P M a r i n e R e f l e c t i o n E x p e r i m e n t (1985) Data from the marine seismic reflection survey, conducted off the west coast of Vancouver Island in 1985, were obtained under the auspices of the Frontier Geoscience Program (FGP) of the Geological Survey of Canada. The primary objectives of the marine experiment were: (l) to clarify the convergent interaction between the Juan de Fuca plate and the America plate, and (2) to better understand the structures associated with accretionary tectonics in this area of plate interaction. Five 30-fold seismic reflection profiles of 16s record length and totalling 520 km were shot over the convergent margin where the Juan de Fuca plate is subducting beneath the America plate. The location of the five marine profiles (lines 85-01 to 85-05) is shown in figure 2.1. Line 85-01 of the marine survey is approximately coincident with the offshore component of line 1 of the VISP refraction program. Some three-dimensional control is provided by lines 85-02 and 85-05 of the marine experiment, and a reflection profile recorded by the United States Geological Survey. Reflection line 85-01 of the marine survey combined with line 84-01 of the LITHO-PROBE land survey provides a high quality reflection profile extending from the deep 15 ocean across Vancouver Island, approximately coincident with line 1 of the VISP refrac-tion program. A preliminary interpretation of some of the FGP marine reflection data is given by Clowes et al. (1987b). 2.4 Instrumentation A detailed description of the instruments used in the 1980 VISP experiment, includ-ing instrument characteristics, the shooting and recording geometries, and a discussion of errors in timing and site location may be found in Ellis and Clowes (1981) and Ellis et al. (1983). Instrumentation used in the 1984 LITHOPROBE survey included a 120-channel Texas Instruments DFS V recording system with four Mertz Model 18 Vibroseis sources. All lines were recorded to 16s with 30-fold coverage. A further description of the 1984 LITHOPROBE reflection program, including the geometry of the survey and the instruments used, is given by Green et al. (1985) and Clowes et al. (1984). Instruments used in the 1985 FGP marine survey included a 50 airgun source array with a total volume of 100 litres tuned to concentrate energy in the frequency band below 50 Hz. Signals were received by a 3000m, 120-channeI streamer, and also were recorded on a 120-channel Texas Instruments DFS V recording system, the same as that used in the land survey (Clowes et a/., 1987b). 2.5 Data Processing 2.5.1 VISP 1980 Data A detailed description of the steps involved in the initial data reduction is to be found in Ellis and Clowes (1981). The 17 P-series shots detonated off the west coast of Vancouver Island were corrected for sea bottom topography and detonation depth by placing all shots at an equivalent datum depth of 2600m (Spence, 1984; Spence et a/., 1985). This involved a time and distance offset correction for each shot. Variations 16 in topography at the receivers were corrected by adjusting the receiver elevation to sea level using a velocity of 5.5 km/s. A correction for basement topography was calculated for shots P8 to P19 in the ocean basin (Spence, 1984; Spence et al., 1985); however, the additional correction to replace the varying thickness of sedimentary layers beneath these ocean basin shots by a uniform layer, as applied by Spence (1984). has been removed since the sedimentary structure as determined by Waldron (1982) is used in the revised model for line I. The distance offset correction associated with removing the varying sedimentary thickness correction ranges from zero to a few tens of meters and has been ignored since it is too small to be observed on any of the sections examined in this study. No basement correction was applied to shots P i to P6 on the continental slope; however the topographic correction applied to these shots was significant since the water depths ranged from 500m to 2100m. Because of the large travel time correction associated with correcting shots P i to P6 to a datum depth of 2600m, the potential for error in these corrections is relatively large compared with the corrections applied to shots P8 to P19 in the ocean basin which ranged in depth from 2425m to 2625m. Additional data processing designed to extract the most information possible from the data is described below. Spectral analyses were performed on representative signal and noise recordings from both the refraction profile run along Vancouver Island and the refraction profile extend-ing from the deep ocean across the island to the mainland of British Columbia. This analysis showed that different segments of the data required different filtering parame-ters to separate the signal from the noise. Figure 2.2 shows a portion of a seismic trace from one of the receiver sections (section X35) before the desired signal has reached the receiver and its corresponding power spectrum. Figure 2.3 shows a seismic trace and its corresponding power spectrum after the signal has arrived. The power spectrum of the desired signal is within the 2.5 . 1 7 (a) HJo o~l I I I I I 1 1 I 1 1 2 0 2 2 2.4 2.6 2 .8 3.0 3.2 3.4 3.6 3.8 4.0 TIME (S) O (b) 24.0 27.0 30.0 FREQUENCY (HZ) Figure 2.2 Noise before arrival on a trace from section X35 (shot-receiver dis-tance « 227 km), (a) portion of noisy trace, (b) power spectrum of (a). to 9 Hz range, while figure 2.2 shows a significant amount of noise below 3 Hz and above 9 Hz. Figure 2.4 shows a segment of a seismic trace from one of the P-series sections (sec-tion Pi3) before the desired signal has arrived and its corresponding power spectrum. Figure 2.5 shows a segment of the same trace that includes the seismic signal and its corresponding power spectrum. We again see that there is a significant amount of back-ground noise below 3 Hz, while the energy in the signal is primarily in the 2.5 to 9 Hz range. The characteristics of these periodograms are typical of all the data examined along line I of the VISP survey. Figure 2.3 (a) Arrival of desired seismic energy on a trace from section X35 (shot-receiver distance w 309), (b) the corresponding power spectrum for that section of trace, and (c) segment of trace in (a) after a 2.5-9 Hz bandpass filter was applied. The arrowhead indicates the location of the first break. The power and amplitude scales used in this figure are different, than those used in figure 2.2. 19 7.8 8.0 TIME IS I 9.0 (b) CD" I • io Q . 6.0 12.0 IS.O 18.0 FREQUENCY (HZ) 21 .0 27.0 30.0 Figure 2.4 Noise before arrival on a trace from section P13 (shot-receiver dis-tance « 201 km), (a) portion of trace before arrival, (b) power spectrum of (a). The same type of analysis as described above was applied to the data collected along line IV. A segment of a trace from section A-North and its corresponding power spectrum before and after the desired signal has arrived are shown in figures 2.6 and 2.7, respectively. This trace is typical of data obtained along the shorter reversed segments (N-A and A-F) of line IV. We see that there is significant background noise extending from 0 to approximately 14 Hz. The energy in the signal is concentrated in the frequency band from 6 to 12 Hz, which unfortunately falls within the range of frequencies where there exists a significant amount of background noise. Fortunately, the majority of the background noise is outside of the 6 to 12 Hz range. Not all the sections from line IV 20 Figure 2.5 Seismic arrival on section P13 (shot-receiver distance « 201 km), (a) portion of trace with arrival present, (b) power spectrum of (a), (c) segment of trace in (a) after a 2.5-9 Hz bandpass filter was applied. The arrowhead indicates the location of the first break. The power and amplitude scales used in this figure are different than those used in figure 2.4. 21 7.8 8.0 TIME (SI 9.0 12.0 15.0 18.0 FREQUENCY (HZ) 21 .0 24.0 27.0 Figure 2.6 Background noise from a trace on section A-North (shot-receiver distance « 81 km), (a) segment of trace before arrival, (b) corresponding power spectrum of (a). required filtering, but the ones that did received a similar analysis to that described above to determine the optimum filtering parameters. An eight-pole Butterworth bandpass filter (Kanasewich, 1981) was applied to most of the refraction data to try and improve the signal-to-noise ratio. A 2.5 to 9 Hz bandpass filter was used on all the sections examined from line I and bandpass filters ranging from 1-10 Hz to 6-13 Hz were applied to data recorded along line IV. Filtering visibly enhanced the data; however, in some cases the data were of such poor quality that even the filtered sections still have a low signal-to-noise ratio (figure 2.8). The filtered sections as well as the respective unfiltered sections were used to make the time picks used in modeling the data. 22 (b) o FREQUENCY (HZ) Figure 2.7 Seismic arrival on a trace from section A-North (shot-receiver dis-tance ^  81 km), (a) portion of trace showing seismic arrival, (b) power spectrum of (a), (c) segment of trace in (a) after a 6-12 Hz bandpass filter was applied. The arrowhead indicates the location of the first break. The power and ampli-tude scales used in this figure are different than those used in figure 2.6. Figure 2.8 Section P8 before and after filtering, (a) section P8 before filtering, (b) section P8 after it has been bandpass filtered from 2.5-9 Hz. Each trace has been multiplied by an amplitude scaling factor of y^j, where d is shot-receiver distance in kilometers, to enhance arrivals at greater distances. 24 2.5.2 The 1984 L I T H O P R O B E And 1985 F G P Surveys The processing parameters used on the 1984 L I T H O P R O B E data set are described by Green et a.1. (1985), and those used on the 1985 F G P marine reflection data are described by Clowes et a i . (1987b). Both migrated and unmigrated data sections were examined in an effort to develop the most plausible interpretation of the data. 25 CHAPTER 3 DATA ANALYSIS 3.1 General Characteristics Of The VISP 1980 Data Thirteen VISP seismic refraction sections were used to model the lithospheric struc-ture along profiles 1 and IV (figure 2.1). The two lines are roughly perpendicular and both have a slight bend near their centers, a feature which is readily seen in figure 2.1. A bend such as that observed in lines I and IV can make modeling the data with a two-dimensional modeling routine very difficult, since the data are actually sampled in three-dimensions. Moving from the west, or shotpoint P19, to the east along line I, we see that the line veers to the north at the west coast of Vancouver Island (figure 3.1). The angular difference between the two portions of the line is about 27°. This means that the seismic energy recorded in the form of first arrivals has probably traveled through material which is geographically north of line I, as shown in figure 3.1. This "bend factor" is only a point to keep in mind when looking at data obtained from shots that were not inline with the receivers on which they were being recorded (e.g. P8, P13, and Pi9). The bend in the line did not have to be considered by Waldron (1982) in his interpretation of the data from OBS's 1, 3 and 5, nor was it a factor in interpreting the shots at shotpoint J, on the extreme eastern end of line 1, since in both cases the receivers on which the energy was being recorded were approximately inline with the shots. Figure 3.1 shows the azimuthal range over which seismic energy from shot P19 might travel in order to reach receivers located on Vancouver Island and the mainland. Shots P8 and P13 would have a slightly smaller azimuthal range than that shown in figure 3.1. The potential problem arises when the seismic energy reaches the west coast of Vancouver Island where the subsurface geology is believed to become more complex 26. Figure 3.J Shaded area represents the subsurface area through which first arrival energy might pass in traveling from a shot at shotpoint Pl9 to the receivers (along line I) on Vancouver Island and on the mainland of British Columbia. The arrow indicates the direction of subduction beneath the shaded region (after Spence et ai., 1985). 27 Figure 3.2 Shaded area represents the subsurface area through which first arrival energy might pass in traveling from a shot at shotpoint F to the receivers along line IV on Vancouver Island (after McMechan and Spence, 1983). (Sutherland Brown and Yorath, 1985) than that seen in the oceanic environment to the west. For line 1, the "bend factor" is not a major problem since line I is approximately parallel to the direction of subduction and lateral variations in the path that energy might take in traveling from a shot to the receivers is along the strike of the subducting plate. In the case of line IV, the bend at shotpoint, A represents an approximate 30° change in profile direction (figure 3.2). This line was shot in two smaller reversed segments, N-A and A-F. The bend did not pose a problem in modeling these shorter reversed segments since the receivers on which the shots were recorded were approximately inline with 28 the shots themselves. There was also a large shot at F recorded along the northern segment N - A as well as a large shot at N recorded along the southern segment A - F , and it is for these two shots that the "bend factor" becomes important. For these two sections we are again faced with the problem that we are sampling the subsurface geology in three dimensions and trying to model it using a two-dimensional model. Figure 3.2 shows a range of paths which seismic energy from a shot at F might follow in traveling to receivers along segment N - A . Unlike line I, which runs approximately perpendicular to the continental margin and therefore approximately parallel to the direction of subduction in the area, line IV is roughly perpendicular to the direction of subduction. In subduction zones, the depth to interfaces associated with the subduction process changes faster along lines parallel to the direction of subduction than along lines perpendicular to the direction of subduction. Since the lateral variations in possible raypaths along line IV are perpendicular to the direction of subduction, there could easily be a significant change in the depth to any interface beneath the shaded area (figure 3.2) that is associated wi th the subduction process. Because of the lateral changes described above and other lateral differences possibly associated with an earlier phase of subduction, a slightly different two-dimensional model was required to fit the data for the long offline sections, F-North and N-South, than was interpreted in fitting the N - A and A - F segments independently. A l l of the data sections presented in this thesis are plotted using a reducing velocity of 8 km/s and such that the relative amplitudes can be compared from one trace to another. The amplitude on each trace is proportional to y—;, where d is the distance from the shot. The filtering parameters discussed in the previous chapter have been applied to all the sections except for those presented in the Appendix , which remain unfiltered. 29 3.1.1 O n s h o r e - O f f s h o r e Data F r o m V I S P L i n e I Seven refraction sections from line I were interpreted in detail to develop a two-dimensional onshore-offshore model, although many other sections had been recorded and were also considered. This was possible since many of these sections exhibited the same characteristics as those sections used in modeling. Even the sections used in modeling often had very similar characteristics, the main difference being the level of background noise and the first break arrival times. Refraction sections recorded along line I but not specifically modeled as part of this thesis are presented in the Appendix. Three of the sections examined were generated by three separate shots (P8, Pi 3 and P19) recorded on all the receivers, while the four other sections represent four separate receivers (X6, X22, X35 and X45) recording the 17 P-series shots fired off the west coast of Vancouver Island. Receivers X6 and X22 were both located on Vancouver Island, while receivers X35 and X45 were located on the mainland of British Columbia. The sections recorded from shots off the coast will be referred to by their shot number, while sections which represent a single receiver recording many shots will be referred to by the receiver number. For example, section P19 will refer to the section generated from shot P19, while X35 will refer to the section produced by recording the 17 P-series shots on receiver X35. 3.1.1.1 T h e P -se r i es S e c t i o n s Record section P13 was filtered from 2.5 to 9 Hz and is shown in figure 3.3. Most of the features present on this section are also seen on sections P8 and P19, so section P13 is the only one presented here. All three sections have a data gap near their center which corresponds to the location of Georgia Strait where no data were recorded. The three sections also show a jump in the first break traveltime of approximately 0.75s across 30 j -I H - i H 1 ' 'i " r - S V , L i •—,—1 1, ' ' i ' i i L - r — 1 1 145 155 165 175 185 195 205 215 225 235 245 255 265 275 285 295 305 315 . . S H O T - R E C E I V E R DISTANCE (KM) x /est East Figure 3.3 Section recorded from the shot at shotpoint P13 (filtered from 2.5-9 Hz), a — a' identifies the approximate location of a possible secondary arrival. Four of the seismograms are labelled by their receiver number: X6, X22, X35, and X45. this data gap. The apparent velocity of the first arrivals across the Vancouver Island receivers is between 7.5 and 7.7 km/s, while across the mainland stations the apparent velocity of the first arrival is closer to 8 km/s. The apparent velocities seen on section P8 and P19 are similar to those seen on section P13. The amplitudes seen on section P13 are slightly larger for receivers located on the mainland side of Georgia Strait than for receivers located on Vancouver Island, and there is a significant fluctuation in amplitude from trace to trace. These features are also seen in sections P8 and P19. A set of secondary arrivals approximately 0.25s after the first arrivals is present on the Vancouver Island stations in section P]3. These secondary arrivals are less clear on section P19 and may or may not exist on section P8. The signal-to-noise ratio is considerably lower on section P8 than on P13 or P19, probably due to the smaller charge used for shot P8 than for shots Pi3 and P19 (200 kg versus 825 kg). 31 3.1.1.2 T h e X - s e r i e s S e c t i o n s Sections X22 and X35 were filtered from 2.5 to 9 Hz and are shown in figure 3.4. Most of the features seen on these sections are also seen in the two other receiver sections (X6 and X45) used in modeling profile 1. There is a clear first arrival present on all four receiver sections, but on sections X35 and X45 it is not easy to follow for distances of less than 200 and 270 km, respectively. The apparent velocity of the first arrival across all the sections is approximately 8.5 km/s. A set of possible secondary arrivals is identified on section X22 as being between a and a' (figure 3.4). These arrivals appear to die out for the traces corresponding to shots closer to the coast than P8. There is also evidence of a secondary arrival on section X6, while the existence of a secondary arrival is less clear on sections X35 and X45. The apparent velocity of the secondary arrivals on sections X6 and X22 is roughly 9.2 km/s. The amplitude of the first and secondary arrivals varies significantly from trace to trace across each of the four sections. The amplitude of the secondary arrival on sections X6 and X22 is, in general, as large or larger than the first arrival. 3.1.2 D a t a C o l l e c t e d A l o n g V I S P L i n e I V Six refraction sections were recorded along line IV and all six were used in modeling. The various sections recorded along this line will be referred to in the following way. With reference to figure 2.1, sections N-North and A-North will refer to a shot at shot-points N and A, respectively, recorded along the northern portion of line IV (segment A-N). Sections A-South and F-South will refer to a shot at shotpoints A and F, respec-tively, recorded along the southern portion of line IV (segment A-F). Section F-North refers to a shot at F recorded on the northern receivers and N-South refers to a shot at N recorded on the southern portion of line IV. In almost all cases, the receivers were 32 310 300 290 280 270 260 250 240 S H O T - R E C E I V E R DISTANCE (KM) 140 230 220 " 210 QurtT_Pt?riri\7irr? niQTAMrtr C I / M \ West Figure 3.4 Receiver sections recorded on Vancouver Island and the mainland of British Columbia along line I. (a) section X22 (filtered 2.5-9 Hz), recorded on Vancouver Island, a—a1 indicates the location of a possible secondary arrival, (b) section X35 (filtered 2.5-9 Hz), recorded on the mainland of British Columbia. The seismograms recorded from shots Pi, P8, P13, and P19 are indicated on both record sections. The distance scale again represents shot-receiver distance, but in this case the distance is measured from a stationary receiver to each of the 17 shots off the coast, and therefore increases from right to left. 200 East placed on bedrock to assure good coupling with the ground. The shots at shotpoint A were detonated in a shallow water-filled quarry, while shots at both N and F were detonated in deeper water. Shotpoint F (figure 2.1) was located in the Strait of Juan de Fuca and the charge was suspended at a depth of approximately 110m. Explosive charges at shotpoint N were detonated on the bottom of a narrow fjord in water depths of 60m and 85m for shots N-North and N-South, respectively. 3.1.2.1 S e c t i o n s P e r t a i n i n g T o T h e T w o R e v e r s e d S e g m e n t s O f L i n e I V Sections N-North and A-North are a reversed pair, as are A-South and F-South. These four record sections have many similar characteristics, so A-South will be the only section discussed at this time; the other three sections will be presented in Chapter 5 together with the interpretation of all the data from line I V . Section A-South was filtered from 6 to 13 Hz and is shown in figure 3.5. A very clear first break is visible across the entire section and is comparable with that seen on the other two sections. The apparent velocity of these arrivals ranges from approximately 5.5 km/s for the small shot-receiver distances to about 7 km/s for the arrivals recorded at the greatest distances. The majority of the first break energy on section A-South has an apparent velocity of roughly 6.5 km/s. The first break energy on the small shot-receiver offset traces seen on section N-North has an apparent velocity closer to 6 km/s. No small shot-receiver offset energy was recorded for section F-South since shotpoint F was approximately 15 km away from the nearest receiver. Other than these two minor exceptions, the apparent velocities associated with the first, breaks are similar on all four sections. In addition to the distinct first arrivals seen on each of the four sections, there are also some less prominent secondary arrivals present. One set of these arrivals is seen in figure 3.5 and is identified on the section as being between a and a'. These arrivals 34 -5 5 15 25 35 45 55 65 75 85 95 105 115 125 135 145 155 165 175 185 Nor th SHOT-RECEIVER DISTANCE (KM) South Figure 3.5 Section A-South (filtered 6-13 Hz) recorded on line IV along segment A-F. a — a' identifies the location of a secondary arrival. may or may not be present on the three other sections. Sections A-North and N-North both seem to show some energy arriving in approximately the same area as that seen in this region of figure 3.5, but it is much less distinct. Figure 3.5 shows additional energy coming in after this secondary arrival but it is very reverberative and has no distinctive phase coherency. The three other sections show a similar pattern. The amplitude of the first and secondary arrivals varies significantly from trace to trace, a characteristic also noted on the three other sections. 3 .1 .2.2 T h e T w o Of f l i ne S e c t i o n s . F - N o r t h A n d N - S o u t h Sections N-South and F-North are shown in figure 3.6: they have been bandpass filtered from 1-10 Hz and 3-10 Hz, respectively. These filtering parameters were chosen to make the low frequency secondary arrivals easier to pick, but in so doing the visibility of the lower amplitude first arrivals has been degraded. The first arrivals (a —a') show up " 3 5 1^50 170 190 210 830 250 270 290 310 330 North SHOT-RECEIVER DISTANCE (KM) South Figure 3 .6 Long offline sections recorded along line IV. (a) section F-North (filtered 1-10 Hz) , (b) section N-South (filtered 3-10 Hz) . A l l traces on both sections have been normalized to a common maximum amplitude, a—a' indicates the approximate location of the primary arr ival; b - b' and c — c' indicate the approximate location of secondary arrivals. best on N-South, but even with a filter tailored to enhance the visibility of these arrivals, they cannot be followed more than halfway across either section. Both section N-South and F-North have been plotted with each trace normalized to a common maximum amplitude. This was necessary because of an extremely large variation in trace to trace amplitude across the sections which interfered with the ability to pick some of the arrivals; therefore, the relative amplitude of arrivals on an individual trace are preserved but the relative amplitude of arrivals on adjacent traces cannot be compared. The apparent velocities of the primary (a -- a') and secondary (b — b' and c — c') arrivals on both sections is approximately 7 km/s and 7.2 km/s, respectively. Precise picks of the arrival times corresponding to the secondary arrivals are impossible given the poor data quality and the complex waveforms, so only approximate arrival times were used in modeling. 3.2 T h e Deep Reflection D a t a Set Deep reflection data were collected in the Vancouver Island region during both the 1984 LITHOPROBE and 1985 FGP marine surveys. The lines along which reflection data were gathered are shown in figure 2.1. Line 1 (line 84-01) of the 1984 experiment combined with line 1 (line 85-01) of the 1985 experiment provide a deep seismic reflection profile extending from the deep ocean, over the convergent margin, and across Vancouver Island. This reflection profile approximately coincides with seismic refraction line 1 of the 1980 VISP survey. Preliminary interpretations of the 1984 LITHOPROBE reflection experiment are given by Clowes et, a/., (1984), Yorath el al. (1985a), Clowes et al. (1986), and Green et al. (1986b). More detailed interpretations are provided by Yorath et al. (1985b), Green et al. (1986a), and Clowes et al. (1987a). A preliminary interpretation of the 1985 FGP marine experiment is given by Clowes et al. (1987b). 37 3.2.1 Data From The 1984 LITHOPROBE Survey Line 1 of the 1984 LITHOPROBE survey is the most important reflection line from the standpoint of this thesis, since it roughly coincides with much of VISP line 1 and intersects VISP line IV. Line 84-03 is also relatively important since it provides some three-dimensional control southeast of line 84-01 and comes very close to VISP line IV. Lines 84-02 and 84-04 of the LITHOPROBE experiment were recorded to resolve structural features on the southern tip of Vancouver Island. Although these lines do not intersect any VISP refraction lines, they do provide some insight into the possible structures near the southern end of VISP line IV. Figure 3.7a shows an immigrated reflection section corresponding to LITHOPROBE line 84-01 (figure 2.1), while figure 3.7b shows a line drawing interpretation of the section in 3.7a. There are two prominent reflecting zones below 3s (two-way traveltime). The top and bottom of the first zone will be referred to as Dj and D 2 , respectively; the top and bottom of the second zone will be referred to as Ej and E 2 , respectively. The two-way traveltimes to the top and bottom of these distinct bands of reflectivity were used in constructing a theoretical earth model that agreed with both the reflection and the refraction data collected along line 1. In addition to these obvious reflections, there are also some less prominent and less continuous reflections (F?) arriving below E 2 . The interpretation of these reflections in terms of geological structure is presented in Chapter 6. The three other reflection sections collected as part of the 1984 LITHOPROBE sur-vey show deep reflecting zones that correlate well with those seen on line 84-01, although many of the shallower reflections are quite different. This would seem to indicate that, while the shallower structures may change, the deeper structures (below about 4s) ob-(a) B 0 ^ C=D1« IIL i 6 5 I-4 ESBij~ t i g ? D2 E2 8 10 •KIP1* 27.2 -52.0 Figure 3.7 (a) Unmigrated seismic reflection section for southwestern two-thirds of line 84-01 recorded as part of the 1984 L I T H O P R O B E experiment, (b) line drawing of reflection section recorded along line 84-01. Letters indicate the location of prominent reflectors (from Clowes et ai., 1987a). 39 served in these four sections are probably common to all of southern Vancouver Island (Green et ai., 1986a). 3.2.2 Data From The 1985 F G P Marine Survey The 1985 marine survey consisted of five lines which are shown in figure 2.1. The one most relevant to this thesis is line 85-01 which roughly coincides with the marine portion of VISP line I. Lines 85-02 and 85-05 are also relatively important since they provide some three-dimensional control on line 85-01 and corroborate important features interpreted from it. Figure 3.8 shows the migrated reflection section corresponding to line 85-01. Many shallow reflectors can be seen and followed for a considerable distance, but the reflector that we are most interested in is reflector JdF which starts at about 5.5s (two-way traveltime) on the western edge of the section. This reflector, although faint in some areas, can be followed discontinuously from the western edge of the section to the eastern edge. Reflector F? identified on the reflection section recorded on line 84-01 of the 1984 LITHOPROBE survey is considered to be the same as reflector JdF (figure 3.8). Reflecting zones D and E identified on section 84-01 merge into a single reflector at the west coast of Vancouver Island (figure 3.7) and correspond to the zone of reflectivity between C and E on section 85-01 (figure 3.8). 3.3 Two Dimensional Modeling 3.3.1 The Modeling Algorithm The modeling algorithm used to calculate synthetic seismograms is based on asymp-totic ray theory (Spence, 1984; Spence et ai., 1984). The algorithm uses a raytracing Figure 3.8 Reflection section recorded along line 85-01 as part of the 1985 FGP marine survey. The letters C, E, and JdF correspond to the boundaries of prominent reflecting horizons. Reflectors C and E are identified in the same way in figure 3.7, while reflector JdF corresponds to reflector F?. R F refers to a ramp fault; DF refers to the deformation front; E B refers to Eocene basalts; M refers to the possible location of the oceanic Moho. The point at which another reflection line (85-05) intersects line 85-01 is indicated by the arrow (from Clowes et al., 1987b). •11 routine, described by Whittall and Clowes (1979), which calculates traveltimes for re-fracted rays, postcritically reflected rays, and "pseudo" head waves through a two-dimensional velocity model. The model is defined by polygonal blocks, each with an assigned velocity and linear velocity gradient. The velocity of each block is defined at the uppermost boundary and the orientation of the gradient is perpendicular to this boundary. Amplitudes are computed using zero-order asymptotic ray theory so that synthetic seismograms can be generated. 3.3.2 Modeling Procedure Starting models for lines 1 and IV were taken from previous interpretations of the data (Spence, 1984; Spence et a/., 1985; McMechan and Spence, 1983) combined with the additional constraints provided by the more recently acquired onshore-offshore deep reflection data. Rays are traced through the model and traveltimes are computed. These theoretical traveltimes are then compared with the observed traveltimes. If the agreement is unsatisfactory, the model is changed by repositioning boundaries and/or changing velocities and gradients within the polygonal blocks. This procedure of ray-tracing and adjusting the model is continued until a suitable traveltime fit is obtained. Figure 3.9 shows the raytracing diagram corresponding to section X35, the record sec-tion corresponding to receiver X35, and the traveltime picks for section X35 with the calculated traveltime curve superimposed. A visual comparison between the observed arrivals and the calculated arrivals was used to iterate to a suitable traveltime fit for all the data sections examined. Figure 3.9c. illustrates the quality of fit between the traveltimes calculated through the velocity model and those picked in the data. The same procedure was used for all the record sections examined and a similar quality of fit was obtained. 42 S H O T - R E C E I V E R DISTANCE (KM) cc -0 10 20 30 40 50 60 70 80 90 100 110 MODEL DISTANCE (KM) Figure 3.9 (a) The raytracing diagram corresponding to receiver section X35. (b) record section X35 (filtered 2.5-9 Hz) with traveltime picks (arrowheads) shown, (c) a comparison between the traveltime picks (dots joined by lines) in (b) and calculated traveltimes (x's). 43 The theoretical amplitudes corresponding to a particular arrival branch must ap-proximately agree with what is seen in the observed data. However, a model which provides a good traveltime fit may not produce a satisfactory amplitude match. The amplitude of a grout) of synthetic arrivals is primarily affected by the gradients in the layers through which the rays travel and by the velocity contrast across the model boundaries encountered by the rays. Therefore, by adjusting the velocities and gradi-ents in the layers along the ray path, the amplitude of a group of synthetic arrivals may be altered to match the amplitudes observed in the data. Ideally a satisfactory match can be obtained between the synthetic section and the observed data section by sim-ply adjusting the gradient in certain layers. A slight adjustment in the gradient often has a relatively small effect on the traveltimes but can have a significant effect on the synthetic amplitudes generated; however, occasionally the velocity structure used to fit the traveltime data is much different than that required to fit the amplitude data and substantial changes must be made in both velocities and gradients in order to fit both the observed traveltimes and amplitudes satisfactorily. 44 CHAPTER 4 INTERPRETATION OF SEISMIC REFRACTION DATA FROM LINE I OF VISP SURVEY 4.1 Initial Constraints The seismic refraction data collected along line I of the VISP survey were originally interpreted by Waldron (1982) and Spence (1984). Waldron modeled the OBS travel-time data using the Whittall and Clowes (1979) raytracer but he used the McMechan and Mooney (1980) ray trace/synthetic seismogram routine to model amplitudes. The McMechan and Mooney algorithm allows amplitude calculations for refracted and re-flected arrivals, but not for headwaves. Waldron's final two-dimensional velocity model was revised by D.J. White (personal communication, 1986) using the Spence et al. (1984) synthetic seismogram routine which makes use of the Whittall and Clowes ray-tracer and allows amplitude calculations for headwave arrivals in addition to refracted and reflected arrivals. This re-evaluation of the OBS data was carried out to model some secondary energy not modeled in the original interpretation of the data and to maintain consistency of format between the model developed for the OBS data and the onshore-offshore model developed by Spence (1984). Spence (1984) used all the refraction data collected on the land based receivers along line 1, as well as Waldron's velocity model, to develop an onshore-offshore model for the region. The interpretation of VISP line IV (McMechan and Spence, 1983) was used to constrain his onshore-offshore velocity model. The 1984 LITHOPROBE and 1985 FGP marine reflection surveys in the Vancouver Island region were designed to aid in the understanding of the structure in this area of plate convergence. As mentioned in Chapter 1, significant differences between the interpreted earth model proposed by Spence (1984) and that required by the reflection data prompted the reinterpretation of the onshore-offshore velocity model along VISP line I. The deep reflection sections recorded along lines 84-01 and 85-01 provide the primary constraints used in the reinterpretation of the refraction data. Figures 3.7 and 3.8 show the deep reflection profiles corresponding to the land and marine segments, respectively, recorded along lines 84-01 and 85-01. The upper and lower boundaries of prominent bands of deep reflections are identified with letters and it is these zones of high reflectivity that delineate the subsurface geometry required by the reflection data. The two-way traveltimes to the top and bottom of these zones of high reflectivity were used to constrain the velocity model developed for the onshore-offshore profile along VISP line I. That is, the vertical two-way traveltimes calculated through the velocity model to boundaries corresponding to these zones of high reflectivity were required to agree with the two-way traveltimes observed in the reflection data. Reflector JdF (figure 3.8) is probably the top of the subducting Juan de Fuca plate, based on the FGP marine reflection results (Clowes el a/., ]987b). This reflector is discontinuous but can be followed from the deep ocean to the west coast of Vancouver Island. The interruption in this reflector beneath the continental slope is probably due to energy loss associated with complex geologic features above the subducting plate. The location of reflector JdF becomes difficult to determine beneath most of Vancouver Island (figure 3.7). The increased dip of the subducting plate at a depth of approx-imately 40 km (figure 4.1) has been drawn to be consistent with the depths to the subducting slab under the Cascade volcanoes suggested by Dickinson (1970), based on the geochemistry of the lavas, and with the average world wide depth of about 100 km (e.g. Isacks and Barazangi, 1977). Reflecting zones D and E are clearly visible on reflection line 84-01 recorded on Vancouver Island (figure 3.7) and visible to a lesser extent on the offshore extension of this line (figure 3.8) where they have merged into a single layer (Clowes et a/., 1987a, 46 West East DISTANCE (KM) CM Figure 4.1 The velocity model corresponding to refraction line I, extending from the deep ocean off the west coast of Vancouver Island, across the island, and onto the mainland of British Columbia, (a) a 1:1 representation of the velocity model, (b) the velocity model with the velocity and gradient assigned to each polygonal block shown. The area indicated by the dashed lines in the upper left-hand corner of the model shown in (b) is enlarged and shown in (c). Velocities (km/s) are given for the top of each block, followed after the colon by the velocity gradient (km/s/km). Heavy lines in (b) indicate the boundaries for which the geometry has been constrained by the deep reflection data collected along lines 84-01 and 85-01. The hollow triangles at 60, 100, 150, 220, and 260 km indicate the location of five velocity depth profiles presented in figure 4.2. 1987b). The velocity between zones D and E in the revised onshore-offshore model (figure 4.1) is constrained by the revised model interpreted for line IV (figure 5.1) which requires a velocity in this region of approximately 7.15 km/s. Both reflectors have been extended as far eastward as the observed data permit. Figure 3.7 shows the two reflectors dying out about three-quarters of the way across the section. The reason they do not extend to the eastern edge of the section is unknown, but the point at which these reflections die out coincides with a major structural change in the velocity model which was required to replicate the traveltime delay seen in the refraction data recorded on the mainland side of Georgia Strait (e.g. sections P8, P13, and P19). 4.2 General Characteristics Of The Final Model The final composite model for the onshore-offshore line (figure 4.1) was derived from analysis of refraction and reflection data collected along VISP line I and along other seis-mic lines in the area. As previously stated, the reflection data have provided constraints on the subsurface geometry beneath Vancouver Island and the continental shelf by re-quiring that, the vertical two-way traveltimes calculated through the onshore-offshore velocity model to boundaries corresponding to zones of high reflectivity observed in the deep reflection data agree with those observed in the reflection sections recorded on lines 84-01 and 85-01. The velocity model is presented in figure 4.1 for reference during the following modeling discussion, and a series of velocity depth profiles corresponding to this model are shown in figure 4.2. The heavier lines beneath Vancouver Island and the continental shelf (figure 4.1) correspond to boundaries for which the geometry has been constrained by the deep reflection data. The two shots fired at shotpoint J (figure 2.1) and recorded on the mainland and Vancouver Island were modeled by Spence (1984). Since no new constraints are available for that portion of the onshore-offshore profile, it remains unchanged and was not remodeled. The portion of the model shown in figure VELOCITY IKM/S) V E I 0 C 1 I Y IKM/S) V E L O C M Y IKM/S) V t ' L O C H i IKM/S) VELDC 1 1 Y IKM/Sl I 3 5 7 91 3 5 7 <J I 3 5 7 9 ] 3 5 7 9 1 3 5 7 60 K M 100 K M 150 K M 220 K M 300 K M Figure 4.2 Velocity depth profiles at five locations indicated in figure 4.1. The distances at the bottom of each V(z) profile correspond to the location of the profile with respect to shot P19 at the western edge of figure 4.1. 49 4.1c was originally interpreted-by Waldron (1982) and subsequently reinterpreted by D.J. White (personal communication, 1986); only minor changes were made during the course of this reinterpretation. As will be seen in the following sections, the principal features of the model which have been derived during the course of this study are: (l) a subducting plate which dips approximately 3° beneath the continental slope and between 14° and 16° beneath the continental shelf and Vancouver Island, and (2) a four layer structure beneath the island which consists of a twice repeated sequence of a low velocity zone (« 6.4 km/s) above a thicker high velocity zone (« 7.1 km/s) for the underplated region directly above the subducting Juan de Fuca plate. 4.3 I n t e r p r e t a t i o n Of T h e P -se r i es S e c t i o n s The data examined in this section were recorded at receivers on Vancouver Island and the British Columbia mainland from shots detonated off the west coast of Vancou-ver Island. The primary and secondary arrivals recorded at the receiver locations are modeled as energy which has traveled through the lower lithosphere of the subducting plate and passed through material beneath Vancouver Island and the mainland on its way to the receiving stations. By modeling these data, the velocity structure in the lower lithosphere of the subducting plate and beneath Vancouver Island and the mainland can be ascertained. The distance scale on all the data sections presented is shot-receiver distance, while the distance scale on the corresponding synthetic seismograms and the raytracing diagrams refers to model distance. Although the two distance scales will be different on all the sections presented except section Pi9, since shot Pi9 was at the ori-gin of the velocity model, they correspond to the same region. Representative wavelets were chosen from the observed data and the wavelet which most closely matched the dominant waveform on a particular observed data section was used in generating the 50 corresponding synthetic section. As mentioned in Chapter 3, all the data sections have been plotted so that the amplitude of each trace is proportional to y f ^ , where d is the distance from the shot. 4.3.1 S e c t i o n P19 Shot P19 was fired at the extreme western end of VISP line I. Figure 4.3 shows the rays traced from shotpoint P19 through the onshore-offshore velocity model, the corresponding traveltime curves superimposed on section P19, and the synthetic section generated. The first arrivals observed on section P19 (figure 4.3b) are relatively easy to follow for the mainland stations, but are very difficult to see on the Vancouver Island stations. Because precise time picks could only be made on a few of the traces, the trend and approximate start of first arrival energy were all that could be modeled. The first arrival traveltimes calculated through the velocity model agree well with the arrivals observed on section P19. Turning rays through the oceanic upper mantle give rise to the first arrivals at model distances which correspond to Vancouver Island and the mainland (figure 4.3a). The oceanic upper mantle velocity ranges from 8.10 km/s to approximately 8.16 km/s in the westernmost block of the velocity model, and from 8.17 km/s to approximately 8.33 km/s in the easternmost block. The division of the upper mantle into a series of blocks which gradually increase in velocity from the west to the east reflects a lateral increase in velocity toward the east which was required to satisfactorily match traveltimes observed in the data. The most conspicuous feature of the observed data, and one that, is evident on all the P-series sections, is the offset in traveltime across Georgia Strait (figure 4.3b). In the model this difference arises from the difference in the raypaths taken to the Vancouver Island stations as compared to that taken to the mainland stations. The rays arriving at the Vancouver Island stations travel through material beneath Vancouver 51 i>>>>>>>>>'?^<?> 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 MODEL DISTANCE (KM) Figure 4.3 (a) The raytracing diagram for a shot at shotpoint P19. (b) the theoretical traveltime curve superimposed on section P19 (filtered 2.5-9 Hz), (c) the synthetic section corresponding to the raytracing diagram in (a). The wavelet used in (c) was chosen to represent the average waveform of the first arrivals in (b). Island which has a higher average velocity than material at the same depth beneath Georgia Strait and the mainland. The velocity depth profiles corresponding to the onshore-offshore model (figure 4.2) illustrate the difference in velocity structure used to model the traveltime offset of 0.75s observed in the data between energy arriving at the Vancouver island receivers and energy arriving at the mainland receivers. The synthetic seismogram section corresponding to the raytracing diagram (figure 4.3a) is shown in figure 4.3c. ln the observed data, the first arrivals at the Vancouver Island stations have, in general, a lower amplitude than the first arrivals at the mainland stations (figure 4.3b). This feature was not reproduced particularly well in the synthetics where the first arrivals at the Vancouver Island receivers have a slightly higher amplitude than arrivals at the mainland receivers (figure 4.3c). The boundary within the upper mantle which intersects the western edge of the model at roughly 27 km depth is modeled as having a velocity of 7.7 km/s. However, the velocity at this boundary is poorly constrained since the observed data do not require refracted rays traveling through material below it. Arrivals reflected from this boundary (figure 4.1) are used to model a possible secondary arrival seen on traces at shot-receiver distances of approximately 208, 220, 240, and 250 km on section P19 (figure 4.3b). Reflected arrivals could not be traced to the mainland stations for any of the P-series shots modeled (P8, P13, and P19) because of the blocky structure required to specify the model; this was not considered a problem since none of the data (sections P8, P13, and P19) require a set of secondary arrivals at the mainland stations. Spence et a/. (1985) also discuss the lack of evidence for secondary arrivals in the data corresponding to the mainland receivers. The amplitude of the secondary arrivals observed at the Vancouver Island stations is the same size or slightly larger than the amplitude of the first arrivals (figure 4.3b). The traveltime difference between these secondary arrivals and the first arrivals becomes less and less, moving from the west to east across the traces recorded on Vancouver Island. The synthetic section corresponding to shot P19 shows no distinct secondary arrival. Because the calculated traveltimes for the primary and secondary arrivals are relatively close, they appear as a single strong arrival on the synthetic section (figure 4.3c). ln addition to the secondary energy modeled for the Vancouver Island receivers, there also appears to be a later arrival at approximately 5s on the westernmost mainland receiver (figure 4.3b); however, it does not appear on any of the other P-series sections examined, and therefore no attempt was made to model it. 4.3.2 S e c t i o n P13 Shot P13 was fired approximately 40 km northeast of shot P19 and at approximately the same depth. Of the three P-series shots used in modeling, the seismic section recorded from shot P13 has the best quality data and most clearly exhibits features that are less clear on the other two sections for which detailed remodeling has been carried out (P8 and P19). Figure 4.4 shows the raytracing diagram for a shot at shotpoint Pi3, the observed data section with calculated traveltime curves superimposed, and the synthetic section corresponding to the raytracing diagram. The first arrivals are relatively easy to follow across the entire observed data section, although the precise start of the first break is not easy to pick. Both filtered and unfiltered data sections were examined to determine the start of first arrival energy. The traveltime offset between the first arrivals on the Vancouver Island receivers and the arrivals on the mainland receivers is clearly observed in figure 4.4b and is the most conspicuous feature of the data. The calculated traveltimes for first arrivals arriving at both the Vancouver Island and mainland stations agree relatively well with the arrivals observed on section P13. A possible set of secondary arrivals is seen on the Vancouver Island stations approximately 54 DISTANCE (KM) 0 40 80 120 160 200 240 280 320 360 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 MODEL DISTANCE' (KM) Figure 4.4 (a) The raytracing diagram for a shot at shotpoint P13. (b) the theoretical traveltime curve superimposed on section P i3 (filtered 2.5-9 Hz), (c) the synthetic section corresponding to the raytracing diagram in (a). The wavelet used in (c) was chosen to represent the average waveform of the first arrival in (b). 0.3s after the first arrivals. These arrivals are modeled as being reflections from the boundary within the upper mantle which was used to generate a similar set of secondary arrivals observed on section P19. These wide-angle reflections traced to the Vancouver Island stations produce theoretical traveltimes which agree reasonably well with the start of a possible secondary arrival (figure 4.4b). The synthetic section corresponding to the raytracing diagram (figure 4.4a) is shown in figure 4.4c. The average amplitude of the first arrival observed on the Vancouver Island receivers is reproduced relatively well in the synthetic section, with the amplitude of the second arrival being a little lower than that seen in the data. The amplitude of these wide-angle reflections is sensitive to the magnitude of the velocity contrast at the boundary from which they are reflected. After assigning velocities which ranged from 7.5 to 8.6 km/s to the material below this upper mantle reflector it was found that a. velocity of 7.7 km/s provided the best overall fit to the data collected along line I; however, a velocity of 8.6 km/s did provide a better fit to the secondary arrivals recorded on Vancouver Island from shots P13 and P19. The average amplitude of the first arrival recorded on the mainland receivers is larger than that produced in the synthetic section (figure 4.4b and 4.4c). 4.3.3 Section P8 Shot P8 was fired over the ocean basin near the base of the continental slope and about 68 km northeast of shot P19. Section P8 is the most difficult of the three sections to interpret because of its low signal-to-noise ratio. Figure 4.5a shows the rays traced from shot P8 through the velocity model to the receivers on Vancouver Island and the mainland. Figures 4.5b and 4.5c show the theo-retical traveltime curves superimposed on section P8 and the synthetic section generated from the raytracing run, respectively. Although the quality of the data is quite low, DISTANCE (KM) 160 200 360 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 SHOT-RECEIVER DISTANCE (KM) 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 MODEL DISTANCE (KM) Figure 4.5 (a) The raytracing diagram for a shot at shotpoint P8. (b) the the-oretical traveltime curve superimposed on section P8 (filtered 2.5-9 Hz), (c) the synthetic section corresponding to the raytracing diagram in (a). The wavelet used in (c) was chosen to represent the average waveform of the first arrival in (b). 57 a clear band of first arrival energy is relatively easy to follow across the section. The calculated traveltimes agree reasonably well with the approximate location of the ob-served first breaks (figure 4.5b). Precise first break picks were not possible because of the low signal-to-noise ratio. Reflected arrivals were generated for the Vancouver Island stations, although no distinct secondary arrivals are observed in the data recorded at these stations. The synthetic seismogram section corresponding to the raytracing diagram (figure 4.5a) is shown in figure 4.5c. The average relative theoretical amplitudes of the first arrivals at the Vancouver Island stations compared to those for the mainland stations match the data quite well. The reflected arrival at the Vancouver Island stations is not required by the data, but adds reverberation which is observed in the data. 4.4 I n t e r p r e t a t i o n O f X - s e r i e s S e c t i o n s The data modeled in this section were recorded on four separate receivers, two on Vancouver Island and two on the mainland. All four receivers recorded each of the 17 P-series shots fired off the west coast of Vancouver Island. The common receiver gather sections (X-series sections) allow an independent check on the velocity structure modeled from the common shot gather sections (P-series sections). Modeling of the X-series data resulted in only minor changes to the velocity model. The energy arriving at the two Vancouver Island receivers is modeled as rays travel-ing through the lower lithosphere of the subducting plate and up through the material beneath Vancouver Island. Energy arriving at the two mainland receivers is modeled in the same way except that the rays travel up through material beneath Georgia Strait and the mainland. The distance scale on all four data sections is still shot-receiver dis-tance, but in this case it increases from right to left since the receiver location is taken as being the origin of the data section rather than the shot location, as in section 4.3. 58 -The distance scale on the corresponding synthetic sections, generated for each observed data section, is again model distance. Although the two distance scales display different values, they both correspond to the same region. 4.4.1 Sections X6 And X22 Receivers X6 and X22 were both located on Vancouver Island. Each receiver recorded the 17 P-series shots which extended from the deep ocean (P8 — P19) to shots over the continental slope (Pi — P6). The sections recorded at receivers X6, near the west coast of Vancouver Island, and receiver X22, near the east coast of the island, were modeled by tracing rays from the receiver location to the approximate location of the line of offshore shots. Figures 4.6a and 4.7a show the final raytracing diagrams cor-responding to receiver sections X6 and X22, respectively. Figures 4.6b and 4.7b show the traveltime curves corresponding to the raytracing diagrams superimposed on the observed data sections; figures 4.6c and 4.7c show the theoretical seismic sections. The first arrival energy observed on both section X6 and X22 is relatively easy to fol-low across each section. The apparent velocities of these arrivals across the ocean basin shots for receiver X6 and X22 are approximately 8.6 km/s and 8.7 km/s, respectively. Turning rays through the lower oceanic lithosphere are used to model the first arrivals observed on both sections. The calculated first arrival times match those observed in the data quite well (figures 4.6b and 4.7b), although these calculated traveltimes are up to 0.22s too late for traces at distances less than 135 and 175 km for sections X6 and X22, respectively. These traces correspond to shots fired over the continental slope (shots P i — P6) and the time correction required to put these shots at the 2600m datum was significant (see section 2.5.1). These topographic corrections were calculated using the velocity structure for the continental slope modeled by Waldron (1982). Because sedi-mentary and upper crustal structures were not as well constrained over the continental 59 D I S T A N C E ( K M ) 0 40 80 120 160 200 240 280 320 360 C\H 1 1 1 1 1 1 1 —1 1 1 f 1 1 1 1 1 1 1 r 1 • 1 ' 0 10 20 30 40 50 60 70 80 90 100 110 M O D E L D I S T A N C E ( K M ) Figure 4.6 (a) The raytracing diagram for receiver section X 6 . (b) the theo-retical traveltime curve superimposed on section X 6 (filtered 2.5-9 Hz) , (c) the synthetic section corresponding to the raytracing diagram in (a). The wavelet used in (c) was chosen to represent the average waveform of the first arrival in (b). 60 DISTANCE (KM) 0 40 80 120 160 200 240 280 320 360 i OJ-t 1 1 1 1 1 1 1 I 1 1 1 1 1 1 1 1 1 1 I 1 1 0 10 20 30 40 50 60 70 80 90 100 110 MODEL DISTANCE (KM) Figure 4.7 (a) The raytracing diagram for receiver section X22. (b) the theo-retical traveltime curve superimposed on section X22 (filtered 2.5-9 Hz) , (c) the synthetic section corresponding to the raytracing diagram in (a). The wavelet used in (c) was chosen to represent the average waveform of the first arrival in (b). 61 slope as they were for the ocean basin (Waldron, 1982), the time corrections associated with placing shots P i to P6 at a datum of 2600m are more poorly constrained than the time corrections required to place shots P8 to P19 at the 2600m datum. Therefore, the fact that the calculated first arrival traveltimes do not agree quite as well with the observed first arrivals on traces corresponding to shots P i to P6 as those for shots P8 to P19 may be related to the less well constrained time corrections applied to these traces. The offset in arrival time observed on the P-series sections is not present on either section X6 or X22 since all the rays traveling from these receivers to the 17 P-series shots pa,ss through the higher velocity material beneath Vancouver Island. Reflected arrivals were generated from the upper mantle reflector and the calculated traveltimes match the observed traveltimes of a possible secondary arrival seen in the data (figures 4.6b and 4.7b). This secondary arrival is most clearly seen on section X22, although it only appears as a distinct arrival on traces at distances greater than 180 kin. The cor-responding traveltimes calculated for comparison with these secondary arrivals extend from approximately 165 km to 250 km (shot-receiver distance). These arrivals could not be generated for distances closer to receiver X22 than 165 km due to the blocky structure required to specify the model. This is not considered a problem since the data on section X22 do not show secondary arrivals closer to the shot than 180 km (shot-receiver distance). The possible secondary arrivals observed on section X6 extend from approximately 129 km to 208 km (shot-receiver distance). The calculated traveltime corresponding to these arrivals extends across the entire section. The synthetic seismograms for receiver X6 and X22 are presented in figures 4.6c and 4.7c, respectively. No attempt was made to try and reproduce the trace to trace fluctuation in amplitudes observed in the data, but the synthetic seismograms for both section X6 and X22 do reproduce the general trend in amplitudes seen in the data. In general, the synthetic secondary arrivals have a slightly lower amplitude than the observed secondary arrivals. 4.4 .2 S e c t i o n s X 3 5 A n d X 4 5 Receivers X35 and X45 were both located on the mainland of British Columbia, and they both recorded the 17 P-series shots iired off the west coast of Vancouver Island. Each section was modeled by tracing rays from the location of the receiver across the continental margin to the offshore portion of line 1. Figures 4.8a and 4.9a show the raytracing diagrams corresponding to X35 and X45 respectively, while figures 4.8b and 4.9b show the corresponding traveltime curves superimposed on the observed data. Figures 4.8c and 4.9c show the synthetic sections generated from the rays traced in 4.8a and 4.9a, respectively. First arrivals for X35 and X45 are relatively clear for traces beyond 240 km and 280 km, respectively. Traces that are closer to the receivers than this are more difficult to pick than they were on sections X6 and X22. This is probably due to the increasing shot-receiver distance and the geologic complexity associated with the continental slope deposits. The apparent velocity of the first arrival across both the observed data sections ranges from approximately 8.0 km/s to 8.9 km/s. Turning rays through the lower oceanic lithosphere were used to model the first arrivals on both section X35 and X45. The traveltimes calculated for these arrivals match what is seen in the observed data relatively well with only small deviations from the calculated traveltime for individual traces (figures 4.8b and 4.9b). It becomes more difficult to assess the first arrival traveltime fit for traces at distances less than 240 km for section X35 and at distances less than 280 km for section X45, since those traces have a much lower signal-to-noise ratio than traces at greater distances. However, the calculated traveltimes corresponding to those traces appears to agree with the probable arrival of first break energy. The 63 0 10 20 30 40 50 60 70 80 90 100 110 MODEL DISTANCE (KM) Figure 4.8 (a) The raytracing diagram for receiver section X35. (b) the theo-retical traveltime curve superimposed on section X35 (filtered 2.5-9 Hz), (c) the synthetic section corresponding to the raytracing diagram in (a). The wavelet used in (c) was chosen to represent the average waveform of the first arrival in (b). 64 DISTANCE (KM) 0 40 80 120 160 200 240 280 320 360 -I 1 1 . i 1 1 '. i i : i i 0 10 20 30 40 50 60 70 80 90 100 110 MODEL DISTANCE (KM) Figure 4.9 (a) The raytracing diagram for receiver section X45. (b) the theo-retical traveltime curve superimposed on section X45 (filtered 2.5-9 Hz), (c) the synthetic section corresponding to the raytracing diagram in (a). The wavelet used in (c) was chosen to represent the average waveform of the first arrival in (b). traveltime jump observed on the P-series sections is not present on section X35 or X45 since the ray paths are not split between traveling through high velocity material beneath Vancouver Island and the lower velocity material beneath the mainland. The synthetic arrivals corresponding to the observed first arrivals for receivers X35 and X45 are presented in figures 4.8c and 4.9c, respectively. The synthetic first arrivals for section X35 agree relatively well with the observed data, although synthetic arrivals at model distances less than 50 km are of lower amplitude than corresponding arrivals in the observed data. The first arrivals generated for comparison with section X45 also agree with the data relatively well, although again the amplitudes of the synthetic sections generated at model distances less than 50 km are in general lower than those for corresponding arrivals in the data. 4.5 Final Velocity Model The final velocity model based on the complete raytracing interpretation was pre-sented in figure 4.1. The model can be divided into three main subsurface regions: the oceanic, island, and continental regions. The determination of these regions has been based solely on subsurface structure and does not necessarily reflect the location of surface features (e.g. Vancouver Island). The oceanic region extends from the western edge of the model to approximately 128 km. This portion of the model (down to about 15 km) was originally interpreted by Waldron (1982). D.J. White (personal communication, 1986) reinterpreted the data in this region and made a. number of modifications to Waldron's original velocity model. No significant changes have been made to this portion of the revised model in reinterpreting the onshore-offshore data along line I. The velocity of the oceanic sediments in the western half of this region range from 1.8 km/s to approximately 2.3 km/s, with virtually no gradient. The continental slope deposits further east are thicker and more complex, 06 with starting velocites ranging from 1.90 km/s to 3.25 km/s in the upper sediments and from 4.4 km/s to approximately 5.9 km/s in the deeper melange unit which overlies the subducting oceanic crust (Waldron, 1982). The location of the top of the subducting oceanic crust has been constrained not only by the refraction data collected along line I, but also by the deep reflection data collected along line 85-01 of the 1985 FGP marine survey. The oceanic crust is approximately the same thickness under all three structiiral components of the velocity model. The velocity of this crustal material ranges from 6.0 km/s to approximately 7.3 km/s. Velocities in the uppermost mantle range from 8.10 — 8.16 km/s in the west, and from 8.17 — 8.33 km/s in the easternmost block. The velocity assigned to the material beneath the upper mantle reflecting boundary is poorly constrained, but has been modeled as 7.7 km/s. The island region extends from 128 km to 240 km. This region coincides with only the western two-thirds of Vancouver Island, while the eastern third of the island overlies what has been designated as the continental region. The subsurface structure of the island region has been constrained not only by the refraction data collected along line 1, but also by the refraction data collected along line IV which intersects the onshore-offshore model at approximately 230 km (figure 4.1). Further constraints on the island region are provided by the deep reflection data collected along line 84-01 of the 1984 LITHOPROBE survey, and it is on the basis of these data that major changes were made in the island region of the revised model compared to the velocity model proposed by Spence et ah (1985). Model velocities down to approximately 10 km are the same as those used by Spence et a). (1985), while below 10 km the velocity structure differs r considerably from that used by Spence. The high velocity (7.7 km/s) wedge proposed by Spence et a/. (1985) has been replaced by a slightly lower velocity (7.15 km/s) wedge. This velocity of 7.15 km/s is constrained by the revised velocity models for line fV (see section 5.3). This high velocity wedge is surrounded by lower velocity material (6.35 67 km/s) (figure 4.1). Directly beneath this is another high velocity (7.1 km/s) wedge underlain by the top of the subducting oceanic crust. The continental region extends from 240 km to the eastern edge of the velocity model (figure 4.1). The upper 20 km of this part of the velocity model was originally modeled by Spence (1984) using data obtained from two shots at shotpoint J, on the eastern end of the model (figure 2.1). No new constraints could be imposed on the upper portion of the continental region, so it remains unaltered with velocities increasing from 5.3 km/s at the surface to 6.95 km/s at a depth of 20 km. Below 20 km the only real constraint on the velocity structure is the requirement that a velocity-thickness combination be used which produces the 0.75s traveltime delay observed at the mainland receivers relative to the Vancouver Island receivers. The velocity structure used in the revised model (figure 4.1) is similar to that used by Spence, with the exception that the continental Moho has been given a slight dip towards the mainland to delay arrivals at mainland receivers. A thick low velocity zone with a positive velocity gradient (6.4;0.03) (figure 4.1) has replaced a similar layer with a negative gradient (6.97;-0.044) used by Spence et al. (1985). The average velocity within this low velocity zone is approximately 0.1 km/s greater than that used by Spence, and since a positive gradient turns rays back toward the surface rather than away from the surface, as a negative gradient does, the positive gradient was preferred. < - 68 CHAPTER 5 INTERPRETATION OF SEISMIC REFRACTION DATA FROM LINE IV OF VISP SURVEY 5.1 Initial Constraints As described in Chapter 3. the seismic refraction data collected along line IV of the VISP survey arc of reasonably good quality for the two reversed segments (N-A and A-F) but of very poor quality for the longer offline sections (N-South and F-North). These data were originally interpreted by McMechan and Spence (1983) and formed the basis of the current interpretation, especially for the upper 10 km. Three particu-lar points indicate the necessity for this reinterpretation. First, a more thorough and careful analysis of the data showed secondary arrivals different, from those modeled by McMechan and Spence (1983). Secondly, the original interpretation was not constrained by the requirement of being consistent with line I at their point of intersection because line I had not been interpreted. Finally, the 1984 LITHOPROBE reflection survey on southern Vancouver Island provided new constraints on the subsurface geometry be-neath the southern portion of line IV. LITHOPROBE line 84-01 crosses VISP line IV, and the southern end of line 84-03 comes very close to line IV (figure 2.1). Although lines 84-02 and 84-04 on the southern tip of Vancouver Island are not particularly close to the southern end of line IV, they do provide insight into the regional structure on the southern part of the island. Figure 3.7 shows the deep reflection section recorded along line 84-01. Line IV intersects this profile at a point approximately halfway along line 84-01. Constraints on the velocity model were imposed by ensuring that the two-way traveltimes determined from the model velocity depth profile at the point of intersection were consistent with those to the upper and lower limits of reflecting zones D and E. The same procedure 69 was used at the southern end of line 84-03 which terminates within 5 km of line IV. Lines 84-02 and 84-04 were used to a lesser extent to control the approximate location of the same reflecting horizons at the southern end of line IV. 5.2 General Characteristics Of The Final Models With the additional constraints provided by the 1984 LITHOPROBE survey, two slightly different velocity models were developed to model the refraction data collected along line IV. As previously stated, the deep reflection data have provided constraints on the subsurface geometry beneath southern Vancouver Island by requiring that the vertical two-way traveltime to zones of high reflectivity agrees with that calculated through the velocity model to the corresponding boundaries. The two velocity models are presented in figures 5.1 and 5.2 for reference during the following modeling discus-sion. The reason two models were needed has to do with the bend in line IV (figure 2.1) and is discussed in section 3.1. Figure 5.1 will be referred to as the hinged model since it is based on data recorded along the two reversed segments of line IV (N-A and A-F, figure 2.1) from shots at shotpoints N, A, and F. The model shown in figure 5.2 is based solely on data from shots at shotpoints N and F, recorded along the full length of line IV with emphasis placed on data recorded along the northern segment N-A from shotpoint F and along the southern segment A-F from shotpoint N. The receivers for these shots were not inline with the shots themselves. Because of this, the rays traveled through material to the west of line IV (figure 3.2). Since every ray sampled a different piece of the shaded region shown in figure 3.2, the model for this area represents the average velocity structure in that region. The heavier lines beneath the southern portion of line IV (figure 5.1) correspond to boundaries for which the geometry has been constrained by the LITHOPROBE reflection data. As will be seen in the following sections, the principal features of North South DISTANCE (KM) (a); 0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 o -( 1 1 1 ' 1 1 1 I • I . J I I I I I I I ^ o S _ m sc cvj E-a. W Q o. ID (b) N DISTANCE (KM) E - W ^ I N E F 0 20 40 60 80 100 120 140*160 180 200 220 240 260 280 300 320 340 i.6:.\ 5.5;.05 5.5;.29 6.35:. 005 7.15:. 00* 4.J5.-.005 6.86:.16 Figure 5.1 Velocity model for the hinged model, (a) a 1:1 representation of the velocity model, (b) the velocity model with the velocity and gradient assigned to each polygonal block shown. The letters N, A, and F indicate the locations of the three shotpoints along line IV. The arrowhead indicates the line of intersec-tion between this model and the onshore-offshore model (figure 4.1). Velocities (km/s) are given for the top of each block, followed after the semicolon by the velocity gradient (km/s/km). The thicker lines represent boundaries for which the geometry has been constrained by the LITHOPROBE reflection data. the model which have been derived during the course of this study are: (l) a well constrained velocity structure to a depth of approximately 15 km along the entire length of Vancouver Island, (2) a velocity structure beneath southern Vancouver Island which is consistent with the LITHOPROBE reflection data, and (3) a model for line IV which agrees with the revised model developed for line I at their point of intersection. For lack of additional information and because line IV runs approximately parallel to the strike of the subducting plate, layers defined beneath the southern portion of Vancouver Island have been continued to the northern end of the model. As shown in the 1:1 71 North South DISTANCE (KM) (a) 0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 S N DISTANCE (KM) 8.17 :.016 Figure 5.2 Velocity model corresponding to the offline model which represents the average velocity structure between line IV and a line connecting shotpoints N and A. (a) a 1:1 representation of the velocity model, (b) the velocity model with the velocity gradient assigned to each polygonal block shown. The letters N and F indicate the locations of the two shotpoints used in modeling this line. representation of both models (figures 5.1a and 5.2a) the general characteristics indicate a sub-horizontal layered medium. 5.3 Interpretation Of The Hinged Model The data examined in this section were recorded on receivers located along line IV (figure 2.1). The primary and secondary arrivals recorded on these receivers are modeled as energy which has traveled for the most part through continental material above the subducting oceanic, plate. By modeling these data, the velocity structure of the continental material beneath Vancouver Island can be ascertained. The distance scale on all the observed data sections is shot-receiver distance. Representative wavelets were chosen from the observed data and the wavelet which most closely matched the 72 dominant waveform on a particular observed data section was used in generating the corresponding synthetic section. As mentioned in Chapter 3, all the data, sections have been plotted so that the amplitude of each trace is proportional to y ^ , where d is the distance from the shot. 5.3.1 Reverse Profile Recorded Along Segment N-A Both section N-North and section A-North were recorded along the northern seg-ment of line IV from shots at N and A respectively. Figure 5.3 shows the rays traced from shotpoint N through the hinged velocity model, the corresponding traveltime curves su-perimposed on section N-North, and the synthetic seismogram section generated. Figure 5.4 shows the raytracing diagram for a shot at A recorded along the northern portion of line IV, the corresponding traveltime curves superimposed on the observed data, and the synthetic seismogram section. The first arrivals observed on both section N-North and section A-North are easy to follow across each section (figures 5.3b and 5.4b). The first arrival traveltimes calcu-lated for section N-North agree well with those observed in the data. The first breaks are modeled as successive families of refractions through layers in the upper 15 km of the model. The shallowest refractions travel within the uppermost layer which ranges between 1 and 3 km thick across the entire model and has velocities which range from 5.5 to 6.2 km/s. The uppermost layer in the model proposed by McMechan and Spence (1983) was between 1.5 and 2 km thick and was discontinous across the model, with a gap of roughly 100 km in the center. The velocity in the uppermost layer is virtually the same as that used by McMechan and Spence except for the northern 30 km of the velocity model, where the revised model includes a velocity of between 6.0 and 6.2 km/s as opposed to the 5.5 km/s velocity used by McMechan and Spence. This higher veloc-ity is required by the data in order to get a closer fit between the calculated traveltime 73 DISTANCE (KM) 0 20 40 60 00 100 120 140 160 MODEL DISTANCE (KM) Figure 5.3 (a) The raytracing diagram for a shot at shotpoint N, (b) the theo-retical traveltime curve superimposed on section N-North, and (c) the synthetic section corresponding to the raytracing diagram in (a). The numbers 1-6 in (a) identify layers discussed in the text. The wavelet used in (c) was chosen to represent the average waveform of the first arrival in (b). 74 20 40 60 80 DISTANCE (KM) 100 120 140 160 180 200 220 240 260 280 300 320 340 160 140 120 100 80 60 40 20 0 SHOT-RECEIVER DISTANCE (KM) 0 20 40 60 80 100 120 140 160 MODEL DISTANCE (KM) Figure 5.4 (a) The raytracing diagram for a shot at shotpoint A , (b) the the-oretical traveltime curve superimposed on section A-Nor th (filtered 6-12 Hz) , and (c) the synthetic section corresponding to the raytracing diagram in (a). The numbers 1-6 in (a) identify layers discussed in the text. The wavelet used in (c) was chosen to represent the average waveform of the first arrival in (b). 75 for rays turning in the first layer of the velocity model and the first refraction branch observed in the data. Turning rays through the uppermost layer of the velocity model generate synthetic arrivals which agree with high amplitude arrivals observed on both sections N-North and A-North. Due to the extremely large amplitude of these arrivals certain traces in the data which were near the shot had to be deleted because the large amplitude arrivals interfered with the ability to see arrivals on adjacent traces. The second refraction branch on both N-North and A-North is modeled as a family of rays refracting through the second layer of the velocity model (figures 5.3a and 5.4a). This layer ranges in velocity from 6.43 km/s to approximately 6.8 km/s. The calculated traveltimes corresponding to section N-North agree with those observed in the data, with the exception of a few individual traces which come in either slightly earlier or later than the calculated traveltime curve. The first arrival traveltimes calculated for comparison with section A-North are in general slightly earlier than the observed first arrival. The general amplitude trend across the second refraction branch on both N-North and A-North (figures 5.3b and 5.4b) is one in which the closest traces exhibit amplitudes which are considerably smaller than those of the refracted arrival through the first layer, and gradually become larger toward the far end of the refraction branch. This general trend is reproduced in the synthetic seismogram sections corresponding to N-North and A-North (figures 5.3c and 5.4c). There is a considerable amount of fluctuation in the amplitude from one trace to another on the observed record section and no attempt was made to try to reproduce anything but the general trend in amplitudes across the traveltime branch. The final branch of refracted arrivals observed on section N-North and A-North is modeled as rays refracting through the fourth layer of the velocity model (figures 5.3a and 5.4a). The third layer is a low velocity zone with velocities ranging from 6.35 km/s 7G to 6.41 km/s and corresponds to reflector band D, seen in the LITHOPROBE reflection data recorded on southern Vancouver Island (figure 3.7). The velocity in the fourth layer ranges from 7.0 km/s to approximately 7.2 km/s. The fourth layer corresponds to the area between the D and E reflectors in the LITHOPROBE reflection data. The theoretical traveltimes corresponding to this final group of refracted arrivals agree well with the observed data for both N-North and A-North (figures 5.3b and 5.4b). The amplitude of these arrivals in the data is relatively low. The corresponding synthetic arrivals are also low amplitude (figures 5.3c and 5.4) and compare favorably with those observed in the data. Reflected arrivals were generated from the top of the fourth layer as well as the four deepest layers in the model for both shotpoints N and A. The theoretical arrival times corresponding to these reflections are superimposed on the observed data in figures 5.3b and 5.4b respectively. The calculated arrival times correspond to similar arrivals seen in the data for the greatest shot-receiver offset traces (130-160 km shot-receiver distance). Although these arrivals cannot be followed (in the case of N-North and A-North) to smaller shot-receiver offset traces, the genera! amplitude trend produced in the synthetics agrees with what is observed in this part of the data (50-110 km on section N-North and 80-110 km on section A-North). The deeper reflected arrivals are generated from boundaries which are included in the model in order to maintain consistency between the hinged model and the onshore-offshore model. The first deep reflected arrival is off the northwestern edge of the conti-nental Moho, while the others are reflected from layers within the underlying subducting plate. The traveltimes for the first two reflections are so close on the synthetic seismo-gram section corresponding to N-North (figure 5.3c) that their wavelets have merged to form a single arrival. The same is true for the norhternmost traces on the synthetic section corresponding to section A-North (figure 5.4c). The location of the continental 77 Moho is not well constrained by the refraction or reflection data and therefore it is not required in the hinged model, but has been included to maintain consistency with the revised onshore-offshore model. The arrivals generated by reflections from these bound-aries appear to agree with similar features seen on section A-North and to a lesser degree with features on section N-North, although the data do not require these arrivals on either section. Figure 5.5 is included to illustrate the difference in the synthetic section generated when the continental Moho in figure 5.3 is removed. Thus the first deep reflected arrivals are generated from the top of the subducting plate rather than the continental Moho as in figure 5.3. The only difference between the synthetics in figure 5.3c and those in figure 5.5c is that the amplitude of the first deep reflected arrival, located at roughly 6.0s at 160 km model distance, is smaller than that generated when the continental Moho is included in the model. 5.3.2 Reversed Profile Recorded Along Segment A-F Sections A-South and F-South were recorded along the southern portion of line IV from shots at shotpoints A and F respectively. Figure 5.6 shows the rays traced from shotpoint A through the velocity model, the corresponding traveltime curves superim-posed on section A-South, and the synthetic section generated from the raytracing run. The raytracing diagram for a shot at F recorded along the southern portion of line IV is shown in figure 5.7 along with the corresponding traveltime curves superimposed on the observed data, arid the synthetic section generated. The first arrivals on both section A-South and F-South are easy to follow across each section. The first breaks are modeled as successive families of refractions through the upper 20 km of the velocity model. 78 DISTANCE (KM) 0 20 40 60 80 100 120 140 160 MODEL DISTANCE (KM) Figure 5.5 (a) The raytracing diagram for a shot at shotpoint N, (b) the theo-retical traveltime curve superimposed on section N-North, and (c) the synthetic section corresponding to the raytracing diagram in (a). The boundary repre-senting the continental Moho has been removed. 79 DISTANCE (KM) 0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 SHOT-RECEIVER DISTANCE (KM) 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 MODEL DISTANCE (KM) Figure 5.6 (a) The raytracing diagram for a shot at shotpoint A , (b) the theo-retical traveltime curve superimposed on section A-South (filtered 6-13 Hz), and (c) the synthetic section corresponding to the raytracing diagram in (a). The numbers 1-6 in (a) identify layers discussed in the text. The wavelet used in (c) was chosen to represent the average waveform of the first arrival in (b). 80 DISTANCE (KM) 0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 140 160 180 200 220 240 260 280 300 320 340 MODEL DISTANCE (KM) Figure 5.7 (a) The raytracing diagram for a shot at shotpoint F, (b) the theo-retical traveltime curve superimposed on section F-South, and (c) the synthetic section corresponding to the raytracing diagram in (a). The numbers 1-6 in (a) identify layers discussed in the text. The wavelet used in (c) was chosen to represent the average waveform of the first arrival in (b). 81 The traveltime corresponding to the refracted arrival through the uppermost layer, on section A-South, blends in with that of the arrival refracted through the second layer (figure 5.6b). The model thickness of the uppermost layer at shotpoint A is 1 km with a velocity of approximately 5.5 km/s. The first refracted arrival from a shot at shotpoint F travels through material of the same velocity, but with a thickness of roughly 2 km. There were no traces recorded on section F-South within shot-receiver distances that could record an arrival refracted through the uppermost layer. This was due to the fact that shot F was detonated in the Strait of Juan de Fuca, off the southwest coast of Vancouver Island, and no receivers could be placed closer than approximately 15 km to the shotpoint. However, in order to achieve the traveltime delay observed in the data, an extension of the uppermost layer to the southern portion of line IV was required. Because of this, synthetic arrivals generated for the arrival turning in the uppermost layer of the model cannot be compared with corresponding arrivals on section F-South. However, section A-South does contain a set of high amplitude arrivals at shot-receiver distances less than 20 km, which agree well with the synthetic arrivals generated by turning rays through the uppermost layer of the model. Refractions through the second layer of the velocity model are used to model the second refraction branch observed in the data. The apparent velocity of the arrivals along this second branch is approximately 6.5 km/s for both A-South and F-South (figures 5.6b and 5.7b). The model traveltimes calculated for this refraction branch agree well with those observed in the data (figures 5.6b and 5.7b). The synthetic first arrivals corresponding to F-South (figure 5.7c) are consistent with those observed in the data. The amplitudes in both the data and the synthetics show an almost constant amplitude arrival out to roughly 140 km shot-receiver distance, or 190 km model distance. On section A-South the observed amplitude varies from extremely low amplitude arrivals, from 50 to 95 km shot-receiver distance, to arrivals which are of much higher amplitude on the greater shot-receiver offset traces (95-135 km shot-receiver distance, figure 5.6b). The synthetic section (figure 5.6c) reproduces these features relatively well, the differences being: (l) the synthetic amplitudes at the start of the second branch of refractions (180 km model distance) are smaller than those observed in the data, (2) the synthetic amplitudes beyond 290 km (model distance) on the second refraction branch are too large compared to what is seen in the data. The final branch of refracted first arrivals observed on section A-South and F-South is modeled as refractions through the fourth layer of the velocity model. The third layer of themodel is intended to correspond to reflecting zone D in the 1984 LITHOPROBE reflection data (figure 3.7). As in the northern portion of line IV, the third layer is modeled as a low velocity zone. The fourth layer again corresponds to the area between reflecting zones D and E in figure 3.7. Because of the added constraints provided by the LITHOPROBE data, as well as the consistency constraint imposed by the onshore-offshore refraction model for line 1, the model structure in the southern portion of line IV is better constrained than that beneath the northern portion of line IV. These constraints control much of the structure of the velocity model from approximately 260 km (model distance), where both line I and line 84-0] intersect line IV, to the southern end of the model (figure 5.]). The calculated traveltimes corresponding to this final refraction branch on F-South agree well with the observed traveltimes in the data (figure 5.7b). However, the calculated traveltimes corresponding to these arrivals on section A-South are as much as 0.3s early (figure 5.6b). The calculated traveltimes for the second refraction branch agree quite well with these arrivals, but the synthetic amplitude generated is too large compared to that observed in the data (figure 5.6c). Traces on section F-South beyond 140 km shot-receiver distance exhibit a first arrival with approximately half the amplitude as those traces closer to the shot. This drop in amplitude is also reproduced in the synthetics coressponding to a shot at shotpoint. F. 83 The synthetic arrivals corresponding to the third branch of refracted arrivals (315-340 km model distance) observed on section A-South agree with the amplitude of arrivals seen on that section, but their traveltimes are as much as 0.3s too early (figure 5.6c). Reflected arrivals from the top of the fourth layer as well as the four deepest layers in the model were generated for both the forward and reverse sections. Calculated traveltimes for a family of reflections from the top of the fourth layer match a set of secondary arrivals (between 65 and 135 km shot-receiver distance) seen on section A-South. The same set of arrivals was also generated for a shot at F, but the data on section F-South do not require these arrivals even though the additional large amplitude reverberations add to the synthetic-observed comparison. Deeper reflected arrivals were generated for both sections from reflecting boundaries required by the onshore-offshore model. On the furthest shot-receiver offset traces (150-190 km shot-receiver distance) there appears to be some weak supporting evidence that some of these arrivals may correspond to features in the data, but for the most part they are not required by the data. Synthetic arrivals were generated which correspond to reflected arrivals from the top of the fourth layer as well as the layers below 35 km. The synthetic arrival reflected from the fourth layer agrees well with a similar arrival on A-South, but no obvious agreement can be seen between this arrival and features visible on section F-South. Although the amplitude of the deep synthetic reflections, in the upper 5s of the synthetic section, agrees relatively well with the amplitude of seismic energy in this area on both section A-South and section F-South. the data show no coherent energy arriving that corresponds to these arrivals. The existence and/or location of the deep reflected arrivals is not entirely clear, therefore no attempt was made to try and match amplitudes between the synthetics and features in the data. 84 The waveforms exhibited on sections A-South and F-South are very different with the frequency content of the arrivals on section F-South being much lower than those on section A-South. This is probably due to the fact that F-South was detonated in a marine environment and much of the high frequency energy may have been absorbed in poorly consolidated marine sediments. 5.4 Interpretation Of The Offline Model Both the forward and reverse profiles recorded along the length of line IV were recorded in two segments. The first segments of both the forward and reverse sections are sections N-North and F-South, respectively. The significant features on each of these data sections are examined in section 5.3. These data sections were combined with the data recorded on N-South and F-North along segments A-F and N-A of line IV, respectively, from shots at N and F, respectively (figure 5.8). The data on sections N-South and F-North are generally of poor quality. By combining the data sets a forward and reverse section along the full length of line IV can be examined. The forward and reverse sections will be referred to as N(N-S) and F(S-N), respectively. Figure 5.9 shows the raytracing diagram corresponding to N(N-S), the observed data section with the calculated traveltime curves superimposed, and the synthetic section generated from the raytracing run. The raytracing diagram for F(S-N), along with the corresponding traveltime curves and synthetics, are shown in figure 5.10. With the distance scaling factor used (755), arrivals at the large offset distances are almost impossible to see; however figure 5.8 shows the data with each trace normalized to a common maximum amplitude which helps make the weaker arrivals easier to see. The velocities and gradients associated with the various layers, in the velocity model, are shown in figure 5.2. This velocity model no longer represents a strictly two-dimensional 85 150 170 190 210 230 250 270 290 310 330 SHOT-RECEIVER DISTANCE (KM) SHOT-RECEIVER DISTANCE (KM) Figure 5.8 (a) The observed data section corresponding to a shot at shot-point N recorded along the southern segment of line IV. (b) the observed data recorded along the northern segment of line IV, from a shot at shotpoint F. The thicker lines on both sections represent calculated traveltimes for the final velocity model superimposed on the data. All traces have been normalized to a common maximum amplitude. 86 Figure 5.9 (a) The raytracing diagram for a shot at shotpoint N, (b) the theo-retical traveltime curve superimposed on section N(N-S) (filtered 3-10 Hz), and (c) the synthetic section corresponding to the raytracing diagram in (a). The wavelet used in (c) is the same as that used in figure 5.3c. 87 DISTANCE (KM) SHOT-RECEIVER DISTANCE (KM) 0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 MODEL DISTANCE (KM) Figure 5.10 (a) The raytracing diagram for a shot at shotpoint F, (b) the theoretical traveltime curve superimposed on section F(S-N) (filtered 1-10 Hz), and (c) the synthetic section corresponding to the raytracing diagram in (a). The wavelet used in (c) is the same as that used in figure 5.7c. 88 model, but now also reflects features to the west of line IV. The reasons for this are discussed in section 3.1. The two features in both N(N-S) and F(S-N) that we are most interested in modeling are the first arrivals, which can be followed more than halfway across both the forward and reverse sections, and the secondary seismic energy showing up on both profiles at times later than 5.5s. The first arrivals for the first half of both N(N-S) and F(S-N) were previously modeled in section 5.3.1 and section 5.3.2, respectively. Although the structure of the offline model differs from that of the hinged model at depths greater than 10 km, this difference does not significantly affect the arrivals previously modeled for sections N-North and F-South. Therefore, the discussion concerning modeling of the arrivals for the first half of section N(N-S) and section F(S-N) will not be repeated. The first arrivals observed on the traces at shot-receiver distances greater than 150 km become increasingly more difficult to follow with increasing shot-receiver distance, even when the amplitudes are enhanced (figure 5.8). The calculated traveltime curves agree quite well with the first arrival traveltimes on both sections out to roughly 270 km on both the forward and reverse sections (figure 5.8); beyond this, however, the calculated traveltime curves continue, but there are no visible first arrivals to compare with the calculated traveltime curve. The synthetic section corresponding to F(S-N) is shown in figure 5.10c. The first, arrivals across the first half of the observed data (0 to 185 km shot-receiver distance), were previously modeled in section 5.3. The amplitude of the synthetic first arrivals beyond 170 km (model distance) is similar to that seen in the observed data beyond 150 km shot-receiver distance; in both cases the amplitude of the first arrival is dying off with increasing distance from the shot. However, on the synthetic section the amplitude of these first arrivals increases slightly between 90 and 0 km model distance. The amplitude of the first arrivals in the synthetic section corresponding to N(N-S) (figure 5.9c) exhibit 89 the same sort of behavior. The synthetic amplitudes are very similar to the amplitudes observed in the data out to approximately 240 km (model distance) at which point there is a slight increase in the synthetic amplitude and a continued decrease in observed first arrival amplitude. Secondary seismic energy arriving later than 5.5s, on both the forward and reverse profiles is modeled as energy being reflected from the three layers below 30 km in the velocity model (figures 5.9b and 5.10b). Of these three layers, the upper two are mod-eled as the top two layers within the subducting slab and the lowest reflecting boundary is modeled as the top of the oceanic Moho. These deep reflectors correspond to similar features in the hinged model, although in this model they occur at a slightly shallower depth, and no boundary corresponding to the continental Moho is present. The dif-ference in model structure reflects the fact that layers associated with the subduction process are shallower to the west of line IV then they are under line fV. Since these reflected rays are traveling through material to the west of line IV, the reflecting bound-aries that appear in the hinged model and are associated with the subducting plate will appear at shallower depths in this model. The calculated traveltimes for rays reflecting off the top of the three deepest layers in the model correspond reasonably well to the arrival of seismic energy on section F(S-N). This is most clearly seen on traces at shot-receiver distances greater than 210 km (figure 5.10b). The calculated traveltimes corresponding to rays reflecting off the top of the three deepest layers in the model correspond equally well with obvious seismic energy arriving on section N(N-S) (figure 5.9b). The calculated arrival times for these reflected rays originating from shots at both shotpoints N and F can be compared more closely with the data in figures 5.8a and 5.8b, respectively. The synthetic arrivals corresponding to rays reflecting off layers in the subducting plate are shown in figures 5.9c and 5.10c for both the forward and reverse profiles 90 respectively. The amplitude of these arrivals corresponds well with the general trend in the amplitude of secondary seismic energy on the observed data sections for N(N-S) and F(S-N). On both the real and synthetic sections there is a gradual decrease in the amplitude of secondary seismic energy with increasing shot-receiver distances. However, in the case of N(N-S) (figure 5.9c) the amplitude of the latest synthetic arrival is too large compared to what is seen in the data. This may be related to the fact that these reflected arrivals are reflecting off the Explorer plate rather than the Juan de Fuca plate, and the subsurface features associated with subduction under the northern portion of Vancouver Island are poorly constrained at depths greater than 15 km. Due to the lack of additional geophysical data on the northern portion of the island, the true structure in that region (below 15 km) could be significantly different from that which has been proposed in this model. 5.5 F i n a l V e l o c i t y M o d e l s The final velocity models based on the complete raytracing interpretation were pre-sented in figures 5.1 and 5.2. Two different velocity models were developed for line TV (hinged model and offline model) based on the arguments presented in section 5.1. 5.5.1 H i n g e d V e l o c i t y M o d e l The hinged model consists of two reversed profiles (A-F and A-N) joined together at shotpoint A (figure 2.1). The uppermost layer in this model ranges in thickness from 1 km to approximately 2.5 km with velocities ranging from 5.5 to 6.2 km/s. The velocity and gradient in the second layer are uniform across the entire model with the actual velocity values ranging from approximately 6.43 km/s to 6.77 km/s. Below a depth of approximately 10 km, the model can be divided into two subsurface regions: the northern region and the southern region. The third, fourth and fifth layers in the hinged 91 model correspond to reflectors seen in the LITHOPROBE data recorded on southern Vancouver Island. The velocity of the third and fifth layer has been modeled as being the same across the entire model. The material between the two layers is modeled as higher velocity material with a. slightly lower velocity in the northern region as compared to the southern region. The southern portion of line IV is better constrained than the northern portion because of the added constraints imposed on the model by the LITHOPROBE reflection data. Line 84-01 crosses VISP line IV at the point indicated in figure 5.1, and the southern end of line 84-03 comes within 5 km of line IV. Lines 84-02 and 84-04 do not intersect line IV, but they do provide a general idea as to what the regional structure at the southern end of line IV might be. The third, fourth, and fifth layers are less well constrained in the northern region where there is no reflection data to confirm their existence or constrain their location. However, the good quality refraction data enable a normal interpretation of the velocity structure; in particular the data support refracted arrivals turning at depths near 15 km. The existence and/or location of the relatively thin (2-3 km) fifth layer (figures 5.6 and 5.7) which ranges in depth from 20 to 30 km across the entire model is much more poorly constrained in the northern region than in the southern region because of the lack of reflection data on the northern portion of the island, and due to the fact that it is not required to model features observed in the data collected along segment N-A. It is included for continuity along strike. The sixth layer in the model, along with the layers beneath it, are not required by the refraction data recorded along line IV, but have been included to maintain consistency with the model developed for VISP line I (figure 4.1). The point of intersection between the two models is shown in figure 4.1 as well as in figure 5.1 and the two models are consistent along their line of intersection. 92 5.5.2 Of f l i ne V e l o c i t y M o d e l The upper 10 km of the offline velocity model (figure 5.2) is exactly the same as that seen in the hinged velocity model. The overall structure along the third, fourth, and fifth layers is more horizontal than the structure seen in the hinged model (figure 5.1). The flattening of these layers was necessary in order to fit first arrivals along the length of line IV. But since the offline model runs approximately parallel to the strike of the subducting plate, it seems reasonable to expect that these layers would be more horizontal than they were in the hinged model, for which shotpoint A is further downdip along the subduction zone than either shotpoints N or F. The thicknesses of the third, fourth, and fifth layers are slightly different, than those shown in the hinged model. The change in thickness on the southern half of the model is supported by the LITHOPROBE reflection data (figure 3.7) which show a thickening of the fifth layer and a gradual thinning of the high velocity wedge toward the west coast of Vancouver Island. The same characteristics have been extended to the northern portion of the island, although there are no reflection data to support the changes in thickness on the northern portion of the model. The velocities .and gradients corresponding to the fourth layer are only slightly different, than those used in the hinged model, but were required to fit first arrivals observed along the full length of line IV. The velocities and gradients corresponding to the sixth layer, as well as the layer beneath it, are the same as those used in the hinged model. However, the locations of these layers are slightly different. The structure along these layers is much flatter than in the hinged model, since this line runs more along strike of the subducting plate. The two layers between 35 and -15 km depth, which represent the subducting plate itself, are slightly shallower than they were in the hinged model. The top of the subducting plate is modeled as being deeper in the northern region than in the southern region. 93 This was required to fit reflected arrivals observed on section N-South. As with the hinged model, the northern region is much more poorly constrained than the southern region. Therefore the structure below 15 km on the northern half of the model must be considered somewhat speculative. 94 CHAPTER 6 DISCUSSION AND CONCLUSIONS 6.1 The Onshore-Offshore Profile The first objective of this study has been to resolve differences between the onshore-offshore velocity model proposed by Spence (1984) and the deep reflection data recorded along approximately the same line during the 1984 L I T H O P R O B E and 1985 F G P sur-veys (figure 2.1). The result is a velocity structure model which is consistent with both the refraction data and the deep reflection data recorded along line I. Figure 6.1a is a generalized version of the more detailed onshore-offshore velocity model in figure 4.1. Figure 6.1b shows a generalized version of the velocity model pro-posed by Spence (1984) for comparison during the following discussion. The velocity model in figure 6.1a is actually a composite model requiring the interpretation of sev-eral associated seismic data sets. As previously noted, both the geologic and velocity structures of the oceanic crust on the western end of the model were determined by Waldron (1982) in an interpretation using marine refraction data from OBS's 1, 3, and 5 (figure 2.1). D.J. White (personal communication, 1986) reinterpreted these data and made some minor modifications to Waldron's original model. Spence (1984) also used the results from Waldron's study, since the results of White's reinterpretation did not become available until later. The upper 18 km at the eastern end of the onshore-offshore velocity model (beyond 240 km) was modeled by Spence (1984) and since no new constraints were available for this portion of the model, the velocity structure re-mains unchanged from that which he used. The basic velocity structure in the upper 10 km of the continental crust was established by McMechan and Spence (1983), based on their interpretation of the refraction data recorded along line IV (figure 2.1). However, 95 Wesi - DISTANCE.(KM) Vancouver Island Easi N-S Line Mainland Figure 6.1 (a) Generalized velocity model interpreted from the onshore-offshore refraction and reflection data collected along line 1. (b) generalized velocity model proposed by Spence (1984) based primarily on the onshore-offshore re-fraction data collected along line I. Numbers assigned to each polygonal block indicate the range of velocities (km/s) at the shallowest and deepest points within that block. The small arrow represents the line of intersection between line 1 and line IV (230 km). The larger arrowheads represent the boundaries of Vancouver Island and the mainland. (M) indicates the location of the continen-tal Moho. • - <K; -in light of the deep reflection results from southern Vancouver. Island and a more de-tailed reinterpretation of the refraction data collected along line IV, much of the velocity structure below 30 km, particularly on the southern half of the island, was found to be inconsistent with features in both the reflection and refraction data, and has been rein-terpreted as part of this thesis. The original model developed for line I (Spence, 1984; Spence et a/., 1985) incorporated results from the model originally developed for line IV (McMechan and Spence, .1.983), but the two models were inconsistent at their point of intersection for depths greater than approximately 1.0 km. A complete discussion of the models developed for line IV is presented in section G.2. The primary differences between the onshore-offshore velocity model proposed by Spence (1984) (figure 6.1b) and that determined in this study (figure 6.1a) are: (1) the velocity structure beneath the Vancouver Island region (140 to 240 km model distance), and (2) the velocity structure beneath the mainland region (240 to 360 km model distance). Of these two primary differences, the revised structure proposed for the area below J.0 km in the Vancouver Island region is the most well constrained. The original model for line .1 (Spence, 1984; Spence el al., 1985) required a block of material with a velocity of 7.7 krn/'s beneath Vancouver Tsland, which was underlain by lower velocity materials (figure 6.1b). The same region in the revised velocity model (figure 6.1a) shows a four-layer structure with two low velocity layers (6.35 km/s) alternating with two high velocity layers (7.15 km/s and 7.1 — 7.18 km/s). The velocity in the uppermost high velocity layer was determined from that required to model refracted arrivals recorded along line IV. The other three layers also appear in the model developed for line IV, but since there were no turning rays modeled within these layers, the velocities were poorly constrained. Thus the velocities within the other three layers were determined from traveltime and amplitude constraints in the refraction data recorded along both lines 1 and IV. As mentioned in section 4.1, the change in structure beneath the Vancouver Island region was required to maintain consistency between the model in figure 6.1a and the onshore-offshore deep reflection profile recorded along a line which is approximately co-incident with line 1 (figure 2.1). The vertical two-way traveltimes to prominent reflecting horizons were used to constrain the velocity structure of the revised onshore-offshore model. Such changes in the velocity structure beneath the Vancouver Island required minor changes in the velocity structure of the upper mantle from that proposed by Spence (1984) in order to fit the refraction traveltime data recorded on Vancouver Is-land. However, the velocity structure in the upper mantle is poorly constrained in comparison with the structures at shallower depths beneath Vancouver Island. The changes made in the velocity structure of the upper mantle made it necessary to alter the velocity structure below 18 km in the mainland region in order to fit traveltime data recorded on the mainland from the P-series shots. Also the positive velocity gradient for the low velocity zone below 18 km in the revised model proved helpful in turning rays up toward the surface. Nevertheless, the velocity structure below 18 km at the mainland stations is probably the most poorly constrained part of the model since: (l) there have not been any deep reflection data recorded over the mainland region that might help constrain the structure, and (2) the existing refraction data do not include rays turning in this area of the model, so no well constrained velocity estimate is available. The location of the subducting plate in the revised model is consistent with the vertical two-way traveltime to the top of a reflector, considered to be the top of the subducting Juan de Fuca plate, observed in the onshore-offshore deep reflection data recorded approximately coincident with line I. The extent of this constraint is shown in figure 4.1. The subducting plate was found to have an average dip of 14° to 16° which is approximately the same as that proposed by Spence (1984) and Spence et al. West East DISTANCE (KM) Vancouver Island Mainland 360 Figure 6.2 Hypocenter cross section across the convergent margin superimposed on the onshore-offshore model determined in this study. "Mag" refers to the earthquake magnitude based on the Richter scale. The arrowheads represent the boundaries of Vancouver Island and the mainland (after G.C. Rogers, personal communication, 1986). (1985). In addition to the constraint provided by the onshore-offshore deep reflection data, there is also seismicity information which places the hypocenters of a number of small earthquakes approximately coincident with the location of the subducting plate as interpreted from the onshore-offshore deep reflection data (G.C. Rogers, personal communication, 1986). In figure 6.2, hypocenters near line I are superimposed on the seismic structural model determined in this study. There is good correspondence be-tween the location and dip of the hypocenters and the position of the subducting plate in the seismic model. Recent surface-wave studies in the Pacific and Atlantic oceans indicate that the average lithosplieric thickness can be no more than 25 km for a region from the ridge axis to an isochron of 5 Ma (Forsyth, 1977). The oceanic lithosphere is between 6 and 7 Ma old at the point where it begins to subduct beneath Vancouver Island (Monger et a/., 1985). Thus, the upper mantle reflector in the onshore-offshore velocity model may be associated with the base of the subducting oceanic lithosphere or equivalently 99 the top of the oceanic asthenosphere (figure 6.1a). As mentioned in section 4.5, the velocity below this boundary is poorly constrained; however, through the modeling of reflected waves and physical considerations, the velocity contrast at this boundary is better constrained and was determined to be approximately 0.5 km/s. Because of this, either a velocity slightly lower or slightly higher than the velocity in the material above this boundary could be used to fit the observed data reasonably well, although a negative velocity contrast was found to provide a better overall fit to the data. A negative velocity contrast across this boundary is also consistent, with what, one would expect between the oceanic lithosphere and asthenosphere. However, a positive velocity contrast did provide a closer fit to data recorded on sections P.13 and P19 (see section 4.3). Figure 6.3 shows the revised onshore-offshore model with a velocity of 8.6 km/s assigned to the region below the upper mantle reflecting boundary. The differences in model structure beneath the mainland region are discussed in forthcoming paragraphs. The amplitudes of reflected rays arriving between 160 and 280 km (model distance) agree with the observed data better than rays reflecting from the same boundary when a velocity of 7.7 km/s is used, rather than a velocity of 8.6 km/s (figure 4.4). The velocity structure below 18 km in the mainland region has been modeled as a thick low velocity zone (6.40 -- 6.95 km/s) overlying the continental Moho (M) (figure 6.1a). There are probably many velocity structures that could be used in this part of the model to fit the observed data. The primary constraint on any velocity structure used in modeling this block is the total traveltime through it, which must be approximately 2.9s. One alternative is shown in figure 6.3. ln this alternative model, the size of the low velocity zone has been reduced, and the range of velocities within this zone is now from 6.65 to 6.93 km/s. The high velocity layer (7.10 — 7.18 km/s) overlying the subducting plate in the Vancouver Island region has been carried across to the eastern end of the model. The traveltimes and amplitudes of rays traveling through this region fit the 100-DISTANCE (KM) 0 40 80 120 160 200 240 280 320 360 n -c\) 4—1—I—1—I—1—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—|—I—I—I—| 1—I—,—I—,—I— 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 MODEL DISTANCE (KM) Figure 6.3 An alternate model which can be used to fit the data collected along line I. The changes relative to figure 4.1 are shown by the stippling, (a) The raytracing diagram corresponding to a shot at shotpoint P13, (b) the corre-sponding traveltime curve superimposed on section P13, and (c) the synthetic section generated from the raytracing run in (a). Velocities (km/s) followed by the velocity gradient (km/s/km) are given for the velocity blocks in the region affected by the alteration. 101 observed data quite well. However, the complete velocity structure of low-high velocity sequences interpreted below 10 km in the Vancouver Island region cannot be continued to the eastern edge of the model since the traveltime delay observed in the data on the P-series sections would not be produced. In addition, an attempt was made to model the data recorded on the mainland by continuing the complete velocity structure, below 10 km in the Vancouver Island region, to the eastern end of the model at a dip slightly less than that, used to model the subducting oceanic crust. This meant that boundaries corresponding to the continental Moho and upper mantle had to be removed. Tracing rays from shotpoint P13 through this model to the mainland produced arrivals which had an apparent velocity significantly lower (ft: 7.0 km/s) than those observed in the data (w 8.0 km/s). • Because of the variability in possible velocity structures that could be used to fit the data, and the lack of constraints on the deep structure in the mainland region, the geologic and tectonic interpretation of this region is poorly constrained. A two-dimensional interpretation of magnetotelluric data collected along lines 84-01 and 84-03 suggest the presence of a highly conductive zone at a depth of approximately 30 km beneath the mainland (Kurtz et al., 1986). This may correspond to the low velocity zone (6.40 - 6.95 km/s) beneath the mainland region required by the seismic data. Heat flow data acquired from the shelf edge to the interior plateau of British Columbia (T. Lewis, unpublished) suggest a rapid increase in heat flow beneath the mainland relative to that, beneath Vancouver Island. The source of these anomalies is uncertain but they would be consistent with an upwelling of partially molten material associated with the volcanic belt. Thus, both the magnetotelluric and heat flow data lend support to the low velocity zone beneath the mainland region as interpreted from the seismic data. However, both the data sets also suggest that the change in structure between the Vancouver Island region and the mainland region takes place farther east than the 102 - -rapid change in velocity structure required by the seismic data. These interpretations of the three data sets all indicate major lateral changes at depth between the Insular Belt and Coast Plutonic Complex. However, the exact location where these changes take place is unclear since the seismic data indicate a change in structure to the west of that indicated by the magnetotelluric and heat flow data. The origin of the anomalous structure below 10 km and above the subducting plate in the Vancouver Island region is. largely unknown, although several possible explana-tions have been suggested (Clowes et a/., 1987a). The letters in the following discussion (C, D 2 , E j , and E 2 ) correspond to the upper and lower boundaries of highly reflective zones observed in the deep reflection data (figures 3.7 and 3.8). These boundaries have been incorporated into the onshore-offshore velocity model (figure 6.1a) and appear below 10 km and above the subducting plate in the Vancouver Island region. One hypothesis is that this sequence of layers may have developed by the addition of newly imbricated materials which were scraped off the top of the subducting plate. Figure 6.4 shows the sequence of events that may have formed the series of layers above the subducting plate. Stage 1: the C — D 2 interval is shown as being formed by the off-scraping of sediment cover on the downgoing plate. Stage 2: two alternatives (options A and B) are given for the possible formation of the D 2 — E] interval. The first alternative (option A) involves the accretion of a single slice of crust and mantle from the subducting plate. One possible problem with this hypothesis is that the episodic accretion of a thick chunk of oceanic lithosphere might be expected to cause a rapid uplift of the overlying subduction complex, and there is no evidence suggesting that this has occurred. The second alternative (option B) (figure 6.4b), is that the D2 — E] interval consists of imbricated slices of mafic rocks derived from the top of the downgoing plate. In this scenario, smaller slices would have been continuously scraped off the top of the 103 ® STAGE 1: OFF-SCRAPING & UNDERPLATING FROM SEDIMENT COVER SEDIMENT COVER (§) STAGE 2: ACCRETION OF MAFIC ROCKS OPTION A: EPISODIC ACCRETION OF OCEANIC CRUST AND MANTLE STAGE 3: ONCE AGAIN, ACCRETION DOMINANTLY RESTRICTED TO SEDIMENT COVER ACCRETED MAFIC 4 ULTRAMAFIC (?) SLICE E1 Figure 6.4 A sequence of events which may have formed the series of layers above the subducting Juan de Fuca plate. Results from this study suggest that stage 2 may currently be repeating itself (from Clowes et a/., 1987a). subducting plate rather than one large slice being scraped off in a single event, as in the case of option A. The high velocity of this region might be accounted for by high pressure metamorphism of the imbricated mafic rocks (Clowes et al., 1987a). Stage 3: the preferred interpretation of the Ej — E 2 interval presented by Clowes et al. (1987a) (figure 6.4c) is that this interval represents the upper part of a zone of active accretion in which material is presently being scraped off the top of the subducting plate. Clowes et al. (1987a) suggest that the top of the subducting plate may be coincident with the E2 boundary, although they acknowledge that seismicity data suggest that the top of the plate is deeper than E2. The refraction data and the 1985 FGP marine reflection data also suggest that the top of the plate must be deeper than the E2 horizon. This 1.04 may mean that the accretionary process that was suggested by Clowes ct a/. (1987a) as an explanation for the creation of the D2 - Ki interval is currently repeating itself, thus creating another thief; high velocity layer (not included in figure 6.4). Another possible explanation for the whole sequence of layers below 10 km and above the subducting plate in the Vancouver Island region is that they are directly related to a westward jump of the subduction zone. The sequence of layers above the current subducting plate may represent pieces of the oceanic lithosphere which were emplaced during two earlier phases of subduction. The Di — Ej interval may represent an ancient oceanic crust and upper mantle sequence which resisted subduction due to any one or a combination of reasons: (]) material encountered beneath the mainland region may have caused subduction to stop, and/or (2) the material being subducted was too light relative to the overlying material to subduct even at a shallow angle of subduction. Whatever the reason, subduction of the oceanic crust corresponding to the D] - Ej interval stopped and the subduction zone jumped westward. The interval between E i and the top of the currently descending plate may represent, the next phase of attempted subduction. The oceanic lithosphere began subducting at a slightly greater angle directly below the Di — Ej interval, but again subduction was aborted, possibly again due to one or both of the reasons previously given for the end of subduction of the Dj — E] interval, in any case, the subduction zone again moved westward, this time to its present location. As modeled, the velocities assigned to the D2 — Ei interval and the interval between E2 and the top of the subducting plate are significantly lower than the velocity assigned to the currently subducting oceanic upper mantle. However, these layers could easily consist of a mixture of relatively low velocity oceanic crustal material and higher velocity oceanic mantle material which would combine to give an average velocity that is lower than that of the currently subducting oceanic mantle. In addition, there could easily be a sliver of material (not included in figure 6.1a) at the base of each of these layers with a velocity similar to that modeled for the currently subducting oceanic mantle since no rays were required to pass through material in the lower right-hand corner of either of these layers (figures 4.3 — 4.9). No matter what the origin of the layers currently overlying the subducting plate, the terminus of all of these layers below the eastern part of Vancouver Island is probably more complex than that illustrated in figure 6.1a. 6.2 N-S Profile Along Vancouver Island The second objective of this study has been to remodel the refraction data collected along line IV and to use the 1984 L I T H O P R O B E reflection data to lielp constrain the revised velocity model wherever possible. In order to satisfactorily model the refraction data collected along line IV, while still maintaining consistency with the deep reflection data, it was found that two velocity models were required. The reasons for using two velocity models rather than one are related to a considerable azimuthal coverage and are discussed in Chapter 3. Generalized versions of these two velocity models are presented in figure 6.5. Figure 6.6 shows the original model as determined by McMechan and Spence (1983) using the McMechan and Mooney (1980) raytracing and synthetic seismogram routine. Figure 6.5a is the hinged velocity model, which was developed solely on data recorded along the two shorter reversed segments of line IV (N-A and A-F). Figure 6.5b shows the offline model which was required to model the data recorded from shots at shotpoints N and F, recorded on receivers along segments A-F and N-A respectively (figure 2.1). 6.2.1 The Hinged Model Of the two models developed for line IV, the hinged velocity model is the most well constrained. The details involved in developing this model are discussed in section 5.2. 106 North DISTANCE (KM) South N A E-W Line F 0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 ',, ,,'V. ,' 1— . . . I... Q_ UJ Q (P) Figure 6.5 Generalized velocity models interpreted from the refraction data collected along line IV and the reflection data obtained in the L I T H O P R O B E experiment, (a) The hinged velocity model, based solely on data recorded along the two shorter reversed segments of line IV (N-A and A-F), and (b) the offline velocity model, based on data collected along the full length of line IV. Numbers assigned to each polygonal block indicate the range of velocities (km/s) within that block. (P) indicates the location of the top of the subducting plate. 107 NORTH N SOUTH A A. x F 5 . 5 " 6.5-10-* 20 gj 30 Q 40-5 . 5 -6.5-6.75,7.05-7.07 -6.6 •7.5 •••.-•.'•Y.— '••.';;-7.8.* : ; ; ; - 7 . 6 ; ; . . . . . -6.2, 7.46 6.75,6.95 6.96 •••6.6 50 100 200 DISTANCE (km) MODEL RELIABILITY: GOOD (REVERSED DATA) MARGINAL (UNREVERSED RAYS BOTTOM, OR PARTIALLY REVERSED) 300 POOR (INFERRED, OR NO RAYS BOTTOM) Figure 6.6 Velocity model proposed by McMechan and Spence (1983) based on their interpretation of the refraction data collected along line IV. Shotpoint locations are indicated by the letters N, A , and F. The letter X indicates the point at which line I intersects line IV. Velocities (km/s) are shown at bound-aries; where two values are provided, these represent the velocities directly above and directly below the boundary. Reliability of the subsurface structure, as de-termined by McMechan and Spence (1983), is indicated by the line type (after McMechan and Spence, 1983). The upper 10 km of both the hinged and offline models is very similar to the upper 10 km of the model proposed by McMechan and Spence (1983). The range of velocities in the upper 10 km of both the revised model and the original model is virtually identi-cal, both models having velocities ranging from 5.5 km/s to approximately 6.6 km/s. However, at depths greater than 10 km, the two models become significantly different (figures 6.5 and 6.6). As noted in section 4.1, the two models were developed using two different raytracing and synthetic seismogram routines; thus, their general appearance varies because of the different manner in which model boundaries are specified. In the 108 McMechan and Mooney (1980) raytracing and synthetic seismogram routine, a series of V(z) profiles are given as input and the points in each V(z) profile corresponding to common boundaries on adjacent V(z) profiles are joined together using a spline inter-polation. The input for the Spence et al. (1984) raytracing and synthetic seismogram routine is a series of polygonal blocks with a velocity and gradient assigned to each one (see discussion in section 3.3.1). The velocity structure below 10 km on the southern end of the hinged model has been constrained by the 1984 L I T H O P R O B E reflection data, while the original model developed by McMechan and Spence (1.983) was based only on the refraction data recorded along line IV. The southern portion of the hinged model (260 to 340 km, figure 6.5a) is relatively well constrained to a depth of approximately 30 km, by the combination of L I T H O P R O B E data and refraction data. Below 30 kin, the structure is designed to be consistent with the structure used in modeling the onshore-offshore data and the structure required by the offline model. As discussed in Chapter 5. there are no reliable secondary arrivals on the shorter reversed sections which could be used to constrain the structure below 30 km. The northern portion of the hinged model (0 — 160 km) is much more poorly constrained than the southern portion. No new constraints were available for this part of the model and the interpretation of this region was primarily based on the refraction data recorded during the 1980 VISP experiment. However, an effort was made to maintain a degree of consistency between the northern and southern portion of the model by extending the boundaries identified in the L I T H O P R O B E data on the southern portion of the island to the northern portion of the island. The velocity structure below 20 km in the northern region is poorly constrained since no turning rays were modeled below 20 km, and because there are no deep reflection data to support or refute the 109 structure proposed. The velocity structure below 30 km is based primarily on the mode] required to simulate the offline data (sections N-South and F-North). With regard to the map which shows the directional offset at A between the northern and southern segments of line IV (figure 3.2), it should be reiterated that the prominent syncline-like structure below A on the model of figure 6.5a is probably the result, at least in part, of a two-dimensional representation of a three-dimensional structure. Because A is the most easterly point in the downdip direction along line IV, structures associated with the subduction process should generally be deeper there than at the ends of the line. Nevertheless, the reflection data on the southern part of Vancouver Island do indicate lateral variations of structure along the strike of the subduction zone (Green et a/., 1986a). The offline structural model of figure 6.5b confirms this interpretation. 6.2.2 T h e Of f l i ne M o d e l The offline velocity model (figure 6.5b) represents the average velocity structure beneath the shaded region in figure 3.2. The details involved in developing this model are discussed in section 5.4. As discussed in Chapter 5, this model was necessary to satisfactorily model data on sections N-South and F-North since the receivers on which the energy was recorded were not inline with the shotpoints. The velocity structure in the upper 10 km of the offline model (figure 6.5b) is identical to that used in the hinged model (figure 6.5a). The velocities below 10 km are equivalent to those used in the hinged model; the primary difference between the two models is the subsurface geometry of the structure that is required to fit, the observed data. The structure below 10 km in the offline model is, in general, more horizontal than that in the hinged model. This was required to fit features in the data, but also agrees with what might be expected since the offline profile runs more along strike of the subducting plate than does the hinged model, as discussed in the previous section. no The structure below 30 km is based not only on the structure needed to model the onshore-offshore data, but also on visible secondary arrivals observed in the refraction data shown in the offline sections (N-South and F-North). Jt was found that the sub-ducting plate under the northern portion of Vancouver Island had to be deeper than it was under the southern portion of the island in order to fit some late arrivals observed on section N-South. There may be other ways to fit these late arrivals, but modeling the subducting plate as being slightly deeper in the northern region than in the southern region agrees with models based on gravity data (Riddihough, 1979). Riddihough (1979) interpreted the free air and Bouguer gravity anomaly along two lines, one which ran across the northern portion of Vancouver Island and one which ran across the southern portion of the island. Both profiles extended from the deep ocean across the convergent, margin to the mainland of British Columbia. The two profiles suggest significantly different subsurface structures. The model proposed for the northern line shows the subducting plate slightly deeper than it is in the southern model, which is what the revised models for line IV suggest. It should also be kept in mind that the plate subducting beneath much of the northern portion of line IV is the Explorer plate rather than the Juan de Fuca plate (the plate subducting beneath the southern portion of line TV). These two plates are believed to be subducting at different rates as demonstrated by the numerous earthquakes associated with their line of contact, the Nootka fault zone (Hyndman et a/., 1979). The Explorer plate is subducting at a rate of less than 2 cm/yr, while the Juan de Fuca plate is subducting at a rate of about 4 cm/yr (Riddihough, 1977; 1984). The Explorer plate is believed to be almost stationary in a hot spot frame of reference, and any subduction is thought to be due to the America plate overriding it (Riddihough, 1984). Thus the subduction regime beneath northern Vancouver Island could be significantly different than that under the southern portion of the island, and the available data seem to suggest differences between the two regimes. I l l The structures derived in the two models developed for line IV are the same as those interpreted in the area beneath Vancouver Island in the onshore-offshore model. The NW-SE profile provides a cross sectional view of the velocity model approximately along strike of the subducting plate, while the onshore-offshore model gives a cross sectional view along a line approximately perpendicular to strike. Explanations as to the possible origin of the structures interpreted for line IV are the same as those discussed in section 6.1. 6.3 C o n c l u s i o n s The structural model developed for the onshore-offshore refraction and reflection profile along line 1 agrees with both the refraction data recorded along line 1 during the 1980 VISP refraction experiment and the vertical two-way traveltime constraints pro-vided by the 1984 L I T r l O P R O B E and 1985 F G P deep reflection data recorded along line 1. Significant changes were made to the original onshore-offshore model proposed for line I (Spence, 1984; Spence et a/., 1985) which was based primarily on the refraction data collected during the VISP experiment. Changes made in the original onshore-offshore velocity model which were directly related to constraints imposed by the deep reflection data include: (l) a significantly different velocity structure for the area below 10 km and above the subducting plate in the Vancouver Island region, and (2) a slight relocation of the subducting plate beneath the continental slope and Vancouver Island. These two aspects of the revised onshore-offshore model are also the most well constrained part of the model. As a result of these changes, the average velocity of the upper oceanic mantle had to be increased slightly (compared to that proposed by Spence (1984)), and some fairly minor alterations were required for the velocity structure below 18 km in the mainland region. 112 The structural models developed for the N-S line along Vancouver Island (line IV, figure 2.1) agree with the refraction data collected along line IV as part of the 1980 VISP experiment, the vertical two-way traveltime constraints imposed by the deep reflection data collected in the 1984 LITHOPROBE survey, and the revised onshore-offshore model developed for line 1. The velocity structure developed for line IV is significantly different at depths greater than 10 km than that originally modeled by McMechan and Spence (1983). Two velocity models were required to model the data collected along the N-S line since shots detonated at the two extreme ends of line IV (shotpoints N and F), and recorded on opposite segments (segments A-F and N-A, respectively) of the line, sampled the subsurface geology in three-dimensions. The structure along the southern half of line IV is relatively well constrained since it is based on the refraction data and structures suggested by the deep reflection data recorded on southern Vancouver Island. Structures modeled for the southern portion of the island have been extended to the northern half of the model although no deep reflection constraints are available there. The velocity structure in the northern region is relatively well constrained by the refraction data to a depth of approximately 20 km. However, below 20 km the velocity structure is based primarily on the better constrained structure beneath the southern portion of line IV as well as the structure required to fit the offline data (N-South and F-North). The plate subducting beneath much of northern Vancouver Island is the Explorer plate, while the Juan de Fuca plate is the plate subducting beneath the southern portion of the island. Gravity data suggest a different subsurface structure beneath the north-ern region than that in the southern region (Riddihough, 1979). The most significant difference between the subsurface structure modeled in the southern half of Vancouver Island and that modeled in the northern region is the depth to the top of the subducting plate. The top of the subducting plate under much of the northern half of Vancouver Island.is modeled as being, on average, 2.3 km deeper than it is beneath the southern half of the island. The two uppermost layers in the N-S models (figure 6.5) as well as the corresponding layers in the onshore-offshore model (figure 6.1a) arc equivalent to the Wrangellia terrane indicating that this terrane, which forms a large part of the Insular Belt, varies in thickness between 10 and 18 km. Such a relatively shallow extent suggests that either the terrane had a less than normal lithospheric thickness when it was accreted, or a substantial volume of its lithosphere has been removed as a result of the subduction process through subduction erosion and/or uplift and erosion. The four underlying layers (figure 6.5) are part of the Cenozoic subduction complex, while layers below 35 km are associated with the currently subducting oceanic lithosphere. The Crescent and the Pacific Rim terranes pinch out before they get as far east as line I V . Additional geophysical study along the onshore-offshore profile and the N-S profile is needed to resolve areas of uncertainty in the velocity models proposed for these areas. The primary area of uncertainty in the onshore-offshore model is the structure beneath the mainland region. An additional deep reflection profile extending from the eastern end of line 84-01 on the east coast of Vancouver Island to shotpoint J (figure 2.1) on the mainland might help constrain the subsurface structure beneath the mainland region. Logistically, this represents a difficult, but not insurmountable, problem. Similarly, additional deep reflection profiles recorded on northern Vancouver Island would help constrain the subsurface structure in that region. 114 R E F E R E N C E S Atwater, T., 1970, Implications of plate tectonics for the Cenozoic tectonic evolution of western North America: Geol. Soc. Am. Bull., 81, 3513-3536. Beck, M.E. and Noson, L., 1972, Anamalous paleolatitude in Cretaceous granitic rocks: Nature, 235, 11-13. Cassidy, J.F., 1986, The 1918 and 1957 Vancouver Island earthquakes: M.Sc. thesis, University of British Columbia, Vancouver, 143pp. Clowes, R.M., Ellis, R.M., Hajnal. Z. and Jones, I.F., 1983, Seismic reflections from subducting lithosphere?: Nature, 303, 668-670. Clowes, R.M., Green, A.G., Yorath, C.J., Kanasewich, E.R., West, G.F. and Garland, G.D., 1984, Lithoprobe - a national program for studying the third dimension of geology: J. Can. Soc. Expl. Geophys., 20, 23-39. Clowes, R.M., Spence, G.D., Ellis, R.M. and Waldron D.A., 1986, in R e f l e c t i o n Se is -m o l o g y : T h e C o n t i n e n t a l C r u s t : American Geophysical Union, Geodynamic Series, eds. Barazangi, M. and Brown, L., 14, 313-321. Clowes, R.M., Brandon, M.T., Green, A.G., Yorath, C.J., Sutherland Brown, A., Kanasewich, E.R. and Spencer, C, 1987a, LITHOPROBE - southern Vancouver Island: Cenozoic subduction complex imaged by deep seismic reflections: Can. J. Earth Sci., 24, 31-51. Clowes, R.M., Yorath, C.J. and Hyndman, R.D., ]987b, Reflection mapping across the convergent margin of western Canada: Geophys. J. Roy. Astr. Soc. in press. Coney, P.J., Jones, D.L. and Monger J.W.H., 1980, Cordilleran suspect terranes: Na-ture, 288, 329-333. Dickinson, W.R., 1970, Relations of andesities, granites, and derivative sandstones to arc-trench tectonics: Rev. of Geophys. and Space Phys., 8, 813-860. Dickinson, W.R., 1976, Sedimentary basins developed during evolution of a Mesozoic-Cenozoic arc-trench system in western North America: Can. J. Earth Sci., 13, 1268-1287. Ellis, R.M. and Clowes, R.M., 1981, Acquisition of crustal reflection/refraction data across Vancouver Island: Earth Physics Branch Open File Report 81-11, 72 pp. Ellis, R.M., Spence, G.D., Clowes, R.M., Waldron, D.A., Jones, I.F., Green, A.G., Forsyth, D.A., Mair, J.A., Berry, M.J., Mereu, R.F., Danasewich, E.R., dimming, G.L., Hajnal, Z., Hyndman, R.D., McMechan, G.A. and Loncarevic, B.D., 1983, The Vancouver Island seismic project: a CO-CRUST onshore-offshore study of a convergent margin: Can. J. Earth Sci., 20, 719-741. Forysyth, D.W., 1977, The evolution of the upper mantle beneath mid-ocean ridges: Tectonophysics, 38, 89-118. 115 Green, A.G., Clowes, R.M. and Yorath, C.J., 1985, L I T H O P R O B E seismic reflection profiles from southeastern Vancouver Island: Earth Physics Branch Open File No. 85-21, Geological Survey of Canada Open File No. 1180. Green, A.G., Clowes, R.M., Yorath, C.J., Spencer, C, Kanasewich, E.R., Brandon, M.T., and Sutherland Brown, A., 1986a, Seismic reflection imaging of the subduct-ing Juan de Fuca plate: Nature, 319, 210-213. Green, A.G., Berry, M.J., Spencer, CP., Kanasewich, E.R., Chiu, S., Clowes, R.M., Yorath, C L , Stewart, D.B., Unger, J.D. and Poole, W.H., 1986b, Recent seismic reflection studies in Canada in Reflection Seismology: A G l o b a l Perspective: American Geophysical Union, Geodynamic Series, eds. Barazangi M. and Brown L., 13, 85-97. Hyndman, R.D., Riddihough, R.P. and Herzer, R., 1979, The Nootka fault zone - a new plate boundary off western Canada: Geophys. J. Roy. Astr. Soc, 72, 59-82. Isacks, B. and Barazangi, M., 1977, Geometry of Benioff zones: lateral segmentation and downward bending of the subducting lithosphere in island arcs, in Deep Sea Trenches a n d B a c k - A r c Basins: American Geophysical Union, Maurice Ewing Series, eds. Talwani, M. and Pitman, W.C., 1, 243-258. Jones, D.L., Silberling, N.J. and Hillhouse, J., 1977, Wrangellia - a displaced terrane in northwestern North America: Can. J. Earth Sci., 14, 2565-2577. Kanasewich, E.R., 1981, T i m e Sequence A n a l y s i s in Geophysics, third edition: University of Alberta Press, Edmonton, 480 pp. Keen, C E . and Hyndman, R.D., 1979, Geophysical review of the continental margins of eastern and western Canada: Can. J. Earth Sci., 16, 712-747. Kurtz, R.D., DeLaurier, J.M. and Gupta, J.C, 1986, A magnetotelluric sounding across Vancouver Island detects the subducting Juan de Fuca plate: Nature, 321, 596-599. McMechan, G.A. and Mooney, W.D., 1980, Asymptotic ray theory and synthetic seis-mograms for laterally varying structures: theory and application to the Imperial Valley, California: Bull. Seism. Soc. Am., 70, 2021-2035. McMechan, G.A. and Spence, G.D., 1983, P-wave velocity structure of the Earth's crust beneath Vancouver Island: Can. J. Earth Sci., 20, 742-752. Monger, J.W.H., Clowes, R.M., Price, R.A., Riddihough, R.P., Simony, P. and Woodsworth, G.J., 1985, Continent-ocean transect B2: Juan de Fuca plate to Alberta plains: Geol. Soc. Am. Bull., Decade of North American Geology, Continent: Ocean Transect 7, 2 sheets, 21 pp. Muller, J.E., 1977, Evolution of the Pacific margin, Vancouver Island and adjacent regions: Can. J. Earth Sci., 9, 2062-2085. Riddihough, R.P., 1977, A model for recent plate interactions off Canada's west coast: Can. J. Earth Sci., 14, 384-396. 116 Riddihough, R.P., 1979, Gravity and structure of an active margin - British Columbia and Washington: Can. J. Earth Sci., 16, 350-363. Riddihough, R.P., 1984, Recent movements of the Juan de Fuca plate system: J. Geo-phys. Res., 89, 6980-6994. Riddihough, R.P. and Hyndman, R.D., 1976, Canada's active western margin - the case for subduction: Geoscience Canada, 3, 269-278. Roddick, J.A. and Hutchinson, W.W., 1974, Setting of the Coa.st Plutonic Complex, British Columbia: Pacific Geology, 8, 91-108. Shouldice, D.H., 1973, Western Canadian continental shelf, in Future petroleum provinces of Canada: Canadian Society of Petroleum Geologists, ed. McCrossan, R.G., Memoir 1, 7-35. Snavely, P.D. and Wagner, II.C, 1981, 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. Spence, G.D., 1984, Seismic structure across the active subduction zone of western Canada: Ph.D. thesis, University of British Columbia, Vancouver, 191 pp. Spence, G.D., Whittall, K.P. and Clowes, R.M., 1984, Practical synthetic seismograms for laterally varying media calculated by asymptotic ray theory: Bull. Seism. Soc. Am., 74, no. 4, 1209-1223. Spence, G.D., Clowes, R.M. and Ellis, R.M., 1985, Seismic structure across the active subduction zone of western Canada: J. Geophys. Res., 90, 6754-6772. Sutherland Brown, A. and Yorath, C.J., 1985, L I T H O P R O B E profile across southern Vancouver Island: Geology and tectonics (trip 8): Geological Society of America, Cordilleran Section Meeting, Vancouver, B.C., Guidebook, 23 pp. Symons, D.T.A., 1973, Concordant Cretaceous paleolatitudes from felsic plutons in the Canadian Cordillera: Nature, 241, 59-61. Taber, J.J. and Lewis, B.T.R., 1986, Crustal structure of the Washington continental margin from refraction data: Bull. Seism. Soc. Am., 76, 1011-1025. Waldron, D.A., 1.982, Structural characteristics of a. subducting oceanic plate: M.Sc. thesis, University of British Columbia, 1.2] pp. White, D.J., 1983, Shallow crustal structure beneath the Strait of Georgia, British Columbia: M.Sc. thesis, University of British Columbia, Vancouver, 121pp. White, D.J. and Clowes, R.M., 1984, Seismic investigation of the Coast Plutonic Com-plex - Insular Belt boundary beneath the Strait of Georgia: Can. J. Earth Sci., 21, 1033-1049. 117 Whittall , K.P. and Clowes, R . M . , 1979, A simple, efficient method for the calculation of traveltimes and ray paths in laterally iii homogeneous media: J. Can. Soc. Explor. Geophys., 15, 21-29. Yole, R .W. and Irving, E., 1980, Displacement of Vancouver Island, paleomagnetic evidence from the Karmutsen Formation: Can. J. Earth Sci., 17, 1210-1228. Yorath, C.J., 1980, The Apollo structure in Tofino Basin, Canadian Pacific continental shelf: Can. J . Earth Sci., 17, 758-775. Yorath, C.J., Clowes, R . M . , Green, A . G . , Sutherland Brown, A. , Brandon, M.T . , Massey, N .W.D. , Spencer, C , Kanasewich, E.R. and Hyndman, R.D. , 1985a, LITHO-P R O B E , Phase 1: southern Vancouver Island: Preliminary analyses of reflection seismic profiles and surface geological studies: in Current Research, Geological Sur-vey of Canada Paper 85-1A, 543-554. Yorath, C.J., Green, A . G . , Clowes, R . M . , Sutherland Brown, A . , Brandon, M . T . , Kanasewich, E.R. and Hyndman, R.D. , 1985b, L I T H O P R O B E , southern Vancou-ver Island: Seismic reflection sees through Wrangellia to the Juan de Fuca plate: Geology, 13, 759-762. 118 APPENDIX Additional E-W Record Sections A. l Common Shot Record Sections The following figures represent the observed data recorded from all the shots deto-nated along line 1 other than shots P19, Pi3, and P8, for which filtered data sections were presented in Chapter 4. The numbers assigned to the P-series shots increase as you move west along the portion of line 1 off the west coast of Vancouver Island (figure 2.1). Shots J l and J2 were located at the eastern end of line I. The sections presented in this Appendix are unfiltered. Amplitudes may be com-pared between shots and all amplitudes have been multiplied by the scaling factor -~, where d represents shot-receiver distance. Times and distances have been corrected in accordance with Spence (1984) to place each of the 17 P-series shots at a datum depth of 2.6 km, and for shots P8 — P19 to correct the sediment layer to a thickness of 1 km and a velocity of 1.8 km/s. S H O T - R E C E I V E R D I S T A N C E (KM) Figure A l . l Observed record section for shot J l . 130 90 50 S H O T - R E C E I V E R DISTANCE Figure A 1.2 Observed record section for shot J2. Figure A 1.3 Observed record section for shot P i , 104 km from shot Pl9. 90 130 170 210 250 SHOT-RECEIVER DISTANCE (KM) Figure A1.4 Observed record section for shot P2, 99 km from shot P19. Figure A1.5 Observed record section for shot P3, 94 km from shot P19. 105 145 185 225 SHOT-RECEIVER DISTANCE (KM) 265 to Figure A1.6 Observed record section for shot P4. 89 km from shot P19. Figure Al.7 Observed record section for shot P5, 83 km from shot P19. 115 155 195 235 275 SHOT-RECEIVER DISTANCE (KM) Figure A 1.8 Observed record section for shot P6, 78 km from shot P19. o n I — — — j • ' — ' '—{ ' 1 ' ' 1 ' ' 1 r f 130 170 210 350 290 SHOT-RECEIVER DISTANCE (KM) Figure A 1.9 Observed record section for shot P9, 63 km from shot P19. SHOT - RECEIVER DISTANCE (KM) Figure A L I O Observed record section for shot P10, 57 km from shot P19. 145 185 225 265 305 SHOT - RECEIVER DISTANCE (KM) Figure A l . l l Observed record section for shot P12, 47 km from shot P19. SHOT-RECEIVER DISTANCE (KM) Figure Al.12 Observed record section for shot Pl4, 36 k m from shot P19. 160 200 240 280 320 SHOT - RECEIVER DISTANCE (KM) Figure A1.13 Observed record section for shot P15, 31 km from shot P19. Figure Al.14 Observed record section for shot PI6, 26 km from shot P19. Figure A1.15 Observed record section for shot P17, 21 km from shot P19. Figure Al.16 Observed record section for shot P18, 10 km from shot P19. ]35 A.2 Selected Common Receiver Record Sections The figures presented in this section represent selected observed data sections recorded along line I, showing all 17 P-series shots recorded on a given receiver. The traces corre-sponding to shots P19 and P i are on the left and right side of each section respectively. Shots P7 and P l l were misfires; therefore the traces corresponding to these two shots are missing. Amplitudes may be compared between all receivers, and all amplitudes have been multiplied by the scaling factor —^, where d represents shot-receiver distance. Times and distances have been corrected in accordance with Spence (1984) to place the 17 P-series shots at 2.6 km depth, and for shots P8 — P19 to correct the sediment layer to a thickness of 1 km and a velocity of 1.8 km/s. co C5 SHOT - RECEIVER DISTANCE (KM) Figure A2.1 Observed record section for receiver X2. Figure A2.2 Observed record section for receiver X 1 3 . co oo S H O T - R E C E I V E R D I S T A N C E ( K M ) Figure A2.3 Observed record section for receiver X15. SHOT-RECEIVER DISTANCE (KM) Figure A2.4 Observed record section for receiver X17. SHOT - RECEIVER DISTANCE (KM) Figure A2.5 Observed record section for receiver X19. Figure A2.6 Observed record section for receiver X23. to SHOT-RECEIVER DISTANCE ( K M ) Figure A2.7 Observed record section for receiver X31. Figure A2.9 Observed record section for receiver X43. 

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}]}"
                            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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0085605/manifest

Comment

Related Items