A SEISMIC R E F R A C T I O N STUDY OF T H E H E C A T E S U B - B A S I N , BRITISH C O L U M B I A by CHRISTOPHER JAMES PIKE B.Sc.(Physics), The University of Toronto, 1982 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Department of Geophysics and Astronomy We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA November 1986 ©Christopher James Pike, 1986 In present ing t h i s thes is i n p a r t i a l fu l f i lment of the requirements for an advanced degree at the Un ivers i ty of B r i t i s h Columbia, I agree that the L ibrary s h a l l make i t f r e e l y ava i l ab le for reference and study. I fur ther agree that permission for extensive copying of t h i s thes is for scho la r l y purposes may be granted by the head of my department or by h i s or her representa t ives . It i s understood that copying or pub l i ca t ion of t h i s thes is for f i n a n c i a l gain s h a l l not be allowed without my wri t ten permission. Department of G*<_of)Kyt^5 4J\JL A~Srcd^oJi^ The Un ivers i ty of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date E-6 (3/81) ii To my grandmother, Nanny Pike, who passed away three short weeks before the completion of this thesis iii A B S T R A C T The Hecate sub-basin is one of two similar sedimentary structures comprising Queen Charlotte Basin, which is located between the British Columbia mainland and the Queen Charlotte Islands. The Queen Charlotte Basin was the locale of an active but unsuc-cessful exploration program, including drill holes, in the 1960's. However, recent studies incorporating modern concepts of plate tectonics have indicated a re-evaluation of the resource potential of the area is warranted. The Hecate sub-basin and its southern counterpart, the Charlotte sub-basin, are filled with Tertiary sediments that are un-derlain by a thick sequence of Tertiary volcanics. Penetration of the latter unit using the reflection method has been difficult. Thus the thickness of the volcanics and the existence or not of more sediments below them has not been established. To address this problem an airgun/ocean bottom seismograph (OBS) refraction survey was carried out across the Hecate sub-basin in 1983. Data from the airgun shots at approximately 160 m spacings were recorded on four OBSs deployed at 20 km intervals to provide a series of reverse profiles extending over 60 km. The principal interpretation procedure involved calculation of theoretical seismograms and travel-time curves for 2-D velocity structure models and comparisons with observed record sections. The interpreted structure model shows significant lateral variations. Low velocity Pleistocene and Pliocene sediments form an upper layer varying between 0.5 and 1.0 km thick. The principal sedimentary unit is the Tertiary Skonun Formation with interpreted velocities of 2.7 km/s and a gradient averaging 0.4 km/s/km, values that are consistent with well log data. These sediments are generally thicker (approximately 2.5 km) on the western side of the sub-basin although they reach their maximum thickness of 3 km in a depression near the central part of the basin. Toward the eastern side of the iv basin, the Tertiary sediments thin to about 1 km as the underlying Tertiary volcanics rise toward the mainland. The maximum sediment thickness in the basin is about 4 km. The upper surface of the volcanic unit shows a pronounced topography which is consistent with the erosional nature of this surface. Velocities for the volcanics vary between 4.8 and 5.0 km/s; thickness of the unit ranges from about 0.2 km to 1.8 km. Below the Tertiary volcanics on the eastern 20 km of the model, a low velocity zone less than 1 km thick had to be introduced to satisfy the data. This zone is inferred to contain Upper Cretaceous sediments. A unit with a poorly constrained velocity of 5.9 km/s which underlies the Tertiary volcanics and low velocity zone on the eastern side is interpreted to be the Paleozoic Alexander Terrane. Most of the characteristics of this model are similar to those determined from an earlier study in the Charlotte sub-basin. An additional component of this thesis project was the development of an interactive procedure for the inversion of densely spaced seismic refraction data by wavefield con-tinuation to derive a l-D velocity-depth profile, and its application to data derived from 2-D structures. The procedure consists of two steps: a slant stack followed by a down-ward continuation. The method was found to yield velocity-depth structures which, when compared with an average velocity-depth structure from the 2-D model, have very similiar gradients and velocity increases. In general the velocity depth curve from the inversion had lower velocities at deeper depths than the averaged 2-D structure. V T A B L E OF C O N T E N T S ABSTRACT ii LIST OF TABLES vi LIST OF FIGURES vii ACKNOWLEDGMENTS ix CHAPTER I INTRODUCTION 1 1.1 Tectonic Evolution of the Queen Charlotte Basin 4 1.2 Subsidence of the Queen Charlotte Basin 10 1.3 Stratigraphy 14 1.3.1 Hydrocarbon potential for the major stratigraphic units . . . 17 1.4 An Outline of the Seismic Refraction Study 18 CHAPTER II DATA ACQUISITION AND ANALYSIS 21 2.1 Data Acquisition 21 2.1.1 Airgun-OBS experiment 21 2.1.2 Description of the OBSs and the airgun 21 2.1.3 Description of the procedure 23 2.2 Data Processing 24 2.2.1 Digitizing, editing and demultiplexing the analog data . . . 24 2.2.2 Basic timing and positioning 25 2.2.3 Special positioning and related timing corrections 25 2.3 Data Analysis and Interpretation Procedures 28 2.3.1 Data analysis 28 2.3.2 Interpretation procedures 28 CHAPTER III INVERSION BY WAVEFIELD CONTINUATION . . . 33 3.1 Introduction 33 3.2 The Linear Transformation 34 3.2.1 The slant stack procedure 34 3.2.2 The downward continuation procedure 39 3.3 Examples 43 3.3.1 The plane-layered synthetic example 44 3.3.2 The 2-D synthetic example 52 3.3.3 A real data example 57 3.3.4 Summary 64 vi CHAPTER IV MODELLING OF THE AIRGUN-OBS DATA 68 4.1 Introduction 68 4.2 Initial Constraints 70 4.3 OBS 1 - OBS 2 Submodel 70 4.3.1 Forward profile 71 4.3.2 Reverse profile 73 4.3.3 Summary 78 4.4 OBS 2 - OBS 3 Sub-model 80 4.4.1 Forward profile 83 4.4.2 Reverse profile 86 4.4.3 Summary 90 4.5 OBS 3 - OBS 4 Sub-model 92 4.5.1 Forward profile 93 4.5.2 Reverse profile 97 4.5.3 Summary 104 CHAPTER V DISCUSSION AND CONCLUSIONS 106 5.1 Discussion of the Final Composite Model 106 5.2 Conclusion I l l BIBLIOGRAPHY • 115 APPENDIX I Summary of Formations for the Queen Charlotte Region . 119 APPENDIX II Horizontal and Hydrophone Component Data 121 v i i LIST O F T A B L E S 2.1 O B S l o c a t i o n a n d d e p t h of a i rgun 24 2.2 O B S p o s i t i o n i n g co r rec t i ons 27 3.1 1-D m o d e l used for invers ion tes t ing 37 5.1 S u m m a r y of s t r a t i g r a p h i c i n te rp re ta t i on 109 viii LIST OF FIGURES 1.1 Location map for the Queen Charlotte region 2 1.2 Late Mesozoic—Cenozoic basins of the Pacific margin 3 1.3 Tectonic evolution of the Pacific Northwest 7 1.4 Present day relationship between Wrangellia and Alexander terranes . . . 8 1.5 Isopaeh of the Tertiary sediments for QCB 11 1.6 Rifting sequence for the QCB 12 1.7 Estimated present heat flow for the QCB 13 1.8 Cross section for the offshore QCB 16 1.9 Structural geological model for QCB 19 2.1 Plan of the 1983 airgun/OBS survey 22 2.2 Power spectra noise and signal + noise 29 2.3 Comparison of unfiltered and filtered data 30 3.1 Theoretical seismograms and slant stack for l-D model 38 3.2 Downward continuation for l-D model with exact v-z function 45 3.3 Downward continuation for l-D model 46 3.4 Comparison of p-z functions for l-D model 51 3.5 Theoretical seismograms and slant stack for 2-D model 53 3.6 Downward continuation for 2-D model 55 3.7 Comparison of p-z functions for 2-D model 58 3.8 Seismograms and slant stack for OBS 3 data 59 3.9 Downward continuation for OBS 3 data 61 3.10 Comparison of p-z functions for real data 65 4.1 Final velocity model 69 ix 4.2 OBS 1 data with travel-times and model 72 4.3 OBS 1 seismograms 74 4.4 OBS 2 reverse profile with travel-times and model 76 4.5 OBS 2 reverse profile seismograms 77 4.6 Velocity cube for OBS 1—OBS 2 79 4.7 Comparison of sonic log with results from OBS 1—OBS 2 model . . . . 81 4.8 Sonic logs from Coho, Tyee and Sockeye wells 82 4.9 OBS 2 forward profile with travel-times and model 84 4.10 OBS 2 forward profile seismograms 85 4.11 OBS 3 reverse profile with travel-times and model 87 4.12 OBS 3 reverse profile seismograms 89 4.13 Velocity cube for OBS 2—OBS 3 91 4.14 OBS 3 forward profile with travel-times and model 93 4.15 OBS 3 forward profile seismograms 95 4.16 OBS 4 data with travel-times and model 98 4.17 OBS 4 seismograms 100 4.18 Comparison of data for the three components of OBS 4 103 4.19 Velocity cube for OBS 3—OBS 4 105 5.1 Final velocity structural model for Hecate Strait 107 5.2 Gravity profile and interpretation for Hecate Strait 112 X A C K N O W L E D G E M E N T S I would like to thank my supervisor, Dr. Ron M. Clowes, for his support and encour-agement over the past years. I am particularly grateful for his patient and painstaking reviews of this thesis. I would also like to thank Dr. Robert M. Ellis for critically reviewing this thesis, particularly in light of the severe time constraints. Without the invaluable help from Kathy Penney and Andy Boland during the final stages of this thesis it wo.uld not have been completed in time. This is also true of David Mackie and Joane Berube who offered much needed assistance in the preparation of figures. David Mackie is also responsible for many of the programs used in processing the origional analog tapes. The efforts of Phil Ross and Elaine Wright are also greatly appreciated. The use of TgX files compiled by Sonya Dehler permitted the quick un-derstanding of this text writing package. The people of the Geophysics and Astronomy Department are responsible for making the time spent at UBC very enjoyable. Financial support for data acquisition and analysis was provided by NSERC through Operating Grant A7707 and Strategic (Oceans) Grant G0738, by D.S.S. contract Q5SU.23235-3-1089 from the Earth Physics Branch, E.M.R., and by D.S.S. contracts 06SB.23445-4-1170 and 11SB.23227-5-0127 from the Pacific Geoscience Centre, E.M.R. Additional support was provided by Chevron Canada Resources Ltd., Shell Canada Resources Ltd. and Mobil Oil Canada Ltd.. 1 C H A P T E R I INTRODUCTION Heca te S t r a i t is s i t u a t e d between the west coast of m a i n l a n d B r i t i s h C o l u m b i a and the Q u e e n C h a r l o t t e Is lands in the l a t i t u d e range 52° to 54° N . It is under la in by the Q u e e n C h a r l o t t e b a s i n , a n o r t h - s o u t h t r e n d i n g depress ion 400 k m long and 100 k m wide w h i c h to the nor thwes t a lso under l ies no r theas te rn G r a h a m Is land and D i x o n E n t r a n c e a n d to the south Queen C h a r l o t t e S o u n d (F igu re 1.1). T h e Q u e e n C h a r l o t t e bas in forms p a r t of the coasta l depress ion bo rde r i ng the western marg in of N o r t h A m e r i c a and is one of m a n y sed imen ta ry bas ins w i t h ages rang ing f r om L a t e Mesozo ie to Cenozo i c (F igure 1.2). In t e rms of the bas in ' s s ize , i t resembles the C o o k Inlet a n d B r i s t o l B a y basins i n A l a s k a a n d the S a e r e m e n t o - S a n J o a q u i n bas in in C a l i f o r n i a ( Y o u n g , 1981). T h e h y d r o c a r b o n po ten t i a l of the a rea was invest igated as ear ly as 1912 w i t h the d r i l l i n g of the T i a n N o . l we l l w h i c h reached a dep th of 490 m and b o t t o m e d in Pa leocene M a s s e t vo l can i cs (see Tab le 1 of A p p e n d i x 1 for f o rma t i on n o m e n c l a t u r e ) . T h i s wel l and o thers d r i l l ed on the i s lands d u r i n g the pe r i od 1950-1961 d id not show any s ign i f icant h y d r o c a r b o n finds. E x p l o r a t i o n was renewed in the 1960's when She l l C a n a d a L t d . d r i l l e d a n u m b e r of wel ls off the west coast of B r i t i s h C o l u m b i a , i n c l u d i n g eight in H e c a t e S t ra i t and Q u e e n C h a r l o t t e S o u n d (F igu re 1.1; S h o u l d i c e 1971, 1973). T h e e x p l o r a t o r y wel ls in H e c a t e S t ra i t were d r i l l ed on s t ruc tu ra l h ighs def ined by re la t ive ly p o o r ref lect ion se ismic d a t a and also d i d not resul t in any s ign i f i cant oi l a n d gas shows. A c t i v i t y ended w i t h the m o r a t o r i u m on exp lo ra t i on in the ear ly 1970's. B e t w e e n 1970 and the present , severa l fac tors have con t r i bu ted to a renewed interest i n the bas ins off the west coast of B r i t i s h C o l u m b i a , and p a r t i c u l a r l y the Q u e e n C h a r -l o t te bas in . T h e re l a t i onsh ip between p late tec ton ics and the w o r l d - w i d e d i s t r i bu t i on 2 132 1 3 0 128 Figure 1.1 Location map for the Queen Charlotte Islands, Hecate Strait and Queen Charlotte Sound. Exploratory wells drilled by Richfield Oil Corporation on the Queen Charlottte Islands and Shell Canada Ltd. in Hecate Strait and Queen Charlotte Sound are indicated by boxed labels ( after Young, 1981). 3 LATE MESOZOIC — CENOZOIC BASINS WESTERN NORTH AMERICA Major sedimentary basins Calc - alkaline plutonic rocks Geosynclinal structural trends Tertiary — Quaternary volcanic cover Spreading center £ Subduction zone / Transform fault Figure 1.2 Late Mesozoic—Cenozoic basins of the of the Pacific margin ( after Young, 1981). 4 of h y d r o c a r b o n s h igh l i gh ted m a n y p rospec t i ve areas (eg. R o n a , 1980). O i l compan ies began to adop t these p r inc ip les in the i r ana l ys i s o f bas in deve lopmen t . T h e concept of acc re ted te r ranes , w h i c h are f r agmen ts of more anc ien t ma te r ia l t ha t has been j u x t a -posed aga ins t o ther ter ranes a n d / o r c ra tons w i t h vas t ly d i f fer ing evo lu t i ons , was easi ly a c c o m m o d a t e d by p la te tec ton ic theory . T h e A l e x a n d e r and W r a n g e l l i a Ter ranes have been iden t i f ied as examp les of th i s process ( B e r g et al., 1972; and Jones et al., 1977 respec t i ve l y ) . T h e s e accre ted ter ranes o r ig ina ted in more equato r ia l env i r onmen ts , con -sequent l y m a y have deve loped o rgan i c - r i ch f o rma t i ons unre la ted to the ad jo in ing ter-ranes , and thus w a r r a n t i n d e p e n d e n t i nves t iga t ion . T h e recent s tud ies of the evo lu t ion of the Q u e e n C h a r l o t t e basin and s u r r o u n d i n g reg ions by Y o r a t h and C h a s e (1981), Y o r a t h and C a m e r o n (1982), Y o r a t h and H y n d m a n (1983), M a c k i e (1985), C lowes and G e n s - L e n a r t o w i e z (1985) and D e h l e r (1986) have con t r i bu ted great ly to the present u n d e r s t a n d i n g of the reg ion. 1.1 Tectonic Evolution of the Queen Charlotte Basin Y o r a t h and C h a s e (1981) rev iewed the wo rk of Su the r l and B r o w n (1968) and S h o u l d i c e (1971, 1973) , and in teg ra ted th is w i t h the i r own and o ther s tud ies to de-ve lop a m o d e l for the evo lu t ion of the Q u e e n C h a r l o t t e bas in . T h e y s imp l i f ied the geology a n d tec ton ics of the region by de f in ing four basic tec ton ic assemblages. T h e Pa leozo i c A l e x a n d e r Te r rane and the M e s o z o i c Wrange l l i a Ter rane compr i se the a l -l o c h t h o n o u s assemblages; the U p p e r Ju rass i c p lu tons and L o w e r C re taceous L o n g a r m F o r m a t i o n (Tab le 1, A p p e n d i x I) m a k e up the suture Assemb lage ; the post suture as-semblage cons is ts o f the M i d d l e to U p p e r C re taceous Q u e e n C h a r l o t t e G r o u p (Tab le 1, A p p e n d i x I); and the m i d - T e r t i a r y p lu tons , M a s s e t vo lcan ics and Neogene S k o n u n f o r m a t i o n c o m p r i s e the r i f t assemblage. 5 Berg et al. (1972) denned the Alexander Terrane as a complex assemblage of sedi-mentary, igneous and metamorphic rocks ranging from Late Precambrian to Late Paleo-zoic in age. Rocks of Late Triassic age unconformably overlie Permian limestones in the southern part of the terrane (Berg et al., 1978). The metasedimentary and metavolcanic rocks along the margin of the Coast Mountains have also been included in the Alexander Terrane by Yorath and Chase (1981). Van der Voo et al. (1980) and Van der Voo and Channel (1980) have obtained paleomagnetic results from Ordovician, Devonian and Carboniferous rocks from southeastern Alaska which clearly indicate the exotic nature of the Alexander Terrane. Displacements of 1800 km relative to eratonie North America between Late Carboniferous and Triassic time have been determined. Jones et al. (1977) described the terrane underlain by Middle to Upper Triassic tholeitie basalts and calcareous sedimentary rocks which occur in the Wrangell Moun-tains and Chichagof Islands as Wrangellia. Yorath and Chase (1981) also included Jurassic volcanic and sedimentary rocks under the definition for Wrangellia. The 4300 m of tholeitie pillow lavas, pillow basalts, equagene tuffs, and the massive basalt flows of the Upper Triassic Karmutsen Formation conformably overlie about 100 m of limestone and argillite of the Upper Triassic to Lower Jurassic Kunga Formation (Sutherland Brown, 1968). This formation, together with the terrigenous elastics and calc-alkaline volcanics of Early to Middle Jurassic age on Queen Charlotte Islands (the Maude and Yakoun Formations), are included in Wrangellia by Yorath and Chase (1981) following upon evidence from Tipper and Cameron (1980), who found ammonite faunas in the Yakoun Formation that developed in a warm-water environment. Paleomagnetic evi-dence from the Nicolai Greenstone belt in the Wrangell Mountains (Hillhouse, 1977) suggests a northward displacement of 3000 km or 6000 km prior to their accretion to 6 the North American continental margin (the different results depend upon whether a northern or southern hemispherical solution is chosen). The model proposed by Yorath and Cameron (1982) for the evolution of the Pacific margin in the vicinity of the Queen Charlotte Islands is depicted in Figure 1.3. By late Triassic time, the Alexander Terrane had reached its present latitude (Figure 1.3a). There was a cessation of the northward movement of the Alexander Terrane following collision with Wrangellia. By some process, the northward movement of the two terranes was converted to movement in a northeasterly direction (Figure 1.3b). The unconfor-mity between the Longarm Formation, characterizing the suture zone, and the Middle Jurassic rocks may be indicative of the Lower Cretaceous-Upper Jurassic collision event (Table 1, Appendix 1). The Longarm Formation consists of sandstones, siltstones and conglomerates (Sutherland-Brown, 1968). Sutherland Brown (1968) noted large blocks of andesitic volcanic rocks (8 to 10 m in size) occurring closely above the unconformable contact. Such large clasts would have to have been derived from the rapid development of substantia] relief in zones adjacent to the suture zone (Yorath and Chase, 1981). Figure 1.4 shows the major faults in the region of the Queen Chatlotte Islands. The offshore trace of the Rennell Sound Fault, which coincides with the trend of a promi-nent gravity anomaly and a seismic reflection survey interpreted by Yorath (Yorath and Chase, 1981), has led the authors to suggest that this fault zone represents the bound-ary between the Alexander and Wrangellia terranes. They propose that the structural depression, as evident from gravity and seismic data, was possibly due to a relocation of compressive stresses in a post-collisional regime. Following the Late Jurassic to Early Cretaceous suturing, the two terranes then docked with North America somewhere between 90 and 40 Ma ago (Figure 1.3c) and may have been responsible for the plutonic uplift within the Coast Mountains. The post > 140 MILLION YEARS AGO WRANGELLIA MOVES NORTHWARD FROM THE SOUTHERN HEMISPHERE ABOUT 140 MILLION YEARS AGO WRANGELLIA COLLIDES WITH THE ALEXANDER TERRANE 90 TO 40 MILLION YEARS A G O COMBINED WRANG.ELLIA AND ALEXANDER TERRANE DOCK WITH NORTH AMERICA Figure 1.3 Schematic diagram for the tectonic evolution of the Pacific Northwest margin showing the collision of Wrangellia with Alexander and then with the Pacific Margin (Yorath and Cameron, 1982). 8 i | Figure 1.4 Present day relationship between Wrangellia and Alexander terranes showing the major . faults and the position of the proto-Queen Charlotte Islands ( after Yorath and Hyndman, 1983). | 9 su tu re assemblage, c o m p o s e d of the M i d to U p p e r C re taceous Queen C h a r l o t t e G r o u p , was depos i ted across the suture zone and overl ies W r a n g e l l i a , the A l e x a n d e r Ter rane a n d the suture assemblage. T h e f o u r t h e v o l u t i o n a r y stage is represented by the M a s s e t a n d S k o n u n F o r m a t i o n s a n d by a series of i n te rmed ia te ep izona l p l u tons , a l l of Ter t ia ry age, denned by Y o r a t h a n d C h a s e (1981) as the ri f t assemblage. A l s o re la ted to the rift assemblage are several pe ra l ka l i ne vo l can i c cent res of the A n a h i m vo l can i c be l t . T h e M a s s e t F o r m a t i o n is composed of py roc las t i c rocks cons i s t i ng p r ima r i l y of a l -kal ic basa l t and sod ic rhyo l i te (Su the r l and B r o w n , 1968). T h e f o r m a t i o n is 1200 to 5500 m th i c k , e r u p t e d subaer ia l l y for the most p a r t , and rests u n c o n f o r m a b l y on a l l o lde r un i ts on the Q u e e n C h a r l o t t e Is lands. T h e age of the f o r m a t i o n var ies between P a l e o c e n e and M i o c e n e (62 and 11 M a ) ( Y o u n g , 1981). T h e mean age of the Masse t vo l can i cs for the Q u e e n C h a r l o t t e Is lands is 27 M a (La te O l i goeene) , wh i le for Heca te S t r a i t and Q u e e n C h a r l o t t e S o u n d i t is 35 M a ( E a r l y O l igoeene) . T h e or ig in of the f o r m a t i o n has been suggested as uppe r man t l e ma te r i a l for the basal ts and remel ted Pa leozo i c p lu ton ic r ocks for the rhyo l i tes ( S u t h e r l a n d B r o w n , 1968). T h e S k o n u n F o r m a t i o n cons is ts of m a r i n e a n d non -mar ine sands , sands tone , shale, l ign i te s t r ingers a n d cong lomera tes ( S u t h e r l a n d B r o w n , 1968). W h e r e f ound on the Q u e e n C h a r l o t t e I s lands , it rests u n c o n f o r m a b l y on the M a s s e t vo lcan ics . T h e fo rma t ion t h i ckens towards the east in to Heca te S t r a i t and Q u e e n C h a r l o t t e S o u n d and ranges in age f r om E a r l y M i o c e n e to L a t e P l i o c e n e . N o r t h of the a s s u m e d suture zone, the S k o n u n a n d Masse t fo rmat ions overly the C r e t a c e o u s a n d / o r Ju rass i c rocks of the post su ture assemblage ( Y o r a t h and C h a s e , 10 1981) . T h e y are also found in fau l t con tac t w i t h the Jurass ic rocks of Wrange l l i a a n d the C r e t a c e o u s s t r a t a of the pos t - su tu re assemblage on the Q u e e n C h a r l o t t e Is lands. 1.2 Subsidence of the Queen Charlotte basin T h e Q u e e n C h a r l o t t e s e d i m e n t a r y bas in is compr i sed of several subs id ia r y bas ins , of w h i c h the H e c a t e sub -bas in and the C h a r l o t t e sub-bas in are cons ide red the mos t i m p o r t a n t . These are separa ted by a basement t o p o g r a p h i c high k n o w n as the M o r e s b y R i d g e , w h i c h r uns i n a nor theas te r ly d i rec t i on f r o m sou thern M o r e s b y I s land (F igu re 1.5). T h e basemen t r idge is l oca ted rough l y in the center of the Q u e e n C h a r l o t t e b a s i n . Y o r a t h and H y n d m a n (1983) have p roposed a m o d e l for the subs idence of the Queen C h a r l o t t e bas in w h i c h comb ines an i n i t i a l stage of r i f t ing , fo l lowed by f lexura l d o w n -w a r p i n g and sed imen t l o a d i n g , w h i c h resul ted f r o m the obl ique convergence of oceanic l i t hosphe re w i t h the N o r t h A m e r i c a n con t i nen ta l m a r g i n . T h e y argue t ha t i f the t race of the A n a h i m hot spo t is e x t r a p o l a t e d backwards in t ime , it wou ld have passed beneath Q u e e n C h a r l o t t e S o u n d (F igure 1.6a). F u r t h e r m o r e , the r i f t ing and c rus ta l ex tens ion w h i c h o c c u r r e d were respons ib le for the w idesp read subarea l M a s s e t vo l can ics . T h i s r i f t i ng ac t i va ted or reac t i va ted the R e n n e l l S o u n d - S a n d s p i t Fau l t , w h i c h a l lowed the p r o t o - Q u e e n C h a r l o t t e Islands to m o v e to the i r present l a t i t ude ( F i g u r e 1.6b). H e a t flow f r o m the Q u e e n C h a r l o t t e bas in of fshore wel ls is high for the n o r t h e r n three wel ls ( F i g u r e 1.7), p a r t i c u l a r l y the Sockeye we l l , w h i c h co inc ides w i t h the p roposed suture be-tween W r a n g e l l i a and A l e x a n d e r te r ranes . T h e depressed heat flow in the four sou the rn we l ls is t h o u g h t t o be due to recent u n d e r t h r u s t i n g of oceanic l i t hosphere . T h e r i f t ing is cons ide red to have occu r red d u r i n g a pe r i od of reg iona l upl i f t and to have c o m m e n c e d a b o u t 21 M a ago and ceased 17 M a ago ( Y o r a t h a n d H y n d m a n , 1983). 1 1 I ISOPACH OF NEOGENE SKONUN FORMATION O B S • WELL * 0 SO 100 km i 1 I ! 1_ 133° 132° 131° 130° 129° Figure 1.5 1971, 1973). Isopach of the Tertiary Skonun sediments for the Queen Charlotte Basin ( after Shouldice, Location of the airgun /OBS survey and the three nearest wells is also indicated. A. 20 TO 17 MILLION YEARS AGO Alexander Terrane (A.T.) » • • • >JfJ>*? Paleozoic rocks • p p p m Wrangellia (W.) rocks Rift zones Mesozoic B. ABOUT 17 MILLION YEARS AGO Trace of Anahim Hot Spot C. M. Continental Margin RSF Rennel Sound Fault SF Sandspit Fault Figure 1.6 Rifting sequence for the Queen Charlotte Basin ( after Yorath and Cameron, 1982). 13 TOO-, 8 0 -'e | 6 0 -S; o 4 0 -u. £ 2 0 -f-H->• Ui oc C L W O SOUTH o Ui _ J cc < X w i t h o u t c o o l i n g b y u n d e r t h r u s t i n g Ui Ui >• 0 r •: r 1 0 0 2 0 0 D I S T A N C E ( k m ) O 8; NORTH 3 0 0 Figure 1.7 Estimated present heat flow based on down hole temperature measurements and thermal conductivities computed from porosity and mineralogy data from the Queen Charlotte Basin offshore wells. The dashed curve of higher heat flow for the southern wells is a possible profile if no underthrust cooling had occurred ( after Yorath and Hyndman, 1983). 14 Yorath and Hyndham (1983) have interpreted the subsidence curves for all the off-shore wells drilled in Queen Charlotte basin. Present depths to biostratigraphic horizons were converted to a basement subsidence history through corrections for sediment com-paction and paleowater depths. The sediment compaction correction was derived using an exponential approximation of the decrease in porosity with depth. Tectonic sub-sidence curves were then estimated, thus correcting for sediment loading. The major uncertainty in their interpretation is in the biostratigraphic data reported by Shouldice (1971). The tectonic curves first showed uplift, which resulted in the unconformity, at the end of the Late Miocene or Early Pliocene. Then 6 Ma ago there was a sudden onset of subsidence which has continued at a decreasing rate to the present. The Queen Charlotte Islands appear to have been uplifted and eroded, particularly on the west coast of the islands. The flexural model developed to explain this second stage of subsidence is based upon the assumption of oblique convergence at the North American continental margin. The subsidence due to the flexural downwarping was further amplified by sediment loading. The Oshawa rise, west of the islands, and the missing Upper Miocene and Pliocene sediments on Graham Island west of the hinge line where subsidence is balanced by uplift are cited by Yorath and Hyndman (1983) in support of this model. 1 .3 Stratigraphy Much of the information on the Neogene Skonun succession was obtained from the exploratory wells drilled on northeastern Graham Island by Richfield Oil Corporation and in Hecate Strait and Queen Charlotte Sound by Shell Canada Ltd. The location of the wells is shown on the map in Figure 1.1 and a cross section derived from the 15 well data is depicted in Figure 1.8. There are several striking features evident on the cross section. As already noted, there are several smaller basins within the Queen Charlotte basin that are controlled by the unconformity at the base of the Tertiary Skonun sediments. In the central part of Hecate Strait, the Moresby Ridge, between the Sockeye and Murrelet wells, separates the main part of the Queen Charlotte basin into the Hecate sub-basin and the Charlotte sub-basin (Figure 1.5). The Tertiary sediments are primarily non-marine in the Hecate sub-basin and marine to the south in the Queen Charlotte sub-basin. Within the Hecate sub-basin, the major source was probably the Coast Mountains with small contributions from the Queen Charlotte Islands (Young, 1981). The Char-lotte sub-basin, of marine Lower Miocene - Upper Pliocene succession, was probably deposited in water depths ranging from shallow to deep (0-50m to 200-1000m) based on biostratigraphic information. Underlying the Tertiary Skonun sediments and Masset volcanics are the Skidegate Formation conformably resting on the Honna formation. The two formations represent the upper and middle units of the Queen Charlotte Group (Sutherland Brown, 1968); they are included in the post suture assemblage of Yorath and Chase (1981). The Honna Formation consists of conglomerate and coarse arkasic sandstone with minor shale or siltstone. The Skidegate Formation consists of fine grained detrital rocks, siltstone, silty shale, fine to medium sandstone and calcareous shale and sandstone. Both formations are estimated to be Late Cretaceous in age with a combined thickness ranging between 800 to 2100 m. Beneath the Queen Charlotte Group, within Wrangellia, the Kunga, Maude and Yakoun Formations locally contain significant quantities of heavy fraction hydrocarbons locally on the Queen Charlotte Islands. The Kunga Formation is primarily composed 16 C. BALL TYEE SOCK. AUK. OSP. COHO MUR. HARL S.L. 4000-r S . L . 1000m -2000m 3 N-M. Ss H H N-M. S l ts t -Sh WMl M. Ss H H M.Sl ts t -Sh • OIL SHOW ™ — FLORA <™™™ FLORA-FAUNA Figure 1.8 Cross section for the offshore wells drilled by Shell Canada Ltd. showing the unconformity controlling the basin. Stratigraphic correlations are based upon flora and fauna from well cores ( after Shouldice, 1971, 1973). 17 of limestone and argillite with its age ranging from Early Upper Triassic to mid Lower Jurassic. The Maude Formation, where present, rests conformably on the Kunga For-mation and is composed of argillite, shale, calcareous shale and lithic sandstone. The contact with the Kunga Formation is gradational, extending over many meters. The age of the Formation is estimated to be Late Jurassic with a total thickness of 225 m. It is seen on the southeastern portion of Graham Island, northeastern Moresby Island and Lyell Island. The Yakoun Formation contains some non-marine, as well as marine sediments. The formation consists primarily of pyroclastic rocks, many of which are formed largely of porphyritic andesite. It also includes volcanic sandstone, some eon-glomerate, shale, siltstone and minor coal. The age of the formation is between Middle Jurassic and earliest Upper Jurassic. Sutherland Brown (1968) described the volcanic cones which rose above sea level during this period to have once been clothed in lush Jurassic forests. The total thickness is approximatley 900 m. Table 1 of Appendix I summarizes each formation's lithology, age and thickness. 1.3.1 Hydrocarbon potential for the major stratigraphic units The Upper Triassic and Jurassic rocks of Wrangellia, the Kunga, Maude and Yakoun formations, are the most favourable prospect for hydrocarbon generation. These are the Kunga, Maude and Yakoun Formations. In some places, Yorath and Cameron (1982) report that these rocks have been described as oil shales. Offshore, they probably occur at considerable depths but the oil may have migrated along faults into either the Upper Cretaceous sediments or the Tertiary sediments. The Cretaceous Honna Formation has good porosity and fair permeability (Shouldice, 1971, 1973) and is thought to be a good reservoir rock. The Tertiary sediments contain the necessary stratigraphic and structural traps for containing the oil which may have migrated into these sediments. 18 Another source of oil may be from the Tertiary sediments within the rift zones (Figure 1.6). If the increased heat flow seen in the wells (Figure 1.7) was higher in the past, oil may have been generated. The heat flux is certainly sufficient to have produced more than an immature gas. However if the heat flow was too high for too long, only dry gas would have been generated (Yorath and Cameron, 1982). Below the Hecate sub-basin north of the suture zone, the Alexander Terrane rocks are not known to contain hydrocarbons. Therefore, only the Tertiary rocks could have generated oil. However, the geothermal gradients are thought to have been normal, as no rifting occurred in the Hecate sub-basin and only small amounts of immature gas may be expected (Yorath and Cameron, 1982). The lack of a sufficient thickness for the sediments combined with their primarily non-marine composition would further support this conclusion. Another prospect, albeit costly, exists within the Cretaceous sediments which lie be-neath the Tertiary lavas. Exploration, using conventional seismic reflection techniques, would be difficult as the volcanic rocks present a strong acoustic barrier to adequate penetration of sound waves. The Tertiary volcanics are known to thin and disappear in some locations (Figure 1.9; Shouldice 1971, 1973), which may aid in the exploration of this stratigraphic zone. 1.4 A n Outline of The Seismic Refraction Study The refraction experiment was designed to determine the seismic velocity structure of the Hecate sub-basin, particularly below the Tertiary sediments where the thickness of the Masset volcanics was unknown. Also, the existence or not of sediments below the volcanics was important to evaluate this region for any further exploration programs. 19 24000* 6000m ip i T SEDIMENTS Wi T VOLCANIC Mz SED.-VOLC. PLUTONS 0 u Mi 100 —i 6 Km 100 Figure 1.0 Structural geological model inferred from the cross section in figure 1.8, industry seismic reflection and refraction data and gravity and magnetics. Projected position of the four OBSs is indicated by arrows ( after Shouldice, 1973). 20 T h e i n i t i a l m o d e l , in ferred by S h o u l d i c e (1971, 1973) f rom wel l d a t a and ava i lab le geophys i ca l d a t a , is shown in F i g u r e 1.9. T h e u p p e r bounda ry and even the presence of the Mesozo i c sed imen ts and vo l can i cs is poor ly def ined beneath the Heca te s u b - b a s i n . T h e pos i t ion of th i s un i t in F i g u r e 1.9 is based u p o n l im i ted we l l log i n f o rma t i on and o u t c r o p s seen on the l and on e i ther s ide of the b a s i n . To ver i fy a n d / o r mod i f y the m o d e l of F i g u r e 1.9, an a i r g u n / O B S survey was car -r ied ou t in 1983. F o u r O B S s were dep loyed at 20 k m in terva ls across Heca te S t ra i t ; a i r gun shots at a b o u t 0.2 k m spac ings were reco rded to p rov ide three reversed prof i les e x t e n d i n g over 60 k m . C h a p t e r II g ives the de ta i l s of the d a t a acqu i s i t i on and a n a l -ys is p rocedures . In C h a p t e r I V , the record sec t ions and the i r i n te rp re ta t i on t h rough c o m p a r i s o n s w i t h theore t i ca l sect ions for 2 -D ve loc i t y mode ls are desc r i bed . T h e three s u b - m o d e l s resu l t i ng f rom in te rp re ta t i on of the i n d i v i d u a l reversed profi les are compos -i t ed to p rov ide a comp le te m o d e l across the b a s i n . A d iscuss ion of the re l iab i l i t y of th is m o d e l a n d i ts r e l a t i onsh ip to the l oca l geology for the Hecate sub -bas in are p rov ided in C h a p t e r V . C h a p t e r III con ta ins an a d d i t i o n a l c o m p o n e n t to th is thesis - the deve lopmen t of an i n te rac t i ve p rocedure for i nve rs ion of re f rac t ion d a t a by wavef ie ld con t i nua t i on . T h e theore t i ca l bas is is ou t l ined a n d and i ts a p p l i c a t i o n to the syn the t i c a n d real d a t a of C h a p t e r I V is s h o w n . 21 C H A P T E R II DATA ACQUISITION A N D ANALYSIS 2.1 D a t a A c q u i s i t i o n 2.1.1 A i r g u n - O B S e x p e r i m e n t T h e a i r g u n / o c e a n b o t t o m se i smograph ( O B S ) re f rac t ion p r o g r a m in Heca te S t ra i t w a s u n d e r t a k e n as one c o m p o n e n t of a larger re f rac t ion p r o g r a m . In coopera t i on w i t h the E a r t h P h y s i c s B r a n c h ( E P B ) and the Pac i f i c Geosc ience C e n t e r ( P G C ) , the U n i -ve rs i t y of B r i t i s h C o l u m b i a ca r r ied out an onshore-of fshore ref ract ion p r o g r a m d u r i n g A u g u s t , 1983. T h i s large 330 k m prof i le was recorded f rom the deep ocean across no r th -e rn M o r e s b y Is land and Heca te S t r a i t , to the m a i n l a n d of B r i t i s h C o l u m b i a . Seventeen se i smographs were d e p l o y e d : 11 land based se ismographs and s ix O B S s ; a n d two energy sources were e m p l o y e d : T N T exp los ives and an a i r g u n . T h e ob jec t ive of t h i s thesis was to inves t iga te the u p p e r 6 k m of the c rus t beneath Heca te S t ra i t . T h e d a t a , recorded on the four O B S s dep loyed across Heca te S t r a i t us ing a 32 l i t re ( 2 0 0 0 m 3 ) a i rgun source, were u t i l i zed to deve lop a se ismic ve loc i t y s t r uc tu ra l m o d e l to meet th is ob jec t ive . T h e l o c a t i o n and p lan of the a i r g u n / O B S e x p e r i m e n t is shown in F igu re 2.1. 2.1.2 D e s c r i p t i o n o f t h e O B S s a n d t h e a i r g u n T h e O B S s were bu i l t at U B C and fo l lowed the design used by A t l a n t i c Geoscience C e n t e r (Heff ler and B a r r e t t , 1979) w i t h some m i n o r mod i f i ca t i ons i n c o r p o r a t e d d u r i n g c o n s t r u c t i o n . T h e O B S cons is ts of four m a i n c o m p o n e n t s : ( l ) the de l ivery sys tem c o m p r i s e d of a glass f l o ta t i on sphere , a pressure case and an anchor ; (2) the se ismic wave d e t e c t i o n sys tem represented by a ver t i ca l a n d a h o r i z o n t a l g imba led 4.5 H z geophone Figure 2.1 Location and plan of the 1983 airgun/OBS survey. OBSs are indicated by darkened circles numbered 1 to 4 and the Coho, Tyee and Sockeye wells are marked by stars. 23 housed w i t h i n the pressure case, a n d a h y d r o p h o n e , m o u n t e d on the side of the flotation sphere ; (3) the reco rd i ng s y s t e m w h i c h uses a four channe l s low speed d i rec t record t a p e un i t t o store the amp l i f i ed s ignal f r o m the two geophones and h y d r o p h o n e ; (4) the c l o c k , ra ted aga ins t W W V B at the t i m e of d e p l o y m e n t a n d recovery, generates a 10 H z a m p l i t u d e m o d u l a t e d t i m e code . T h e ope ra t i on of the O B S is governed by a m i c r o p r o c e s s o r and the s igna ls f r o m the three c o m p o n e n t s a long w i t h the t i m e code were reco rded on cassette t ape . T h e passband of the O B S s lies between 4.5 and 30 H z (C lowes , 1985). T h e se ismic source was a 32 l i t re ( 2 0 0 0 m 3 ) a i rgun w h i c h p rov ided energy equ iva len t to a b o u t 4 kg w i t h i n the se ismic p a s s b a n d . T h e a i r gun f i r ing was con t ro l l ed by a m ic rop rocesso r w h i c h t r iggered' the a i rgun at one m inu te in te rva ls us ing a c lock w h i c h was ra ted aga ins t W W V B t ime code a t the s tar t of the s h o o t i n g . 2.1.3 Description of the procedure T h e O B S s were dep loyed a t a p p r o x i m a t e l y 20 k m spac ing across H e c a t e S t ra i t in wa te r dep ths rang ing f r om 22 m for O B S 1 in the west to 162 m for O B S 4 in the east ( F i g u r e 2.1). T h e dep ths were d e t e r m i n e d us ing the sh ip ' s dep th s o u n d i n g sys tem as reco rded on an E P C l ine scan recorder . T h e la t i t udes a n d long i tudes were de r i ved f rom read ings of the sh ip ' s p r i n c i p a l nav iga t i on s y s t e m , L o r a n C , at the t ime of d e p l o y m e n t . Tab le 2.1 l ists the d e p t h and p o s i t i o n for each O B S . T h e sh ip s ta r ted f rom the p o i n t o f d e p l o y m e n t of O B S 4 in the west . T h e a i rgun was towed beh ind the sh ip at a d e p t h of 18 met res un t i l i t neared O B S 2, where the d e p t h was decreased to 14 m for the r e m a i n i n g po r t i on of the prof i le. T h i s firing rate of one m i n u t e , c o m b i n e d w i t h the sh ip ' s speed , resu l ted i n a source in te rva l spac ing 24 OBS LOCATION (N. Long. : W. Lai.) DEPTH AIRGUN DEPTH 1 53.005 : IS 1.4616 22 U 2 53.8008 : 131.2342 82 14 3 58.8060 : 130.9579 122 18 4 53.5158 : 130.7155 163 18 T a b l e 2.1 O B S l oca t i on and dep th and dep th o f a i rgun ( dep ths are g iven in mete rs ) . of a p p r o x i m a t e l y 160 m . R e v e r s e d prof i les for segments of the a i rgun l ine between n e i g h b o u r i n g O B S s arose n a t u r a l l y f r om the design of the expe r imen t . 2.2 D a t a P r o c e s s i n g 2.2.1 D i g i t i z i n g , e d i t i n g a n d d e m u l t i p l e x i n g t h e a n a l o g d a t a T h e d i rec t record casset te t apes were d ig i t i zed using the P D P l l / 3 4 - b a s e d ana log-t o - d i g i t a l convers ion fac i l i t y . T h e d a t a were conver ted at a 120 H z s a m p l i n g rate w i th va r ia t i ons in ana log tape speed taken in to accoun t . T h e d ig i t i zed d a t a were then edi ted i n t o 25 segments cen t red a b o u t the largest a u t o m a t i c a l l y de tec tab le event w i t h i n each 60 s reco rd i ng i n t e r va l . T h e t i m e of f irst s a m p l e for each event was de te rm ined w i th the a i d of a c o m p u t e r p r o g r a m w h i c h keyed on the O B S t ime code. T h e t ime code was not exp l i c i t l y read because the high pass band response of the h y d r o p h o n e to the w a t e r wave had been cross- fed to the t ime code channe l . T h i s fac i l i ta ted the water wave r e c o g n i t i o n , bu t in ter fered w i t h the t ime code s igna l over a range of d a t a po in ts . T h i s 25 p r o b l e m was c i r c u m v e n t e d by de tec t i ng only m i n u t e m a r k s , keeping t rack of the t ime to the po in t where the ed i ted d a t a segments b e g a n , and then s k i p p i n g over the requis i te n u m b e r of d a t a b locks to bypass the water wave in ter ference. T h e ed i ted d a t a were then d e m u l t i p l e x e d and conver ted to an I B M - c o m p a t i b l e f o rma t for p l o t t i ng . 2.2.2 Basic timing and positioning C o r r e c t i o n s were app l i ed to the t ime of the first samp le of each s e i s m o g r a m for dr i f t i n the O B S and a i rgun c locks re la t i ve to W W V B , de lay in the ac tua l firing of the a i rgun af ter the t r igger pu lse and a d j u s t m e n t s for the skewness of the reco rd ing head. S h i p pos i t i ons were de te rm ined by L o r a n C readings logged a t 10 m i n u t e in te rva ls . 2.2.3 Special positioning and related timing corrections T h e nav iga t i on d a t a show tha t the shot l ine d i d not pass d i rec t l y over the O B S s nor d i d the nearest offset, as def ined by these da ta , agree w i t h the offset as de te rm ined f rom first a r r i v a l obse rva t i ons . T o enab le comp i l a t i on of the se ismic d a t a i n to app rop r i a te r eco rd sec t ions for w h i c h 2 - D prof i le record ing is assumed' , d is tance and t i m e cor rec t ions were requ i red . A l s o , the m o s t accura te d a t a f r o m w h i c h shot - rece iver d is tances can be d e t e r m i n e d is the first a r r i va l a n d / o r d i rec t wa te r wave a r r i va l on the se i smograms , r a the r than in te rpo la ted nav iga t i on pos i t ions. A fu l l descr ip t ion of the t i m i n g and p o s i t i o n i n g cor rec t ions a long w i t h a comp le te set o f nav iga t i on and t ime of first sample tab les is g i ven in C lowes (1985). T h e spec ia l pos i t i on i ng and t i m i n g w i l l be restated here as they represent a d j u s t m e n t s w h i c h in f luence a l l the t races pa r t i cu la r l y for the s o u r c e / r e c e i v e r d is tances t ha t are less t h a n 2.0 k m where in l i ne co r rec t ions are most no t i ceab le . 26 C o r r e c t i o n s for s h o t / r e c e i v e r pos i t i ons a n d d is tances were made in three s tages, a n d necess i ta ted a co r respond ing a d j u s t m e n t to t rave l t imes . T h e f i rst stage made use of the n a v i g a t i o n d a t a to place the shots and O B S s a p p r o x i m a t e l y in l ine a n d p rov ide the usua l t race spac ing assoc ia ted w i t h a one m i n u t e f i r ing in te rva l . S i m p l e p l a n a r geometry was used to pro jec t the O B S pos i t ions on to the shot l ine. T h e rev ised (shor tened) shot rece iver d is tances d i d requ i re tha t a c o r r e s p o n d i n g t rave l t i m e co r rec t i on for the e x t r a t rave l pa th be removed . T h i s was the substance of the second stage w h i c h used first a r r i va ls to de te rm ine the t rave l t imes to the O B S s . D i r e c t wa te r waves were not i n fac t the first a r r i va ls , even at short d i s tances , due to the effects of the sha l low water and re f rac t ion t h r o u g h the sub b o t t o m . Consequen t l y , a two-layer m o d e l (water p lus sub b o t t o m , for w h i c h the a p p r o x i m a t e ve loc i t y was de te rm ined f r o m first a r r i v a l da ta ) was used to de te rm ine the best p ro jec ted l oca t i on of the O B S s re la t i ve to the sho ts a long the prof i le . T h i s was accomp l i shed by i t e ra t i ve a d j u s t m e n t s of c a l c u l a t e d in te rcep ts to agree w i t h measu red t rave l t imes . K n o w i n g the d e p t h of the O B S and the d e p t h of the wa te r be low the a i r gun ( f rom 3.5 k H z p ro f i l i ng ) , w h i c h was near ly cons tan t for a few k i l ome te r s on e i ther side of the O B S s , a s imp le ca l cu la t i on us ing P y t h a g o r a s ' t heo rem enab led de te rm ina t i on of the app rop r i a te t ime co r rec t i on . T h e t i m e cor rec t ions app l ied to each O B S a m o u n t e d to less t h a n 0.45 s for the O B S r e q u i r i n g the largest i n l i ne co r rec t i on at zero offset. A t 1 k m offset th is co r rec t i on (wh ich was for O B S 2) had d r o p p e d to 0.11 s. These in l ine t i m i n g correct ions were genera l ly less for the o ther three O B S s and were large c o m p a r e d to the p i ck ing of first b reaks . H o w e v e r , the large shi f ts do not co r respond to a large unce r ta in t y in first a r r i va l p i cks for the af fected t races bu t d o affect secondary a r r i va l pos i t i ons , render ing t h e m un re l i ab le where the cor rec t ions were large (less than 1 k m sou rce / rece i ve r d is tance) . S e c o n d a r y a r r i va l s were on ly m o d e l l e d b e y o n d the 5 k m range where the co r rec t ions 27 for travel time were not significant. This was borne out in the comparison of velocities determined from the corrected traces (2.0 km/s) versus velocities derived from some well logs in the area which yielded velocities of 2.17 km/s at 300 m depths. Details of the calculations can be found in Clowes (1985). The time correction is only applied to the plotting parameter which controls the sample representing zero time. The third stage involved a simple lateral shift to produce agreement between the nearest offset based upon navigation and the nearest offset based upon observed data. Although the relative accuracy of Loran C is approximately + or - 200 m in far offshore regions, it may deteriorate near station locations in near shore areas. Furthermore, there was no check for lane jumps which can result in 1 to 2 kilometer errors in the relative position of the ship. To determine the lateral shift, two methods were applied; one which uses the velocity, and one which used1 the predicted time-distance intercept for a two layer model. Table 2.2 contains a summary of the positioning corrections for each OBS. The details of the calculations for the lateral shift are given in Clowes (1985). Appendix II contains all the corrected sections plotted in variable area format. OFFLINE CORRECTION (km) OBS NAVIGATION BASED ARRIVAL BASED LATERAL SHIFT (km) 1 0.810 0.190 0.995 2 1.170 0.619 1.156 S 0.460 0.124 1.990 4 0.656 0.154 1.160 Table 2.2 OBS positionin g corrections. 28 2.3 Data Analysis and Interpretation Procedures 2.3.1 Data analysis T h e f requency con ten t of the se ismic s ignal and noise recorded on an O B S is i l -l u s t r a t e d by the power s p e c t r a shown in F i g u r e 2.2 . T h e s igna l - to -no ise ra t i o for the r e c o r d sect ions was very good a n d only ce r ta in cases requ i red the use of f i l te red record sec t i ons . T h e noise is concen t ra ted in two b a n d s (F igure 2.2a), the 14 to 24 H z fre-quency b a n d and the 0 to 5 H z f requency b a n d . T h e r e is also noise present in the 5 to 12 H z f requency band bu t , as i l l us t ra ted in F i g u r e 2.2b, the s ignal con ta ins more power t h a n the noise for most cases. Howeve r , somet imes noise obscures any recorded seismic s igna l as shown in F i g u r e 2.3a for O B S 4 h y d r o p h o n e c o m p o n e n t . In th is ease filtering w i t h i n the seismic passband p roved he lp fu l . F i g u r e 2.3b shows the same d a t a (p lo t ted us ing the same scale factors) af ter bandpass filtering between 5 and 12 H z . A n event pa ra l l e l to the d is tance ax i s at a b o u t 1.8 s is now d i s t i ngu i shab le . A r r o w s h igh l i gh t i ng the event were pu rpose l y o m i t t e d to p e r m i t the reader to j udge the resu l ts w i t h o u t too m u c h bias f r om the a u t h o r . T h i s pa r t i cu l a r e x a m p l e proved to be a va luab le guide in the even tua l i n t e rp re ta t i on of the d a t a recorded on the ve r t i ca l and h o r i z o n t a l channe ls of O B S 4. 2.3.2 Interpretation procedures T h e record sect ions in A p p e n d i x II are in a f o rm ready for i n te rp re ta t i on . C o n -ven t i ona l i n te rp re ta t i on of re f rac t ion d a t a usua l l y invo lves a fo rward m o d e l in the fo rm of two -d imens iona l m o d e l l i n g schemes. T h e process begins by d e t e r m i n i n g a s ta r t i ng m o d e l , usual ly based u p o n a d i p p i n g layered m o d e l de r i ved f rom s t ra igh t l ine fits to PERIODOGRRH POUER o 0. 20 30 40 FREQUENCY (HZ) 9.000 8000 X 7000 6000 5000 | cr UJ '£ 4000 i 3000 2000 X 1000 10 PERIODOGRRM POUER "20 30 FREQUENCY (HZ) 40 50 60 Figure 2.2 Power spectra (a) background noise (b) signal + noise for OBS 3. Periodograms were computed two second window (see text for explanation). 30 O B S 4 H y d r o p h o n e O B S 4 H y d r o p h o n e Figure 2.3 Comparison of unfiltered (a) and filtered (b) record sections for OBS 4 hydrophone. Th< record sections are plotted with the same scale factors The filter used for (b) was a 5-12 Hz bandpas: filter. 31 f o rwa rd and reversed t rave l t imes of the p r i m a r y a r r i va l s . T h e s ta r t ing m o d e l may also be de r i ved f r o m m o d e l s ob ta i ned in o ther e x p e r i m e n t s near the area of i nves t iga t ion . T h e i n i t i a l m o d e l is then p e r t u r b e d by ad jus t i ng i t s parameters to ach ieve a closer fit be tween the c o m p u t e d t rave l t imes t h r o u g h the m o d e l , and the observed t rave l t imes , based u p o n p r i m a r y a r r i v a l events. T h i s p rocedu re is i nheren t l y non -un ique , bu t may be cons t ra ined by adherence to o the r i n f o rma t i on gathered f rom wel ls d r i l l ed near the a rea of s tudy , o ther se ismic d a t a a n d / o r mode ls a n d the regional geology of the area. A f u r the r re f inement to the final t rave l t ime mode ls m a y be a c c o m p l i s h e d by the mode l l i ng of the observed a m p l i t u d e s in the record sect ion . M o d e l s are very sens i t i ve to changes in a m p l i t u d e and care m u s t be taken when c o m p a r i n g mode l l ed a m p l i t u d e s w i t h observed a m p l i t u d e s where the d a t a have been comp i l ed f r o m m a n y di f ferent receivers a n d / o r sources. F u r t h e r m o r e , any p o s i t i o n i n g and t i m i n g cor rec t ions app l ied to t races, such as i n - l i ne co r rec t i ons , do no t i nc lude cor rec t ions for the amp l i t ude of the sh i f ted a r r i va l . T h e d a t a in the present s tudy were comp i l ed in a c o m m o n receiver f o rma t f rom a s tab le cons is ten t sou rce , w h i c h lends i tse l f we l l to a m p l i t u d e mode l l i ng . T h e in l ine cor-rec t ions app l i ed to the d a t a prec lude re l iable a m p l i t u d e m o d e l l i n g of a r r i va ls w i t h i n 4 k m of the O B S . T h e i n te rp re ta t i on of the d a t a in th is s tudy emp loyed a two -d imens iona l ray t r a c i n g rout ine deve loped at U B C (Spence , W h i t t a l l a n d C lowes , 1984). T h e c o m p u t e d t r ave l t imes m a y be c o m p a r e d w i t h those p icked f r o m the d a t a by d i r ec t l y over lay ing the c o m p u t e d values on to the observed da ta . T h e advan tage of d i rec t l y over lay ing the d a t a w i t h the c o m p u t e d t rave l t ime cu rve is tha t the c o m p u t e d resu l ts can be compared in the con tex t o f the ent i re record sec t i on . V a l u a b l e ins igh ts can resul t f r o m fo l lowing th i s p rocedu re as the in te rp re te r is cons tan t l y r e m i n d e d of other t rends w h i c h may ex is t i n the da ta . T h i s is not the case w h e n t rave l t imes are read in to the p r o g r a m since the in te rp re te r is t hen concerned w i t h fitting only an iso la ted set of po in ts . T h i s idea is 32 not new to refraction modelling, since in refraction interpretation, particularly for poor data, the interpreter will often return to the data after gaining a clearer insight into the structure under investigation and repick the first arrivals. To refine the model derived from the travel-time fit to the data, theoretical seis-mograms were calculated for the theoretical travel-time curves. Relative amplitudes across the synthetic record section were visually compared to the observed amplitudes. Amplitudes are particularly sensitive to velocity gradients and this feature provided additional control on the choice of velocity gradients for the model. Velocity gradients for the various layers in the model were altered until the amplitudes for the theoretical seismograms matched those in the data. Amplitude modelling also provided critical information for the analysis of particular phases observed in the data. An alternative method for determining a l-D velocity-depth structure directly from data sections was presented by Clayton and McMeehan (1981). The author followed their procedure and developed an interactive computer program to implement the method. The work represents a new development at UBC. It also provides examples additional to, and different from, those illustrated in the original paper and insight into the effects of applying a l-D interpretation to a data set which clearly represents a 2-D structure. Therefore, the method and its application to the airgun/OBS refraction data from Hecate Strait will be described in the next chapter. 33 C H A P T E R III INVERSION B Y WAVEFIELD CONTINUATION 3.1 Introduction Inversion by wavefield continuation is an image analysis procedure by which a 1-D veloeity-versus-depth structure can be extracted directly from a seismic record section. This is accomplished by projecting the observed wavefield, as represented by arrivals recorded at increasing offsets, back to zero offset. This backward projected wavefield then represents the arrival of wavefronts as a function of two-way travel time at the surface. This delay time function for the arrivals, along with the known stacking lines along which the backward projection was carried, may then be utilized to downward project the wavefield at the surface to the appropriate slowness medium from which it originated. The practical application of this method proceeds through two linear transformations; a slant stack (the back projection) and a downward continuation (the downward projected wavefield). 1 The inversion of refraction data by wavefield continuation has the desirable quality of an unbiased inversion, with the exception of the final velocity depth-pick. The observed data represent the input required by the program and the velocity depth function is obtained directly from the imaged wavefield. Inversion of refraction data by wavefield continuation has been applied to real data examples exhibiting weak lateral velocity variations (Clayton and McMechan, 1981). The theory is based upon the solution to the one dimensional wave equation for the 1-D velocity versus depth model. Clayton and McMechan (1981) demonstrated the robustness of the algorithm in the treatment of earth models with weak two dimensional velocity structures. 34 A c o m p u t e r p r o g r a m based u p o n the theory as presented by C l a y t o n a n d M c M e c h a n (1981) has been w r i t t e n w i t h two ob jec t i ves i n m i n d : ( l ) to test the app l i cab i l i t y of the p rocedu re for e x a m p l e s w i t h a greater degree of of l a te ra l he terogene i ty ; and (2) to a p p l y the p r o c e d u r e to a selected segment of the a i r g u n / O B S d a t a set for the purpose of c o m p a r i n g the resu l ts w i t h those ob ta ined f rom the 2 - D m o d e l l i n g in te rpre ta t ion in C h a p t e r I V . T h e d a t a under inves t iga t ion in th is thesis offer an idea l o p p o r t u n i t y to mee t these ob jec t i ves for two reasons . F i r s t , the d a t a set sat isf ies the basic assump t i ons o u t l i n e d in the d e v e l o p m e n t of the theory . Second ly , the d a t a have been fu l ly in terpreted us ing the U B C 2 - D ray t r a c i n g scheme (Spence , W h i t t a l l and C l o w e s , 1985). Therefore the resu l t i ng i nve r s i on , i nvo l v i ng c o m p a r i s o n of t heo re t i ca l and observed se ismograms, m a y be eva lua ted based u p o n these 2 - D resul ts . O t h e r i nve rs ion p rocedures w i l l not be invest igated as they are beyond the scope o f th is thesis. T h e reader is referred to var ious papers by Bessonova et al. (1974, 1976); G a r m a n y et al. (1979); W e n z e l et al. (1982); C a r r i o n et al. (1984) and m a n y o the rs for more t h o r o u g h d i scuss ions concern ing the invers ion of re f ract ion da ta . T h e fo l l ow ing d e v e l o p m e n t was unde r t aken as a separate s tudy of in terest , subsequent to resu l ts de r i ved f r o m the 2 -D m o d e l l i n g i n te rp re ta t i on desc r i bed in C h a p t e r I V . 3.2 The Linear Transformations 3.2.1 The slant stack procedure To ob ta in the s lant s tack , the a m p l i t u d e s are s u m m e d a long l ines of cons tan t s lope a n d in te rcep t ; the wavef ie ld is t hen cons t ruc ted by sweep ing t h r o u g h a l l s lopes a n d in te rcep ts on a se ismic record sec t ion . T h e s lope of the l ine in the t rave l t ime-o f fse t 35 ( t -x) d o m a i n is the inverse of the ve loc i t y (^) , a lso k n o w n as the ray parameter p; the in te rcep t is the two-way t rave l t ime at zero offset, usua l ly referred to as r. T h e a m p l i t u d e s u m m e d a long l ines of cons tan t s lope p and in te rcep t r is then stored or p l o t t ed at i ts new coo rd i na tes (p, T). H a v i n g desc r i bed how the s lan t s tack is p e r f o r m e d , i t may not be appa ren t how i t ar ises. A n u n d e r s t a n d i n g of the s lant s tack p rocedure can be a t t a i ned by cons ide r i ng the se i smogram presen ta t ion i tself . A func t ion P ( t , x ) , where P represents the a m p l i t u d e at the p o i n t ( t ,x) in the t — x d o m a i n can be de f ined. A s t ra igh t l ine th rough these p o i n t s , def ined by desc r ibes the t rave l t i m e for po in ts a long the l ine . If th is l ine in te rcep ts an a r r i v a l , t hen the t rave l t i m e to a receiver at an offset of x can be d e t e r m i n e d . F u r t h e r m o r e , th is a r r i va l a lso has a ve loc i t y in the f o r m of the s lope of the l ine th rough tha t po in t . H o w e v e r , th i s does not have m u c h s ign i f icance since a n u m b e r of l ines can pass t h rough th is a r r i v a l d e p e n d i n g u p o n the choice of in te rcep t and s lope . T h e s lan t s tack a l lows a we igh t to be a t t ached to the a r r i v a l to de te rm ine the s ign i f i cance of the assoc ia ted ve loc i t y . In prac t ice th is is done by rede f in ing P{t,x) by s u b s t i t u t i n g t i n (3.1) to ob ta in P { T i + P i x i x) a n d then a d d i n g al l the a m p l i t u d e values w h i c h fa l l a long the l ine def ined by Ti + p i x. T h i s we igh t i ng func t ion is def ined by t = Ti + pi x, (3.1) (3.2) z a n d for a set of a r r i va l s w i th the same appa ren t ve loc i t y as tha t of the slope of the l i ne , S w i l l a d d to large values whi le unco r re la ted events w i l l des t ruc t i ve ly inter fere. 36 W e i g h t i n g func t ions c a n be def ined for the ent i re se i smogram by v a r y i n g b o t h p and r. T h i s is the essence of the s lant s tack and i n i ts p r a c t i c a l f o rm is g iven by S{T,P) = J2P{T + PX,X). (3.3) z M o r e f o rma l l y s t a t e d , the s lant s tack is g iven by /+ oo P{T + px,x)dx. (3.4) - oo where S is the (p — r ) wavef ie ld . T h e s lant s tack is more easi ly ca r r ied ou t in the t ime d o m a i n when t race spac ing is not cons tan t ( M c M e c h a n and O t t o l i n i , 1980). A cor rec t ion for the f requency de-pendence of the stack mus t be m a d e i f wave fo rms are to be preserved . P h i n n e y et al. (1981) showed how the f requency dependence ar ises th rough the i r d e r i v a t i o n of the i n -verse s lant s tack . T h i s dependence c a n be i n tu i t i ve l y rea l i zed by c o n s i d e r i n g tha t the s u m m a t i o n o f a m p l i t u d e s a long d i f ferent po r t i ons of the wave le t , for m a n y wavelets , e f fect ive ly spreads the t rans fo rmed wave le t . H o w e v e r , since the invers ion scheme under cons ide ra t i on is on ly concerned w i t h the t a u cu rve desc r ibed by the locus i n the i — p d o m a i n , th is cor rec t ion o f the f o r m H(t) t~1'2 (Ph inney et al., 1981) need not be ap-p l i ed . F u r t h e r m o r e , the non- idea l 2 -D case requi res the s lant s tack and invers ion to use a l l va lues of p a long a pa r t i cu l a r r a y p a t h ( C h a p m a n , 1 9 8 1 ) w h i c h is beyond ' the scope of th is s tudy . T h e s lant s tack decomposes the observed se i smogram in to i ts fixed p c o m p o n e n t s w h i c h m a y then be d o w n w a r d con t i nued separa te ly . F i gu re 3.1a shows a theoret ica l se ismic sec t ion generated f r om the a s y m p t o t i c syn the t i c se i smog ram rou t ine o f Spence et 37 al. (1984) for a p lane layered ea r t h m o d e l w i t h four layers. T h e four a r r i va l branches can be seen i n the theore t i ca l s e i s m o g r a m . O n l y the re f rac ted a r r i va ls were cons idered; no p r e c r i t i c a l nor w ide ang le re f lec t ions were i n c l u d e d . Tab le 3.1 shows the charac ter is t i cs of t he 1-D m o d e l . LAYER DEPTH (km) VELOCITY (km/s) GRADIENT (km/s/km) I 0.0 2.0 O.S II 0.7 2.7 0.5 III 2.0 4.8 0.5 IV 4-0 6.0 O.S T a b l e 3 .1 1 -D m o d e l used for invers ion tes t ing. T h e a r r i va ls f r o m each layer have been labe l led w i t h the co r respond ing layer number i n F i g u r e 3.1a. T h e c o r r e s p o n d i n g s lant s tacked a r r i va l branches have been labe l led by lower case r o m a n numera l s . If the a r r i va l b ranches were represented by s t ra igh t l ines, each a r r i v a l b ranch w o u l d m a p to a po in t co r respond ing to the in te rcep t and slope of the s t a c k i n g l ine. Howeve r , ex tended l o c i of energy arise when ex tended wavelets are used a n d , as in th is case, w h e n the a r r i v a l b ranches possess cu rva tu re due to ve loc i ty g rad ien t s . T h e re la t i onsh ip be tween the b ranches and the i r t r ans fo rmed values can be eas i l y seen. T h e layer I a r r i va l b r a n c h t r a n s f o r m s to the p — T locus i wh i ch has i ts m a x i m u m energy concen t ra ted a t a po in t w h i c h represents the average ve loc i t y in the layer . S i m i l a r re la t i onsh ips can be seen for b ranches II and III . T h e last a r r i va l b ranch ( I V ) was i n c l u d e d to d e m o n s t r a t e the effects of s tack ing low a m p l i t u d e s w i t h apparen t ve loc i t i es close to a preced ing b r a n c h . T h e rays b o t t o m i n g in th is layer are few and a c o r r e s p o n d i n g l y lower a m p l i t u d e of the i r a r r i va l s is ind ica ted . T h e low grad ient and T R A V E L TIME CURVE (True Time) P - T A U C U R V E Distance (km) F igure S . l Slant stacked wavefield (b) for theoretical seismograms (a) computed for l - D model in Table 3.1 . Lower case roman numerals are used to show the corresponding p - T mapping for the equivalent arrivals identified by uppercase arrivals in (a). The slope for slant stack lines was incremented by 0.004 s/km and the intercept times were incremented by 0.0083 s or the reciprocal of the sampling rate (120sps). Stacking velocities ranged between 1.8 km/s and 6.5 km/s. 39 the far offset also affect the a m p l i t u d e of these a r r i va ls . W h e n these a m p l i t u d e s are s u m m e d , because no w i n d o w i n g was used, they are abso rbed essent ial ly by the effects f r o m b r a n c h III. T h u s , they are not c lear ly seen in the t rans fo rmed wavef ie ld . W h e r e they s h o u l d occu r has been i n d i c a t e d by i v . T h e y a lso have the effect of s l i gh t l y lower ing the a m p l i t u d e s for the t r a n s f o r m a t i o n of b ranch III. T h e ar t i fac ts seen as d iagona l , low a m p l i t u d e events r u n n i n g f r om left to r igh t resu l t f r om the f in i te aper tu re used , spat ia l a l i as ing a n d / o r the lack of w i n d o w i n g the s tack . 3.2.2 T h e d o w n w a r d c o n t i n u a t i o n p r o c e d u r e T h e d e v e l o p m e n t fo l lows t ha t of C l a y t o n a n d M c M e c h a n (1981), excep t they i m -p l e m e n t e d the d o w n w a r d c o n t i n u a t i o n in the f requency d o m a i n , bu t f o l l ow ing the i r r e c o m m e n d a t i o n , th is s tudy i m p l e m e n t s it in the t ime d o m a i n . T h e d o w n w a r d con t in -u a t i o n is s im i l i a r to a dep th m i g r a t i o n bu t is a p p l i e d in the offset d o m a i n as opposed to the c o m m o n m i d p o i n t d o m a i n . C l a e r b o u t (1976) and G a z d a g (1978) show tha t the downward con t i nua t i on of the wavef ie ld observed at the sur face can be i m p l e m e n t e d by a phase ro ta t i on in the fre-q u e n c y d o m a i n w h e n the ve loc i t y var ies on l y w i t h dep th . W r i t i n g the wave equat ion in the f requency d o m a i n gives dh2 d2 + 4 2 ]p (w,M) = o. T h e s o l u t i o n to th is equat ion was g iven by C lae rbou t (1976 ) and G a z d a g (1978) P[u,,kk,z) = P(w,fcfc ,0) e x p - i 2 z v2{z) 4 2 (3.5) 40 whe re u; is the t e m p o r a l f requency and kh is the h o r i z o n t a l wavenumbe r . T h e first m i n u s s ign i n equa t ion 3.5 i nd i ca tes tha t u p c o m i n g waves are be ing i m a g e d . T h e Fou r i e r C e n t r a l S l ice t h e o r e m ( C l a y t o n a n d M c M e c h a n , 1981) is invoked to recas t equa t i on 3.4 in the f requency d o m a i n w h i c h demons t ra tes , m o r e c lear ly , the r e l a t i o n s h i p be tween equa t ion 3.5 and the s lan t s tack . R e w r i t i n g equa t ion 3.4 S ( w , p ) = P{w,-2wp). (3.6) E q u a t i o n 3.5 m a y then be conver ted to i ts s lowness f o r m by subs t i t u t i ng — 2up for k^. R e w r i t i n g th is equat ion y ie lds F ( « , - 2 W p ^ ) = P ( W , - 2 w p , 0 ) e - i u * ( p ' z ) , (3.7) whe re * (p, z) = 2 / y/v-2(z) - p 2 dz. (3.8) U s i n g equa t ion 3.6, equa t i on 3.7 m a y be w r i t t e n as 5 ( u ; , p , 2 ) = 5 ( a ; , p , 0 ) e - ^ * ^ ^ (3.9) Inverse t r a n s f o r m i n g th is equat ion g ives S{r,p,z) = j S{oj1p,0)e-iu'^{p^'^dw. (3.10) T h i s equa t ion a l l ows the spec i f i ca t ion of the T — p wavef ie ld a t any dep th z. S ( a / , p , 0) is the s lan t s tacked wavef ie ld , ob ta ined in sec t ion 3.2.1, wh i ch represents the delay t ime 41 f unc t i on for u p c o m i n g waves at the surface. T h e e x p o n e n t i a l func t ion is the downward c o n t i n u a t i o n ope ra to r w h i c h is c o m m o n in po ten t i a l field ana lyses . F o r the p r o b l e m d e s c r i b e d , i t is not necessary to have the wavef ie ld for every dep th z, on l y those depths sa t i s f y ing the imag ing c o n d i t i o n for th is p r o b l e m mus t be met ; the d o w n w a r d con t i nua t i on process mus t s top w h e n a l l p lane wave c o m p o n e n t s have reached the i r m a x i m u m dep th of pene t ra t i on or b o t t o m i n g po in t . M a t h e m a t i c a l l y , th i s o c c u r s w h e n r = 0 ( the p of the ray equa ls the t rue slowness of the m e d i u m S e t t i n g T = 0 in equat ion 3.10 y ie lds E q u a t i o n 3.11 is in the f o r m u t i l i zed by C l a y t o n and M c M e c h a n (1981) in the i r a l g o r i t h m for i m p l e m e n t a t i o n . T h e i r p r o g r a m was run on an a r ray processor but they suggest recas t ing equa t ion 3.11 in the t i m e d o m a i n for genera l purpose mach ines . T h e p r o g r a m w r i t t e n for th is s t u d y uses the t i m e d o m a i n representa t ion of equat ion 3.11 w h i c h is g iven by E q u a t i o n 3.12 ar ises by t a k i n g the first m i n u s sign in equa t ion 3.10 ins ide the b racke ts , us ing the shi f t ru le ( C h a p m a n , 1978) and w r i t i ng equat ion 3.12 d i rec t l y . T h e doma in in w h i c h equa t ion 3.11 or 3.12 is app l i cab le has a b r a n c h cu t wh i ch is remed ied by a l te r ing the de f in i t ion of to (3.11) whe re s (p , z) is def ined as the s lowness p lane. s{p,z) = S[T - * ( p , z ) , p , 0 ] . (3.12) o (3.13) 42 which prevents attenuation of the wavefield below depths where p is greater than As recommended by Clayton and McMechan (1981), a phase shift of 57r/4 was then applied to the downward continued wavefield. This phase shift embodies three correction factors to compensate for (l) the far-field radiation condition (Aki and Richards, 1980, p 417); (2) the 2-D representation of wave propagation in three dimensions (Chapman, 1978); and (3) an average factor of 7r/2 to compensate for the range of reflection coef-ficients expected for reflections and refractions (Clayton and McMechan, 1981). This last point may cause confusion as reflection coefficients have been implied for refraction. The image formed by applying equation 3.11 is composed of wavelets whose shapes are defined by the phase shift associated with the reflection coefficient at each z (Clayton and McMechan, 1981). McMechan and Ottolini (1980) show that the p— r curve for a refraction branch is formed by the envelope of p— T curves for the reflections. Chapman (1978) also states that a refracted ray can be treated as having a reflection coefficient of —isgn{u>)>. The downward continuation procedure represented by equations 3.11 and 3.12, with the phase shift applied, was implemented on the Amdahl 5850 computer at the UBC Computing Center. Since the downward continuation equation requires a velocity versus depth curve as an input, it is necessarily an iterative procedure. The program has been designed to automatically input an initial velocity versus depth curve for the downward continuation of the p — T wavefield. The velocity versus depth curve used was 1.8 km/s for all depths. The downward continued wavefield is then displayed on the terminal. To operate the procedure in an interactive mode, a means of inputting the subsequent user picked velocity depth curves was required. The user obtained these new curves by picking points along the locus of maximum amplitude, now at minimum depth (z) in the downward continued wavefield or p — z wavefield. This was implemented using the cross 43 ha i rs on the g raph ics t e r m i n a l . T h e p r o g r a m i ns t ruc t s the compu te r to accept g raph ic i n p u t a n d the c o m p u t e r responds by d i s p l a y i n g cross ha i rs . T h e user then pos i t ions the cross ha i rs over the des i red a m p l i t u d e s and hi ts any key to i npu t the values. T h e p i ck ing session ends when the user selects a p va lue less t h a n 0.55 s / k m (this can be a l tered i f r e qu i r ed ) . T h e con t ro l then re tu rns to the p r o g r a m w h i c h conver ts the p icked p va lues to the ve loc i t i es and resamples t h e m at the desi red s a m p l i n g in te rva l t h rough a s imp le l i near i n t e r p o l a t i o n . T h i s new cu r ve is used to d o w n w a r d cont inue the or ig ina l p — T wavef ie ld a g a i n . T h u s , any new cu rve does not depend on the resul ts of the p rev ious c u r v e s ince the p r o g r a m a lways re tu rns to the o r ig ina l p — r wavef ie ld. T h e new p - z wavef ie ld is then d i sp layed and the process can be con t i nued . Usua l l y 4 to 5 i te ra t ions are necessary to ob ta in a, p — z wavef ie ld t h a t converges to a single so lu t i on . T h e advantage of i m p l e m e n t i n g the inve rs ion in an i n t e r a c t i v e m a n n e r is tha t the resu l tan t p — z cu r ve , for each d o w n w a r d c o n t i n u a t i o n i t e ra t i on , is immed ia te l y d i sp layed . Hence , the user can q u i c k l y ga in exper ience in i n t e rp re t i ng the p — z wavef ie lds. Changes observed by the user in a sho r t pe r i od can be be t te r ass im i l a t ed and an acceptab le so lu t ion is reached sooner . E a c h p-depth curve is o u t p u t to a file for later rev iew. 3 .3 Examples M c M e c h a n and O t t o l i n i (1980) and P h i n n e y et al. (1981) p rov ide an exce l lent d i scuss ion on the ana lys is of t r ans fo rmed record sect ions and the reader is referred to these pape rs . Howeve r , two speci f ic po in t s i n the a p p l i c a t i o n of the invers ion p rocedure shou ld be no ted . T h e resu l ts of the invers ion de te r io ra te for cases in wh ich (i) the d a t a are s p a t i a l l y a l i ased and (ii) where the wave fo rms are not phase corre lated or coherent 44 a long a s lan t s tack . S p a t i a l a l i as ing of the d a t a can be avo ided i f the source / rece ive r separa t ion is less t h a n one hal f of the wave leng th of the h ighest f requency one wishes to reso lve . 3.3.1 The plane-layered synthetic example T h e wavef ie ld in F i g u r e 3.2 was ob ta ined by d o w n w a r d con t i nu ing the p— T wavef ie ld of F i g u r e 3.1b. S u p e r i m p o s e d on the figure is the p— z f unc t i on used for the downward c o n t i n u a t i o n . T h i s inverse ve loc i t y dep th func t i on is also the one used to cons t ruc t the p lane - l aye red m o d e l f r om w h i c h the theore t i ca l se i smograms were der ived (F igure 3.1a). T h i s examp le i l l us t ra tes the re la t i onsh ip between the d o w n w a r d con t i nued wavef ie ld a n d the t rue ve loc i t y dep th func t i on . In th is e x a m p l e the p — z wavef ie ld images the t rue p -dep th cu rve bu t there are some m i n o r d i f ferences. T h e p — z cu rve (sol id l ine) between 0.5 and 1 0.455 s / k m does not co inc ide w i t h the large a m p l i t u d e event . T h i s cou ld be due to s p a t i a l a l ias ing and the effects of us ing an average phase shi f t of 7 r / 2 for the ref lect ion coef f ic ients (see Sec t i on 3.2.2). T h e agreement imp roves for the deeper va lues . T h e sma l l ve l oc i t y j u m p s at 4.0 k m d e p t h are d i f f icu l t to see bu t m a y be i n d i c a t e d by the s m a l l d o w n w a r d shi f t in the first a r r i va ls for the last two t races. In any event , i t wou ld be missed in a real d a t a case. To i l l us t ra te the i te ra t i ve na tu re of the p rocedure a fu l l r un us ing the p lane- layered syn the t i c se i smogram was pe r fo rmed and the resu l ts are dep i c ted in F i g u r e 3.3 a to e. T o ob ta in these resu l ts the theore t i ca l se ismic sect ion was read in to the p r o g r a m , s lant s tacked (decomposed i n to i ts fixed p componen t s ) and the s lant s tacked wavef ie ld was then d o w n w a r d con t i nued (each p separate ly) us ing a cons tan t ve loc i t y of 1.8 k m / s for the ze ro -o rde r i t e ra t i on for a l l dep ths . O n c e the d o w n w a r d con t inued p — z wavef ie ld 45 P - D E P T H C U R V E P (s/km) 0.60 0.55 0.50 0.45 0.40 0.35 0.30 0.25 0.20 0.15 0-0 I llrlllNlllliytu^AJIlj'llllllllllltlMllllllllll'lllllllllllllllliiillllii'iiiiiiiiiiiiliiii i 1.0 2.0 3.0 4.0 5.0 6.0 J Figure 3.2 Downward continuation of the p— r wavefield in figure 3.1b computed for the exact p-depth function (thick solid line). The relationship between wavefield and input p-depth function is illustrated for this best case in which the true p-depth function is known. Figure 3.3 The figure panels (a) to (e) on this and the following two pages depict the iterative series for the downward continuation of the p — T wavefield in figure 3.1b. Each panel shows the p — z wavefield computed from the p — r wavefield using the p — z function (thick solid line) selected from the previous downward continued (p — z) wavefield. An initial p — z wavefield was computed using a constant p or inverse velocity for all depths. This wavefield was then picked and the resulting p — z function (thick solid line in (a)) was used to compute the wavefield in (a). This wavefield was then picked (the dashed line in (a)) for the next iteration (b) where it is now plotted as a thick line. This process continues until convergence is obtained (see text for detailled explanation). Figure 3.3 c & d Downward continued p - r wavefield computed for p - z function from Iteration 2 (c) and Iteration 3 (d). P - D E P T H CURVE oo Figure S.S e Downward continued p - r wavefield computed for p — z function from Iteration 4. 49 was d i s p l a y e d on the t e r m i n a l , con t ro l re tu rned to the user. T h e user then proceeded to i n p u t new ve loc i t y versus d e p t h curves by p i ck i ng values as per sec t ion 3.2.2, un t i l convergence was o b t a i n e d . A s each new i te ra t ion is car r ied out , the so lu t i on osci l lates be tween h igh p a t low z va lues a n d low p at h igh z values (F i gu re 3.3a-e). Conve rgence can be recogn ized by two means . T h e first occur rs when a single p — z cu rve is ob ta ined for two or mo re i te ra t ions . T h e second occurs when two curves or success ive i t e ra t i ons have the p rope r t y such tha t i n p u t t i n g the h igh p, low z curve wi l l y i e l d the p rev ious low p, h igh z cu rve and vice versa. C l a y t o n and M c M e c h a n (1981) use th is p r o p e r t y to select a so lu t ion w h i c h is the average of these two curves . T h e two cu rves f o r m an enve lope w h i c h conta ins an o p t i m a l so lu t i on , bu t th is is not to be con fused w i t h u n c e r t a i n t y in the resu l t wh i ch is based on the w i d t h of the p — z image at convergence ( C l a y t o n and M c M e c h a n , 1981). T h i s is s im i la r to o ther invers ion schemes w h i c h a t t e m p t to def ine the enve lope of a l l poss ib le so lu t ions ( M c M e c h a n a n d W i g g i n s , 1972; B e s s o n o v a et al., 1974). A fu r ther ind ica t ion of convergence is the focuss ing of the a m p l i t u d e s a long the p—z image . A s convergence is a p p r o a c h e d , th is image 'b r igh tens ' . F i g u r e 3.3a has been l abe l l ed i t e ra t i on 1 and is the resul t ob ta ined by p i ck ing the image in the i n i t i a l p — z wavef ie ld for the cons tan t ve loc i t y dep th cu rve . Therefore , i t e ra t i on 1 refers to tha t i t e ra t i on w h i c h fo l lows f r om the first user p icked p—z cu rve , i n d i c a t e d by the so l id l ine. T h e dashed l ine in F i g u r e 3.3a is the curve p icked for th is d i s p l a y of the wavef ie ld . T h i s convent ion is fo l lowed for a l l the examp les g iven; the dashed l ine represents the present p ick whi le the sol id l ine represents the prev ious p ick . In the p i c k i n g of the p—z images for th is and the o ther examp les , the po in t co r respond ing to the m a x i m u m a m p l i t u d e at m i n i m u m z was se lec ted. A l s o , a sma l l n u m b e r of po in ts were p i cked and the p r o g r a m l i nea r l y i n te rpo la ted between t h e m to give a resu l t s im i l a r to t h a t used to def ine the m o d e l ( layers w i t h l inear ve loc i t y g rad ien ts ) . 50 In Figure 3.3a the basis for picking the p-z curve for the first 0.7 km in depth is clear. At this depth the amplitudes of this first locus continue but it must be remembered that we are looking for the 1-D v — z curve. Thus, the point chosen was the maximum peak at the next monotonically lower z value. This is at a value of 0.370 s/km. The next point chosen was at the lower end of this second major locus, but such that it was not lower than the amplitude for the wavelets in the third major locus. This is the general procedure used for the entire picking process for all the examples given in this chapter. The results of picking the dashed curve in Figure 3.3a are shown in Figure 3.3b. The p — z image now mostly appears above that of the previous pick (the solid line, Figure 3.3b) for depths below 1 km. This process continued for Figure 3.3 b to e. The final panel (Figure 3.3e) was not picked as it represents the final iteration based upon the previous pick. This is how the interactive session would normally terminate since the result of any pick is applied before control returns to the user. During the picking process, a possible guide to jumping the low amplitude regions from one major locus to the next are the "knees" which form at the base of each linear gradient (see Figure 3.3c at z = 0.6 and p — 0.45). The p — z curves from the last two iterations (the solid lines in Figure 3.3d and e) are shown in Figure 3.4 along with the correct velocity depth function. The lack of agreement in the upper 1.0 km of depth is probably due to spatial aliasing of the data. For these depths and velocities, a trace spacing of 110 meters would be required, but for this example a trace spacing of 200 m was used. Between a depth of 1 and 2 km, agreement between the correct curve and the downward continued wavefield is very good. Below this depth the last two iterations form bounds on the correct solution. The resolution of the graphics video display device used in the interactive session also contributed to the poorer results for the velocity 51 P - D E P T H C U R V E Figure 3.4 Comparison of p — z functions from figure 3.3d (line 2) and e (line 3) with exact function (line 1) from 1-D model. 52 va lues p icked at dep ths greater t h a n 2.0 k m . F u r t h e r m o r e , as noted in sect ion 3.2.1, the rays d i d not dense ly samp le the layer and th is con t r i bu tes to the poo r l y de te rm ined values at deep z. 3 . 3 . 2 T h e 2 - D s y n t h e t i c e x a m p l e T h e 2 - D syn the t i c examp le is f r o m the mode l l i ng resul ts for the O B S 3 to O B S 2 reco rd sec t ion in chap te r I V . I nve r t i ng the 2 - D syn the t i c examp le served as an i ns t ruc -t i ve gu ide pr ior to i nve r t i ng the rea l d a t a examp le . K n o w i n g the ac tua l s t ruc tu re of the m o d e l f r o m w h i c h the se ismograms were c a l c u l a t e d a l lowed for c o m p a r i s o n s between the two m e t h o d s , p a r t i c u l a r l y w i t h respect to the effect of a p p l y i n g th is invers ion to a 2 - D case. T h e syn the t i c s e i s m o g r a m sec t i on , genera ted by the asympto t i c ray theory m e t h o d as desc r ibed by S p e n c e et al. (1984), is shown in F i g u r e 3.5a. T h e s lan t s tacked wavef ie ld is i l l u s t r a t e d in F igu re 3.5b. T h e t r ans fo rmed branches c o r r e s p o n d i n g to the t rave l t i m e curves have been labe l led as in Sec t i on 3.3.1. T h e ex tended nature of the t r ans fo rmed b ranches is due to the cu rva tu re of the t rave l t ime b r a n c h wh i ch resul ts f r o m the ve loc i t y g rad ien ts in the m o d e l . T h e t rave l - t ime branches I I a n d I I I appear to be a s ing le b r a n c h bu t a s l igh t change can be observed at a m o d e l d i s tance of 6 . 0km. S i m i l i a r l y for the t r ans fo rmed b ranches , a s l ight change can be seen between i i and i i i . A r r i v a l b ranch I V t r ans fo rms to the c lear , bu t low a m p l i t u d e , p — r b ranch i v . T h e m u l t i p l e event V t r ans fo rms to the p — T event v appea r i ng lower in the t rans fo rmed wavef ie ld and therefore does not in ter fere w i t h the p r i m a r y p-r event . B y s lant s tack ing the a m p l i t u d e s w i t h i n a speci f ied w i n d o w the des t ruc t i ve inter ference effects f rom the ex tended wave le ts of events ou ts ide th is w i n d o w c o u l d be d i m i n i s h e d . T h i s wou ld have TRAVEL TIME C U R V E (True Time) P - T A U CURVE Distance (km) Figure 3.5 Slant stacked wavefield (b) for theoretical seismograms (a) computed for 2-D velocity sub-model of Chapter IV, section 4.3.2 (see figure 3.2 for explanation of symbols). 54 the des i rab le effect of i nc reas ing the range of larger a m p l i t u d e s for each t r a n s f o r m e d b r a n c h . A n examp le wou ld be the p — r b ranch i v . T h e a m p l i t u d e s at the uppe r end of the b r a n c h c o u l d have been depressed by the in ter ference f r om the a m p l i t u d e in the wave le t for t r ave l - t ime b r a n c h III. For the d o w n w a r d c o n t i n u a t i o n of the p — r wavef ie ld shown i n F i g u r e 3.5b, the same p rocedu re as ou t l ined in sec t ion 3.3.2 was fo l lowed. Fo r each i t e ra t i on , the m a x i m u m a m p l i t u d e at m i n i m u m z was p i cked f r om the p — z wavef ie ld and used for the nex t i t e ra t i on . T h e final three i t e ra t i ons are d i sp layed in F i g u r e s 3.6 a to c. T h e so l id l ine in F i g u r e 3.6a was the i n p u t p — z p ick used to d o w n w a r d con t inue the wavef ie ld p resented in th is figure. T h e dashed l ine represents the p ick made f r o m th is wavef ie ld in the i n te rac t i ve session for the nex t i t e ra t ion s h o w n in F i g u r e 3.6b (it now appears in th is figure as the sol id l i ne ) . A s im i l a r desc r ip t i on app l ies to F i g u r e 3.6e. A dec is ion was then made to choose w h i c h of these resul ts represented convergence. T h e dashed l ine in F i g u r e 3.6b was chosen as the first i nd i ca to r . T h i s choice was made because the a m p l i t u d e s inc reased s ign i f i can t l y between the wavef ie lds shown in F i g u r e 3.6a and b. T h e s m a l l ve loc i t y j u m p , w h i c h recurs a t a dep th of 3.8 k m was based u p o n the weak a m p l i t u d e s e x t e n d i n g h o r i z o n t a l l y , c o m b i n e d w i t h the severe decrease in a m p l i t u d e observed af ter the t race i nd i ca ted by the a r row. A n a l y s i s of these a m p l i t u d e s is on ly made poss ib le by the noise-free d a t a . T h e second cu rve p i cked (dashed l ine, F i g u r e 3.6b) was chosen as the second and final i n d i c a t o r for convergence. It is shown as the so l id l ine in F i g u r e 3.6c. In genera l , the a m p l i t u d e s are weaker t h a n those seen in the wavef ie lds for F i g u r e 3.6a and b. B a s e d upon these two observa t ions , the sol id l ines of F i g u r e 3.6 b and c (or converse ly the dashed l ines of F i g u r e 3.6 a and b) were chosen as the two curves def in ing convergence. P-DEPTH CURVE P-DEPTH CURVE Figure 3.6 The figure panels (a) to (c) depict the final three iterations for the downward continuation of the p -wavefield in figure 3.5b . (see figure 3.3 and text for explanation) / r P-DEPTH CURVE 57 Five p-depth functions were selected from the 2-D model, where it was sampled by the rays, and averaged. The average p-depth function and those for the last two iterations are shown in Figure 3.7. Comparing the three curves in this figure shows how the downward continued wavefield solutions (curves 2 and 3) appear to have averaged the velocities across the model. This result can be described as a l-D kinematic equivalent for the actual 2-D velocity structure. The minor discrepancies observed can be attributed to the effects of not windowing the stack, spatial aliasing, the resolution of the graphics terminal at high p, low z values, and the 2-D nature of the model itself. A plane-layered model constructed from an average of these curves could easily be perturbed to fit the 2-D model. We might then expect that for the real data example the imaged result will represent a l-D kinematic equivalent for the 2-D earth model. As the subsurface structure more closely approaches a laterally homogenous model, the result will begin to reflect a more accurate representation of the subsurface velocity distribution. 3.3.3 A real data example In this last example, observed data, the interpretation of which led to the theoretical data used in section 3.3.2, were used in the wavefield transformation procedure. The record section and the slant stacked record section are shown in Figure 3.8a and 3.8b. The primary arrivals transform to the maximum amplitude locus at minimum T. The large amplitude event, lower amplitude event and the multiple event are labelled as in previous diagrams. The larger amplitude artifacts appearing above the primary p — T event in Figure 3.8b may in part be due to transformed precritical reflections not clearly evident in the seismogram. Clayton and McMechan (1981) observed similiar 58 P - D E P T H C U R V E P (s/km) 0.60 0.55 0.50 0.45 0.40 0.35 0.30 0.25 0.20 0.15 6.0 J Figure S.7 Comparison of p - z functions from figure 3.3b (line 2) and (c) (line 3) with average p — z function for 2-D model (line 1). The average p-z function was calculated for five p-z curves selected along the 2-D sub-model in figure 4.11b where it is sampled by simple refracted ray paths. T R A V E L TIME CURVE (True Time) P - T A U C U R V E P (s/km) Distance (km) Figure 3.8 Slant stacked wavefield (b) for OBS 3 reverse profile (a) (see section 4.3.2) plotted from east to west, (see figure 3.1 and text for explanation of figure) 60 features for the i r e x a m p l e s . T h e reverbera to ry na tu re of the se ismic s igna l con t r ibu tes to r eve rba t i ons in the p — T sec t i on , bu t be low the t r ans fo rmed p r i m a r y ar r iva ls . T h e d a t a gaps do not a p p e a r to have had an adverse effect on the s lant s tack . T h e i r m a i n effect w o u l d p robab l y be r e d u c i n g the po ten t i a l a m p l i t u d e of a r r i va ls w h i c h fel l a long the s lan t s tack l ine . T h e d o w n w a r d con t i nua t i on steps are dep i c ted in F i g u r e 3 .9a-e. T h e in te rp re ta t ion of the rea l d a t a set presents some new p rob lems . T h e recogni t ion of the t rue t rans fo rmed p r i m a r y is some t imes confused by the noise in the d a t a . Howeve r , one c lear gain in the rea l d a t a case is in t e rms of the f ocus i ng effect on the d o w n w a r d con t inued wavef ie ld . It is m u c h more p r o n o u n c e d t h a n in the syn the t i c e x a m p l e and th is helps to coun te rac t the adverse noise effect. In F i g u r e 3.9a, the sol id l ine shows the peak f r o m the zero-o rder (cons tan t ve loc i t y ) d o w n w a r d con t i nued wave f ie ld . A t a dep th of 2.0 k m the p d e p t h cu rve was ex tended to fill the d a t a space. T h i s is s imp l y a requ i remen t of the r e s a m p l i n g rou t ine in the p r o g r a m . T h e dashed l ine shows the p ick for the use in the nex t i t e ra t ion . T h e p icks are based u p o n the m a x i m u m a m p l i t u d e at m i n i m u m z. W h e r e the a m p l i t u d e sh i f ted downward i n z for an ad jacen t t race , a po in t was selected at the top a n d the b o t t o m of th is j u m p (see the a r row in F i g u r e 3.9a). O t h e r w i s e , the same procedure as descr ibed in the p rev ious sect ion was fo l lowed. T h e b l o c k y s t ra igh t l ine app roach was used where poss ib le . T h a t is, not every peak was p i c k e d , nor need be, but key po in t s in the curve were p i cked and the p rog ram i n t e rpo la ted between these po in ts . T h e process con t inued un t i l the p—z wavef ie ld was deemed to have converged . I te ra t ions 4 a n d 5 (F igure 3.9 d and e) were ident i f ied as the two convergent ind ica to rs . In th is e x a m p l e the cu rve , p i cked f r o m the wavef ie ld in F i gu re 3.9d, ac tua l l y fits the d o w n w a r d con t inued wavef ie ld o f F i g u r e 3.9e. T h i s resul t figured s t rong ly in the decis ion to t e rm ina te the i te ra t i ve Figure 3.9 Figure panels (a) to (e) depict the iterative series for the downward continuation of the p - r wavefield in figure 3.8b.(see figure 3.3 and text for explanation) i P-DEPTH CURVE c P (s/km) 0.80 0.55 0.50 0.45 0.40 0.35 0.30 0.25 0.20 0.15 Figure 3 .9 c ic d downward continued p — r P-DEPTH CURVE d P (s/km) 0.60 0.55 0.50 0.45 0.40 0.35 0.30 0.25 0.20 0.15 wavefield for Iteration 3 (c) and Iteration 4 (d). / P-DEPTH CURVE 64 process . T h e f ina l two resu l ts ( the so l id l ines f r o m F igures 3.9 d and e) are shown in F i g u r e 3.10 a l ong w i t h the same p-depth cu rve f r o m the prev ious sect ion (see F i g u r e 3.7). T h e ve loc i t y g rad ien ts a n d ve loc i t y increases for the invers ion curves m a t c h those for the averaged ve loc i t y -dep th cu rve f r o m the 2 - D mode l qu i te we l l . In general the i nve rs ion gave p—z values tha t were lower t han the p — z cu rve for the 2 - D m o d e l l i n g . T h i s cou ld be a t t r i b u t e d to the p rev ious l y o u t l i n e d p rob lems a n d for th is ease, p rob lems assoc ia ted w i t h the accura te iden t i f i ca t ion of the m a x i m u m event and m i n i m u m dep th for the t r a n s f o r m e d p r i m a r y a r r i va l event . No ise m a y have con t r i bu ted to i nco r rec t l y i d e n t i f y i n g the lower a m p l i t u d e p - z events d u r i n g the p i ck ing process. Fo r dep ths be low 2.0 k m , the s t rong la te ra l he terogene i ty of the 2 - D m o d e l wou ld ind ica te tha t th is fac to r has the largest effect on the f ina l resu l t (see F i g u r e 4 .11b) . 3.3.4 Summary T h e inve rs ion p rocedure y ie lded the cor rec t ve loc i t y dep th curve for the p lane lay-ered ease. T h e resu l t was ob ta i ned q u i c k l y a n d in an unb iased fash ion . W h e n the subsur face ear th m o d e l d iverges f r o m a p lane layered case the solut ion ob ta ined f r o m the wavef ie ld con t i nua t i on appears to represent a l - D k inemat i c equ iva lent for the 2 - D ve loc i t y m o d e l . E f fec ts of no ise, a t least at low leve ls , and reverbera t ions in the d a t a do not a p p e a r to degrade the resu l ts for the e x a m p l e cons idered . T h i s is p r inc ipa l l y because the first a r r i va l a m p l i t u d e s in the t— x d o m a i n t r a n s f o r m in to amp l i t udes w h i c h have m i n i m u m p and z values. W h a t are the imp l i ca t i ons of these findings? The re appear to be severa l . A s shown by C l a y t o n a n d M c M e c h a n (1981), the i nve rs ion of se ismograms recorded over areas k n o w n to have a near ly p lane- layered ve loc i t y s t ruc tu re y ie lds va l id resul ts . A second 65 P - D E P T H C U R V E 6.0-1 Figure 3.10 Compasrison of p — z functions from figure 3.9d (line 2) and (e) (line 3) with average p function from 2-D model as per figure 3.7 . — z 66 resu l t f r o m th is s tudy is t ha t the m e t h o d can be app l ied to ve loc i ty s t ruc tu res w i t h an even h igher degree of l a te ra l va r ia t i on i f an a p p r o x i m a t e func t ion is the ob jec t i ve . In o the r w o r d s , i f on ly an i n i t i a l m o d e l is des i red for fu r the r p e r t u r b a t i o n by a 2 - D m o d e l l i n g scheme, th is m e t h o d c a n p rov ide a qu ick a n d eff icient means of o b t a i n i n g the s ta r t i ng m o d e l . A d i p p i n g l aye red m o d e l m a y be ob ta i ned by i nve r t i ng b o t h the f o rwa rd a n d reverse prof i les. T h e dens i t y of rays s a m p l i n g the subsur face at e i ther end of the prof i le w i l l be we igh ted for t h a t e n d . If there is s t r uc tu re , then the resu l ts f r om a f o r w a r d a n d reverse prof i le invers ion w i l l be s l igh t l y di f ferent and cou ld be used as the basis for an i n i t i a l 2 -D m o d e l A n i m p o r t a n t obse rva t i on is t h a t ex t reme cau t ion is adv i sed in the app l i ca t i on of th is i nve rs ion m e t h o d to a d a t a set recorded over an area where l i t t l e or no th ing is k n o w n a b o u t the subsur face . C l a y t o n and M c M e c h a n (1981) inver ted a rea l d a t a e x a m p l e as a demons t ra t i on of the robustness of the m e t h o d for cases where weak la te ra l ve loc i t y var ia t ions ex is t . T h e d a t a i n d i c a t e d only weak loca l ve loc i t y effects a n d an in te rp re ta t ion f r o m a nea rby prof i le s u p p o r t e d the i r resul ts . If the theoret ica l s e i s m o g r a m s of F igu re 3.5a are cons ide red , the p r i m a r y a r r i va ls are observed to increase s m o o t h l y a long the t ra jec to ry desc r i bed by these a r r i va ls . Y e t the m o d e l (see F i g u r e 4.11b) f r o m w h i c h these s e i s m o g r a m s were generated is 2 - D . These observa t ions suggest t ha t the i n d i s c r i m i n a n t a p p l i c a t i o n of a 1 -D invers ion to d a t a j udged by the i r own mer i t as rep resen t ing a r r i va ls f r om a p lane layered ear th wou ld lead to an er roneous so lu t i on . It is c lear t ha t the real d a t a set i nver ted in th is chap te r i nd i ca ted no severe effects due to a s t rong ly t w o - d i m e n s i o n a l ve loc i t y s t r uc tu re . Y e t subsequent resul ts f rom the 2 - D m o d e l l i n g i nd i ca te o therw ise . A l t h o u g h the mode l l i ng process is non -un ique , i t is c o n s t r a i n e d by fo rward and reverse profi les and suppo r ted by resul ts f r o m prev ious 67 s tud ies ( S h o u l d i c e , 1971, 1973; and S tacey and Stephens,1969) as desc r ibed in C h a p t e r I V . B a s e d on the p r o g r a m w r i t t e n for th is s tudy , and i t s app l i ca t i on to syn the t i c and rea l d a t a , severa l r e c o m m e n d a t i o n s c a n be made . Fo r the s lant s tack ope ra t i on , the s tack s h o u l d be w i n d o w e d to reduce the effects f rom a r r i va ls not assoc ia ted w i t h the s t a c k i n g ve loc i t y . Second ly , for the a i r gun s igna tu re w h i c h has a s t ronger amp l i t ude in the second cyc le of the a r r i v a l , the p r o g r a m cou ld be des igned to use th is a m p l i t u d e for the i nve rs ion . T h i s cou ld be a c c o m p l i s h e d by a p p l y i n g a phase shi f t to move th is peak to the r igh t d e p t h . C l a y t o n and M c M e c h a n (1981) do someth ing s im i la r w h e n they use m u l t i p l e s to ex t rac t the ve loc i t y vs d e p t h cu rve . T h e y accomp l i sh th is by us ing the fac t t ha t the mu l t i p les have tw ice the r of c o r r e s p o n d i n g p r imar i es and i m p l e m e n t the a p p r o p r i a t e phase ro ta t i on by d o u b l i n g the f requency. T h i s a l lows the mu l t i p l es to be used to ex t rac t the p -depth cu rve . The re fo re , a phase shif t of poss ib ly a quar ter of a wave leng th may a l low the user to u t i l i ze the s t rong m a x i m u m in the a i rgun s ignature to e x t r a c t t he p -depth cu rve . F i l t e r e d d a t a and p — T sect ions were tested us ing a zero-phase bandpass f i l ter, bu t the side lobes in ter fer red w i t h the recogn i t i on of the first m a x i m u m a m p l i t u d e s at m i n i m u m z in the d o w n w a r d con t i nued wavef ie ld . T h e zero phase charac te r i s t i c of the filter m a y have been the p r o b l e m . Fu r t he r invest igat ion is r e c o m m e n d e d . 68 C H A P T E R IV M O D E L L I N G OF T H E A I R G U N - O B S DATA 4.1 Introduction T h e a i r g u n - O B S d a t a were m o d e l l e d in th ree segments us ing the 2 - D ray t race syn the t i c se i smogram a l g o r i t h m d e s c r i b e d by S p e n c e et al. (1984). T h e segments were de f ined by the d a t a sets be tween O B S 1 a n d O B S 2, O B S 2 and O B S 3 a n d O B S 3 a n d O B S 4. T h e f o rwa rd and reversed prof i les were mode l led s imu l t aneous l y for each segment of the a i r gun l ine. N o n e of the se ismic record sect ions presented i n th is chapter have been filtered un less o the rw ise i n d i c a t e d . A r 1 / 2 geometr ica l sp read ing fac to r , where r is the offset, has been app l i ed to b o t h the t heo re t i ca l and observed a m p l i t u d e s for each s e i s m o g r a m . T h e a m p l i t u d e s on the observed se i smog rams were oversuppressed by i ts a p p l i c a t i o n for se i smog rams w i t h i n 3.0 k m of the receiver . T h i s occurs because the a c t u a l rece iver was never d i rec t l y be low the source whereas th is must be the case for a 2 - D m o d e l . H o w e v e r , th i s ove rsuppress ion d i m i n i s h e s beyond the 3.0 k m sou rce -rece iver d is tance . T h e t r ave l t ime fits shou ld be re l iab le after a 1.0 k m source- rece iver d i s t a n c e . T h e m o d e l l i n g for each segment w i l l be d i scussed separate ly for the fo rward and reverse prof i les. T h i s w i l l be fo l lowed by a s u m m a r y of the m o d e l tha t was deve loped for t ha t segment p lus a br ie f geolog ica l i n t e r p r e t a t i o n . T h r o u g h o u t the fo l l ow ing d iscuss ion on ly the ver t i ca l c o m p o n e n t for each d a t a set is p resented. However the h o r i z o n t a l and h y d r o p h o n e c o m p o n e n t s were used when they p rov ided cons t ra in ts for ques t ionab le events on the ve r t i ca l c o m p o n e n t s e i s m o g r a m . A p p e n d i x II i nc ludes record sect ions for a l l c o m p o n e n t s . T h e final ve loc i t y m o d e l is p resented in F i g u r e 4.1 for reference d u r i n g the fo l l ow ing d i scuss ions . W h e r e va lues for ve loc i t i es a n d / o r g rad ients are uncer ta in OBS 1 0 OBS 2 OBS 3 2oo (0.25) ^ D is tance (km) f 20 25 30 35 40 10 15 1.80 (0.25)' 2.35 (0.25) OBS 4 1.49 (0.001) f 45 1 s Q .90 (0.25 7.20 (0.30) F igu re 4.1 Final velocity model for airgun/OBS survey. Velocities for each polygonal block apply to the upper boundary or the uppermost boundary segment when indicated. Velocity gradients for the blocks are enclosed in brackets and increase downwards along a line perpendicular to the upper boundary. The location of the OBSs with respect to the final composite model are indicated at the top of the model. The model is plotted with a vertical exageration of 5:1 . Uncertainties in the values derived from the model are indicated by (*) while undefined regions of the model are marked by (?). — 70 have been i n d i c a t e d by an aster isk . Q u e s t i o n m a r k s are used to ind ica te regions where l i t t le con t ro l ex is ts for the ve loc i t ies and grad ien ts used . 4.2 Initial Constraints T h e i sopach for the Ter t i a ry S k o n u n sed imen ts (Shou ld iee 1971, 1973) der ived f r om i ndus t r y re f lec t ion a n d re f rac t ion se ismic d a t a a long w i t h wel ls d r i l l ed in the Queen C h a r l o t t e bas in p rov i ded cons t ra in t s for the m o d e l . T h e sonic logs f r om these wel ls p rov ided the necessary jus t i f i ca t ion for the choice of large ve loc i ty g rad ien ts . W h e r e d a t a p rocess ing affected the near offset t races (see C h a p t e r II) such tha t in terpre ted ve loc i t ies were no t re l iab le , sonic logs f r o m wel ls near the a i r gun l ine were used to cons t ra in the ve loc i t y for the uppe r sed imen ta ry layers . A g rav i t y survey (Stacey a n d S t e p h e n s , 1969), near ly co inc iden t w i t h the a i rgun l ine , also p rov ided j us t i f i ca t i on for the gross s t r u c t u r a l charac te r i s t i cs o f the m o d e l observed at i ts eastern e n d . T h e in i t i a l m o d e l was cons t ruc ted by pro jec t ing the pos i t i on o f the O B S s on to the m o d e l presented by Shou ld i ee (1971) and us ing the ve loc i t i es f r om the wel ls c o m b i n e d w i th the ve loc i t ies de r i ved f r o m cons ide r ing the fo rward and reverse prof i les. 4.3 O B S 1— O B S 2 Submodel F o r the purposes of th is and the fo l l ow ing sec t ions , the Forward profile refers to the record sect ion where the source- rece iver d is tance increases f rom west to east. T h e Reverse profile refers to source- rece iver d i s tances t ha t increase f rom east to west. T h e re la t i ve pos i t i on of the O B S that recorded the se i smogram is i nd i ca ted by the insert in e i ther the uppe r left corner of the record sec t ion , as for the Forward profile or the uppe r r i gh t hand corner , as for the Reverse profile. A n y d is tances quoted in the fo l low ing 71 discussions will be indicated by the qualifiers MD for model distances and SRD for source-receiver distances. The theoretical travel time curves have been superimposed as solid curves on the observed seismograms and are labelled by lower case letters. The model through which rays are traced appears directly below the seismogram. The ray groups reach the surface below their corresponding travel time curves on the seismogram. In a following figure the data are also presented above the theoretical seismograms derived from the model. In this case, the travel time curves have been omitted from the real data display so as not to detract from the amplitude comparison between theoretical and observed seismograms. 4.3.1 Forward profile The record section for the forward profile is shown in Figure 4.2A. Figure 4.2B shows the ray traced model for the OBS 1 to OBS 2 segment, see Figure 4.1 and Figure 2.1. The analog data were digitized for only the near offset arrivals (first 7.0 km SRD ) as the noise level completely obscured the seismic signal beyond this range. The high noise levels for OBS 1 may be attributed at least partly to the shallow depth of this particular OBS. The model is separated into two basic units, the lower velocity sedimentary unit and the higher velocity crystalline unit although the latter is not defined by the forward profile but by the reversed one (section 4.3.2). The lower velocity sedimentary unit is further divided into three low velocity layers. The energy from the upper layer was modelled by refractions through a layer with a velocity of 1.8 km/s. The theoretical travel time branch, a, matches the observed arrivals but the sonic logs from nearby 4 6 8 10 12 14 16 18 Source/Receiver Distance (km) Model Distance (km) 20 Bi o 2 -ffi 4 cu Ed 8 r Figure 4.2 The data for OBS 1 (A) are displayed above the ray tracing diagram for the OBS 1—OBS 2 sub-model (B). The source/receiver distance SRD is plotted along the horizontal axis for the data. Model distance MD is plotted along the horizontal axis for the ray tracing diagram. The data are plotted in reduced time format such that arrivals with an apparent velocity of 6.0 km/s appear horizontal. The travel time curves, labelled a to c, superimposed on the data were computed from the corresponding ray groups appearing below each curve. The relative position of the OBS is indicated by the label in the upper part of the record section (see text for discussion). 73 wel ls i nd i ca te a ve loc i t y of 2.0 k m / s for the uppe r sed imen ts . T h e d i sc repancy may be a t t r i b u t e d to the p rocess ing of the d a t a . G o o d f ixes on the pos i t ion of the O B S s imp rove where sp l i t sp read i n f o r m a t i o n is ava i l ab le . F u r t h e r m o r e the near offset t races are more s t rong l y af fected by res idua l 3 - D effects assoc ia ted w i t h the i r t rue offl ine pos i t ion af ter the 2 - D pro jec t ion (see C h a p t e r II). Re f rac t i ons th rough the 2.35 k m / s layer in the m o d e l are not seen to emerge as p r i m a r y a r r i va l s bu t the p rec r i t i ca l ref lect ions f r o m the top of the 2.75 k m / s layer , cu rve b, do m a t c h the large seconda ry a r r i va ls seen between 1.0 and 1.7 k m SRD. T h e re f rac t ions t h r o u g h the 2.75 k m / s layer , b ranch c , emerge as the f irst set of p r i m a r y a r r i va l s seen af ter b ranch a . T h e th inness of the 2.35 k m / s layer , less t han 100 meters , p r o b a b l y a c c o u n t s for th is . T h i s layer m a y or m a y not be present at th is end of the s u b - m o d e l bu t is observed in the o the r s u b - m o d e l s and therefore has been inco rpo ra ted i n to the i n te rp re ta t i on . S ince the accu racy of the resul ts a t near offset are affected by the d a t a p rocess ing , e l i m i n a t i o n of th is layer was not jus t i f i ed . T h e d a t a are noisy bu t a b a n d p a s s filtered ( 5 -12 H z ) reco rd sect ion p lus the g o o d response on the ho r i zon ta l geophone , a ided in the d e t e r m i n a t i o n of the first b reaks . T h e theo re t i ca l se i smog rams are c o m p a r e d w i t h the observed d a t a in F i g u r e 4.3. A g rad ien t of 0.50 k m / s / k m in the 2.75 k m / s layer was chosen to g ive a good re la t ive m a t c h between the observed a n d the syn the t i c a m p l i t u d e s . For the Queen C h a r l o t t e bas in , g rad ien ts in excess of 1.0 k m / s / k m are obse rved f rom the sonic wel l logs. 4 . 3 . 2 R e v e r s e p r o f i l e T h e reversed prof i le shown in F i gu re 4 .4a recorded se ismic energy out to 17 k m SRD. O n l y th is prof i le ex tended across m o s t of the range between O B S 2 to O B S 1. 0 2 4 6 8 10 12 14 16 18 20 Source/Receiver Distance (km) Model Distance (km) Figure 4.3 Data for OBS 1 (A) is compared with the synthetic seismograms (B) computed for the ray tracing diagram in figure 4.2B. Both the data and the synthetic seismograms are plotted with the same scale factors. The MD plotted along the horizontal axis of the synthetic seismograms corresponds to those of the ray tracing diagram. 75 T h e i sopach of the Te r t i a r y sed imen ts a n d the k n o w n bas inwa rd d i p e x t r a p o l a t e d f r om the geology observed on the Queen C h a r l o t t e Is lands p rov ided the needed cons t ra in ts for the s u b - m o d e l . T h e a m p l i t u d e s of the f irst a r r i va l s are s t r ong ou t to 10 k m SRD where they then d r o p off qu i te sudden l y . B e y o n d 10 k m SRD, the se ismic s ignals are very weak. T h e onset of a h igher ve loc i t y b ranch can be seen between 10 k m a n d 11 k m SRD. R a y s t r aced t h rough the 2.0 k m / s layer generated theo re t i ca l t rave l t imes tha t m a t c h e d the observa t ions between the 0 a n d 1.0 k m SRD, cu rve a i n F i g u r e 4 . 4 A . T h i s va lue is cons is ten t w i t h the ve loc i t i es for the u p p e r sed iments measured by the sonic logs. E n e r g y re f rac ted t h r o u g h the 2.35 k m / s layer , s l igh t l y th i cker near O B S 2, emerge as p r i m a r y a r r i va ls be tween 1.0 k m and 1.8 k m SRD, (curve b ) . A g rad ien t of 0.35 k m / s / k m was requ i red for th is layer to m a t c h the observed a m p l i t u d e s a n d p roduce a m o d e l cons is ten t w i t h tha t f r om the ad jo i n i ng s u b - m o d e l ( O B S 2 — O B S 3). T h e a r r i va l s be tween 2.0 a n d 10.0 k m SRD were mode l l ed by rays t raced th rough a th i ck layer w i t h a ve loc i t y of 2.75 k m / s a n d a g rad ien t of 0.50 k m / s / k m (curve c ) . T h e theore t i ca l se i smograms and the obse rved d a t a are shown in F i g u r e 4.5. T h e theore t i ca l a m p l i t u d e s t ha t c o r r e s p o n d to curve c a c t u a l l y r e m a i n large for an a d d i t i o n a l number of t races beyond where the a m p l i t u d e s decrease a b r u p t l y . C o n s i d e r i n g the a m p l i t u d e s a t th is offset (between 6 and 8 k m MD ) t hey a p p e a r to be large a n d end more abrup t l y t han for the real d a t a a t equ iva len t offsets. T h e a c t u a l s t ruc tu re a long the top of the 5.0 k m / s layer may be respons ib le for th is bu t the unreversed po r t i on of th is segment and the weak a m p l i t u d e s for a r r i va ls f r om th is lower un i t d id not jus t i f y a more comp lex b o u n d a r y . Figure 4.4 Data for OBS 2 reverse profile (A) and ray tracing diagram for OBS 1—OBS 2 sub-model (B). (see figure 4.2 for explanation) -20 - 16 -14 -12 -10 -8 -6 -4 Source/Receiver Distance (km) -2 0 —r-2 4 6 8 10 12 14 16 Source/Receiver Distance (km) 18 20 Figure 4.5 Comparison of data for OBS 2 reverse profile (A) with synthetic seismograms (B) computed for ray tracing diagram in figure 4.4B. (see figure 4.3 for explanation) 78 The velocity and depth of the 5.0 km/s and 5.8 km/s layers were constrained by the sonic log from the Tyee well (Figure 2.1) and the isopach trend. The neighbouring sub-model provided an additional constraint by requiring that the model features be continuous beneath OBS 2. The theoretical amplitudes, Figure 4.5B, show the weak amplitudes for the arrivals from these two layers. This indicates that the seismic signal in the data may be obscured by the noise. The theoretical travel time curve, e, coincides with the onset of the higher velocity arrival near the beginning of the curve. This energy can not be seen clearly to extend beyond about 11 km SRD, although some weak coherent energy may be discernible. 4.3.3 Summary Figure 4.6 shows a three dimensional representation of the velocity changes along the sub-model between OBS 1 and OBS 2. Velocity and depth profiles were selected from different locations along the model and plotted in a 3-D display. The selected velocity-depth profiles are the solid curves increasing in a particular velocity-depth plane at fixed distance values. This presentation of the velocity information illustrates the overall velocity structure of the model, including velocity gradients. The geological interpretation for each velocity unit is also shown. The thick 2.75 km/s unit is interpreted as the Tertiary Skonun sediments (Sutherland Brown, 1968). Thickening of this unit across the basin is supported by the isopach for the sediments (Figure 1.1). The sonic log from the Tyee well (Figure 2.1) is compared in Figure 4.7 with the velocity versus depth curve from the 15 km MD location. A visually smoothed average of the sonic log is also indicated by the thick solid line. The two compare quite favourably. The velocity from the sonic log has measured the characteristics of OBS 2 _ Water Figure 4.6 Velocity cube for OBS 1—OBS 2 sub-model with a brief geological interpretation. Model distance is plotted at the top of the diagram along the axis projecting into the page. The depth axis is plotted along the vertical axis. Velocity is plotted at the base of the diagram along the axis projecting out of the page. The velocity cube was constructed by selecting five velocity depth profiles from the sub-model at 0, 7, 15 and 20 km MD. The relative position of the two OBSs is indicated at the top of the diagram. 80 ve r t i ca l wave p ropaga t ion over d is tances of me te rs whereas the ref ract ion m e t h o d tends to p rov ide h o r i z o n t a l ve loc i t y i n f o rma t i on over d i s tances of k i l ome te rs . T h e ve loc i ty and the g rad ien ts can change s ign i f i can t ly over shor t d i s tances , a fact wh i ch also cou ld be respons ib le for the di f ferences in the ve loc i t y d e p t h f unc t i on between the two di f ferent l oca t i ons . S u c h a charac te r i s t i c is obv ious when one compares the three sonic wel l logs for the C o h o , T y e e , and Sockeye wel ls w h i c h are near the a i rgun l ine (F igu re 4.8). G r a d i e n t s i n excess of 1.0 k m / s / k m are observed in the Sockeye we l l . T h i s observa t ion p rov i ded some of the j us t i f i ca t i on for the large g rad ien ts tha t were somet imes requ i red to m o d e l the a m p l i t u d e s of the observed a r r i va ls . T h e 5.0 k m / s un i t u n d e r l y i n g the Te r t i a r y sed imen ts of F i gu re 4.7 is i n te rp re ted as T e r t i a r y vo l can i cs , poss ib ly the M a s s e t f o rma t i on ( S u t h e r l a n d B r o w n , 1968). T h e 5.8 k m / s un i t be low th i s , a l t hough not wel l def ined for th is s u b - m o d e l , is in te rp re ted as the Pa leozo i c A l e x a n d e r Te r rane ( Y o r a t h and C h a s e , 1981). T h e ve loc i t y j u m p below th is has on ly been i n c l u d e d as a c o n t i n u a t i o n of a un i t f r om the ad jo in ing s u b - m o d e l and does not represent a t rue subsur face b o u n d a r y for th is s u b - m o d e l . 4.4 OBS 2—OBS 3 Sub-model In F i g u r e 4.1 the s u b - m o d e l for th is segment of the a i rgun l ine cor responds to the range between 20 a n d 41 k m MD as i n d i c a t e d by pos i t ions of the two O B S s . P re -c r i t i ca l re f lec t ions cou ld not be mode l l ed as they were no t observed in the d a t a due to thei r r eve rbe ra to r y na ture . Figure 4.7 Comparison of velocity versus depth profile (curve 1) from OBS 1—OBS 2 sub-model at 15 km MD and the sonic log from the Tyee well (curve 2) and its visually smoothed version (curve 3). 82 SHELL ANGLO COHO 2. 3. DEPTH (KM) SHELL ANGLO TYEE 2. 3. DEPTH (KM) SHELL ANGLO SOCKEYE B-10 1. 2. 3. DEPTH (KM) 4. Figure 4.8 Sonic logs from the three wells (Coho, Tyee and Sockey B-10, figure 1.1) nearest the airgun/OBS line. 83 4.4.1 Forward profile Results of the travel time fit to the data and the corresponding ray trace model are shown in Figure 4.9A and 4.9B, respectively. Curve a shows the fit to the observed travel times modelled by tracing rays through the upper 2.0 km/s layer. A gradient of 0.3 km/s/km was required to produce the proper curvature and match the amplitudes of these arrivals beyond the 3.0 km 5/?Z?(Figure 4.10b). The second theoretical travel time curve, b, modelled the energy arriving from the 2.30 km/s layer. The gap in the data between 3.8 and 5.8 km SRD resulted from an airgun malfunc-tion for approximately 20 minutes. A n emerging higher velocity branch can be seen between 3.0 and 3.8 km SRD which can be followed andi picked up along a continu-ous curve through the gap at 5.8 km SRD. Curve c models these arrivals as refractions through the Tertiary sediments (referred to in the previous sub-model). The amplitudes of the observed arrivals beyond 8.0 km SRD, Figure 4.9A and 4.10A, are weaker and, at the level of the noise, degrading the first break picks for this branch. These amplitudes die off sooner and more gradually than their theoretical counterparts. This may be due in part to the interference of other arrivals such as refracted multiples of possibly diffracted energy. The cut off in amplitudes was modelled by requiring that the volcanic unit (5.00 km/s layer) truncate the ray group for branch c (Figure 4.9). The theoretical travel time, curve d, for the group of rays that have travelled through the Tertiary vol-canic unit emerge as a slightly weaker set of arrivals extending to 15.4 km SRD (Figure 4.9A and 4.10A). The lower apparent velocity for these arrivals was modelled by the drop off in depth of the surface of the volcanics. Previous studies (Shouldiee, 1971, 1973; Yorath and Hyndman, 1983) show the top of the Tertiary volcanics to have been eroded during a period of uplift. There were even instances where the volcanics were o Cd © CO \ Q I E-< B o 4 6 8 10 12 14 16 18 Source/Receiver Distance (km) Model Distance (km) 4 6 8 10 12 14 16 18 20 22 20 2J CL. H Q Figure 4.9 Data for OBS 2 forward profile (A) and the ray tracing diagram for OBS 2—OBS 3 su model (B). (see figure 4.2 for explanation) 85 o Cd cn, o CO \ Q I' E-1 4 6 8 10 12 14 16 18 Source/Receiver Distance (km) 20 22 B o © eo* \ Q I E-2A l A 0 0 6 8 10 12 14 16 Model Distance (km) 18 20 22 Figure 4.10 Comparison of data for OBS 2 forward profile (A) and synthetic seismograms (B) puted for ray tracing diagram in figure 4.9B (see figure 4.3 for explanation). 86 not present in some of the wells. Shouldice (1973) also indicates that the steep dips seen on the reflection seismic data may also be fault related. The lower unit of the model, 5.9 km/s layer, models the energy thought to have just sampled the upper part of the interpreted Alexander Terrane, curve e. Observed arrivals can not be clearly seen in the data but when these data were bandpass filtered weak correlated arrivals could be discerned at the appropriate range and time, but the location of the first breaks could not be determined. The synthetic amplitudes for curve e Figure 4.9B indicate that these arrivals are weak. The theoretical seismogram, Figure 4.10B, and the observed seismogram, Figure 4.10A, compare well in their overall response. 4.4.2 Reverse profile OBS 3 recorded strong seismic signals, with a good signal to noise ratio (SNR) (Figure 4.1 IA and 4.12A). It was located 21 km from OBS 2 at the eastern end of the sub-model. The model at this end can be seen to be changing slightly. The upper two sedimentary units were modelled as for the forward profile, except that the lower unit required a lower velocity (2.25 km/s) beneath OBS 3 to provide some continuity between the adjacent sub-models at this location. The theoretical travel times for arrivals from these two layers in the sub-model are shown in Figure 4.11A, curves a and b. The water depth increases towards the east and a corresponding thickening of the uppermost sediments is observed in the sub-model. The theoretical travel time curve, c, modelled observed arrivals as due to ray travel paths through the Tertiary sedimentary layer (2.7 km/s). To match the observed arrivals out to 9.0 km SRD, the sediments were required to thicken towards the west. This thickening of the sediments 87 Source/Receiver Distance (km) Model Distance (km) 8 — — 1 Figure 4.11 Data for OBS 3 reverse profile (A) and ray tracing diagram for OBS 2—OBS 3 sub-model (B). Arrivals e and f correspond to the complex set of internally reflecting ray groups, (B), introduced to model the prominent secondary arrivals in (A) (see text for explanation). 88 was accomplished by thinning the underlying volcanic unit. A further thinning of this unit was then required to match the low apparent velocity of the arrivals from this layer, curve d. A velocity of 4.8 km/s was required and would seem reasonable, as the depth of burial decreases beneath OBS 3. At greater depths, the velocity increases to the 5.0 km/s value observed for the forward profile. The theoretical seismograms (Figure 4.12) match the amplitudes for both the c and d arrivals. The cutoff in the amplitude of the arrivals for branch c show very good agreement for this ease. The undulating nature of the arrivals (curve d), as modelled by rays through the volcanic layer, was achieved by slight variations of the upper boundary of the 2.25 km/s layer and the upper surface of the Tertiary volcanics. Unfortunately, the airgun malfunction referred to in the previous sections is also responsible for the premature end of the seismic traces at the western end of the record1 section. Rays traced through the model, Figure 4.11B, sparsely sample a region of high velocity (5.9 km/s). A portion of the travel-time curve for this sparse group of trays is indicated by the very short solid line near the end of the data. The complete travel time curve for this ray group was omitted beyond the range of the recorded data. The clear distinct secondary arrivals recorded by this OBS provided a good oppor-tunity to investigate the nature of these arrivals (see arrivals along curves e and f). The record sections for this and other marine surveys have been found to exhibit these secondary arrivals. The amplitudes are relatively strong and the arrivals appear as multiples of the near offset primary arrivals. They also exhibit some move out, similiar to that observed for reflections. These events are subsequently interpreted as multiple refractions, generated when the refracted energy was internally reflected during its up-coming travel path from a boundary with a large velocity contrast across it. Three of Source/Receiver Distance (km) Model Distance (km) Figure 4.12 Comparison of data for OBS 3 reverse profile (A) and synthetic seismograms (B) computed for ray tracing diagram in figure 4.11B. Arrivals labelled d, e and f correspond to similiarly labelled travel time curves in figure 4.11A. 90 these boundaries can be found in typical marine surveys: the air-water interface, the water-sediment interface and the sediment-basement interface. The group of curves, labelled f, are the travel time curves for rays internally reflected from the air-water interface and the sediment-water interface. The amplitudes of these arrivals are strong and persist for all displayed traces (Figure 4.12B). The result of reflecting the energy refracted through the 2.7 km/s layer, from the interface between it and the overlying layer, matches the weaker of the secondary arrivals on the data (see curve e). The arrivals labelled on the synthetic seismogram correspond to the curves shown in Figure 4.11 A . The amplitudes for the e and f arrivals match the relative amplitudes of the secondary arrivals in the data remarkably well. The weaker amplitudes for arrivals labelled e are expected since the contrast for this internally reflected set of rays was smaller than that for the air-water or water-sediment interface. The modelling of these secondary arrivals gives an added degree of confidence to the models since the model is now doubly sampled for some of the upper units. In general, theoretical travel time curves matched observed arrivals to within the error of picking the first breaks for these arrivals. The worst case showed a misfit of no more than ±0.20s for the secondary arrival fits. 4.4.3 Summary The velocity profile cube for this sub-model is shown in Figure 4.13. Seven profiles were selected along the model since it exhibited large lateral velocity variations. Geo-logical interpretation of the various units is shown; the units follow continuously from the OBS 1-OBS 2 sub-model considered in section 4.3 . The lateral velocity variations 92 can be followed easily in this display. The thinning of the Tertiary volcanic layer can be seen as the 4.8 to 5.0 km/s velocities shift to greater depths. Other features also become evident — the upper sedimentary layers thicken and the thicker sedimentary unit begins to show less of a velocity contrast with the overlying thin units. As with the previous sub-model, the lower unit, indicated by the ?, is not defined for this sub-model. 4 .5 OBS 3—OBS 4 Sub-model This third and final sub-model is located between 41 km and 62.5 km MD in Figure 4.1. The forward and reverse profiles recorded across this segment of the airgun line contain several striking anomalous features and a large noise level. One would expect to see unusual events in the data as the subsurface structure has been shown to change rapidly at the eastern margin of the basin (Shouldiee, 1971, 1973; Staeey and Stephens, 1969; Yorath and Cameron, 1982; Young, 1981). However some features were not antic-ipated. Modelling for a consistent travel time and amplitude match for both the forward and reverse profiles proved to be difficult. This was due to the obvious differences in the data sections recorded at OBS 3 and OBS 4 (Figures 4.14A and 4.16A). 4 . 5 . 1 Forward profile The data for the forward profile and the ray trace model are shown in Figure 4.14. Strong amplitudes extend to 9 km SRD for the refracted arrivals. The amplitude sud-denly drops off, then a delayed signal appears between 14 and 18 km SRD. These characteristics are shown more clearly by the hydrophone component (Figure 11.11, Appendix II). Figure 4.14 Data for OBS 3 forward profile (A) and the ray tracing diagram for OBS 3—OBS 4 sub-model (B). Some reflected refraction ray paths have been included in an attempt to model particular features (see figure 4.2 and text for explanation). 94 M o d e l l i n g for the sed imen ta ry layers p roceeded as before and the t rave l t imes are s h o w n , curves a and b, in F i g u r e 4 . 1 4 A . T h e ve loc i t y of the m i d d l e sed imen ta ry un i t , t ha t is t r u n c a t e d by the th i ck Te r t i a r y un i t , requ i red a lower ve loc i t y of 2.2 k m / s . T h e ve loc i t y con t ras t be tween the u p p e r layers and th is m i d d l e un i t d im in i shes to zero towards the eastern end of the m o d e l due to the ve loc i t y g rad ien t a n d i nc reas ing th i ckness of the u p p e r m o s t layer (see F i g u r e 4 .19) . R a y s t h r o u g h the lower sed imen ta ry u n i t , the T e r t i a r y sed imen ts , (ve loc i ty 2.6 k m / s ) generate the theore t i ca l a r r i va l b ranch e. In th is pa r t of the m o d e l , the in te rp re ted Te r t i a r y vo l can i c layer is sha l lower , as r equ i r ed by the t rave l t ime charac te r i s t i cs of the d a t a and the i sopach i n fo rma t i on (Shou ld iee , 1971). C u r v e d shows the theore t i ca l t rave l t imes for th is un i t . T h e sudden decrease in the a m p l i t u d e at 9 k m SRD necessi tated the i n t r oduc t i on in the m o d e l of a feature w h i c h c o u l d cause the observed charac te r i s t i c . A fau l ted b lock or a low ve loc i t y layer are t y p i c a l m e c h a n i s m s chosen to cause the desired effect. T h e fau l t so lu t ion was cons idered bu t d i d not y ie ld a m o d e l cons is ten t w i t h the reverse prof i le . F u r t h e r m o r e , there is no ev idence to s u p p o r t m a j o r fau l t i ng in th is reg ion . A low ve loc i t y zone w h i c h p inches out to the west was i n t r o d u c e d immed ia te l y be low the Te r t i a r y vo lcan ics . T h e syn the t i c se i smogram sect ion (F igu re 4 .15B) shows tha t the a m p l i t u d e for the ar r i va ls for phase d drops off sha rp l y a t abou t 9 k m it S R D , as obse rved on the d a t a sec t i on . A ve loc i t y of 3.5 k m / s was chosen for th is low ve loc i t y zone t o be cons is ten t w i t h a s im i l a r l y i n te rp re ted low ve loc i t y layer f r om the C h a r l o t t e sub -bas in (C lowes and G e n s - L e n a r t o w i c z , 1985). T h e low v e l o c i t y / t h i c k n e s s c o m b i n a t i o n was cons is ten t w i t h the de lay observed in the a r r i va l s beyond 11 k m SRD. In a d d i t i o n , the i n t r o d u c t i o n of the low ve loc i t y zone was cons is ten t w i t h d a t a f rom the reverse prof i le as d iscussed in the next sec t ion . 95 Source/Receiver Distance (km) Model Distance (km) Figure 4.15 Comparison of data for OBS 3 forward profile (A) and synthetic seismograms (B) com-puted for ray tracing diagram in figure 4.14B (see figure 4.3 for explanation). 96 T h e s o m e w h a t i so la ted bu rs t o f h igh a m p l i t u d e energy observed on the record sec t ion at a b o u t 2.5s and between 14 and 15 k m SRD, is more p ronounced on the h y d r o p h o n e c o m p o n e n t (F igure 11.11, A p p e n d i x II). S i m p l e re f rac t ions th rough the A l e x a n d e r ter rane (5.9 k m / s layer) be low the low ve loc i t y zone d id no t generate a m p l i -tudes c o r r e s p o n d i n g to those on the observed sec t ion . C o n s e q u e n t l y the burs t of energy was in te rp re ted to represent a l oca l f ocuss ing effect. A n a t t e m p t was then made to focus the energy by select ing a n u m b e r of t r ave l pa ths to g ive the same t rave l t imes . F i r s t , rays were in te rna l l y ref lected f r o m the base of the 2.2 k m / s layer . T rave l t imes for th is ray g roup are shown in F i g u r e 4 .14A as cu rve g . B u t because of the t ime spent in the T e r t i a r y sed imen ts and the low ve loc i t y con t ras t between th is layer and the 2.2 k m / s layer , these ar r i va ls were de layed by too m u c h and had weak a m p l i t u d e s (F igu re 4 .14A a n d 4 . 1 5 B ) . N e x t rays were ref lected f r o m the base of the s e d i m e n t - b a s e m e n t boundary ( F i g u r e 4 .14B and cu rve h in F i g u r e 4 . 1 4 A ) . These a r r i va ls represent the po r t i on of cu rve l i out to 16 k m SRD a n d , as can be seen in F i g u r e 4 .15B , con t r i bu te the mos t to the large a m p l i t u d e s tha t are obse rved . T h e d i rec t a r r i va l s t h rough the A l e x a n d e r te r rane , curve e , are mode l led for the weak a m p l i t u d e a r r i va ls observed at t h i s range. T h e h y d r o p h o n e componen t shows the a r r i va l s more c lear l y , bu t the first b reaks cou ld not be de te rm ined because of the low a m p l i t u d e s and the noise. These low a m p l i t u d e s are ev iden t as first arr iva ls on the theo re t i ca l se i smograms and are c lear ly not respons ib le for the foeussed energy a t th i s range . There fo re , rays were ref lected f r om the base of the A l e x a n d e r Terrane to a u g m e n t those on cu rve h. These ref lect ions f o r m curve f and the rest of cu rve h, w i t h some over lap between the two g roups m a k i n g up cu rve h . T h e lower set of ref lected energy con t r i bu te to the overa l l a m p l i t u d e s observed in the theo re t i ca l se ismograms. A po r t i on of these rays , f r o m the deep ref lect ing g roup , ac tua l l y refract th rough the 97 u p p e r m o s t par t o f the un i t (7.5 k m / s ) where ve loc i t ies and g rad ien ts are uncons t ra i ned by the d a t a . T h e concave shape of the lower re f lec t ing b o u n d a r y is s ign i f i can t for the resu l t i ng t rave l t imes . In order to get an a p p r o p r i a t e set of a r r i va ls , the s t r uc tu re had to have a concave shape . In the mode l ! w i t h o u t th is s h a p e , the c o m b i n a t i o n of the t h i n n i n g of the sed imen ta r y layer and the sha l l ow ing of the Ter t ia ry vo lcan ic un i t caused the theore t i ca l a r r i va ls to have larger appa ren t ve loc i t ies . To m a t c h the appa ren t ve loc i t y of the observed a r r i va l s , a means of i nc reas ing the t rave l t imes as the offset increased was needed. T h e concave s t ruc tu re p rov ided the necessary m e c h a n i s m and focussed the energy. Howeve r , the t rave l t imes do not appea r to m a t c h any speci f ic set o f a r r i va ls in the energy burs t (F i gu re 4.14A). Never the less , the theoret ica l s e i s m o g r a m s do m a t c h the overa l l charac te r i s t i cs of the observed se ismic sec t ion (F igu re 4.15). It is th i s observa t ion w h i c h leads to the conc lus ion t h a t these a r r i va ls represent a f ocuss ing of energy, most l i ke ly due to cons t ruc t i ve a n d / o r des t ruc t i ve in ter ference f r om var ious a r r i va ls . 4 . 5 . 2 R e v e r s e p r o f i l e T h e reversed prof i le was m o d e l l e d s imu l t aneous l y w i th the f o rwa rd prof i le . T h e d a t a and ve loc i t y s t r uc tu re , w i t h t r aced rays , are dep ic ted in F i g u r e 4.16. No i se levels were re la t i ve ly h igh for th is d a t a set a n d requ i red analys is of record sect ions for al l th ree c o m p o n e n t s . B a n d p a s s f i l te r ing was app l ied when necessary. T h e mode l l i ng of the p r i m a r y a r r i va l s be tween 0 and -8 k m SRD fo l lowed tha t for the fo rward mode l l i ng b u t now, a r r i va ls f rom the u p p e r m o s t sed imen ta ry un i t , are i m m e d i a t e l y fo l lowed by those f r om the th i ck Te r t i a r y sed imen ts (curves a and b ) due to the t r unca t i on of the m i d d l e un i t . T o m o d e l the f i rst a r r i va ls , cu rve a , and generate the necessary ve loc i t y 98 Source/Receiver Distance (km) Model Distance (km) Figure 4.16 Data for OBS 4 (A) and ray tracing diagram for OBS 3—OBS 4 sub-model (B) (see figure 4.2 for explanation and the text for an explantion of events labelled D). 99 con t ras t to m o d e l the b u m p - l i k e feature observed in the d a t a between 2 a n d 4 k m SRD, the v e l o c i t y of the sed imen ts be low O B S 4 were decreased to 1.9 k m / s . B y d i rec t ing the rays t h r o u g h a s t r u c t u r a l h igh on the Ter t i a ry sed imen ta ry layer af ter t rave l l i ng th rough the 1.9 k m / s layer , the a r r i va l s exper ience an increase in apparen t ve loc i ty , cu rve b . T h i s is i m m e d i a t e l y fo l lowed by a decrease in appa ren t ve loc i t y as the rays enter the uppe r 2.0 k m / s layer . T h e t heo re t i ca l se i smograms m a t c h the b u m p - l i k e feature we l l in amp l i t ude a n d shape of the t rave l t ime cu rve (F igu re 4.17) . T h e 2.2 k m / s layer is p i nched out by the r ise of the s t r uc tu re represen t ing the Te r t i a r y sed iments . T h e ex tens ion of b as a secondary a r r i va l resu l ts f r om ray pa ths th rough the Te r t i a r y sed imen ts , the 2.6 k m / s layer . T h e c o m p u t e d t r a v e l t imes , cu rve c, for the 4.8 k m / s layer are t r unca ted by the ex tens ion of the low ve loc i t y layer i n t r oduced in the fo rward prof i le. T h e amp l i t ude of the a r r i va l s in the d a t a do not a p p e a r to d rop off as fast as those f rom the fo rward profi le bu t th is cu t off is a lso suppor ted 1 by the filtered1 h y d r o p h o n e c o m p o n e n t (F i gu re 11.12, A p p e n d i x II). T h e o r e t i c a l a m p l i t u d e s for the a r r i va l s f r o m the Ter t ia ry vo lcan ics are la rger t h a n the first break a r r i va ls for the observed da ta . However , for the hyd rophone c o m p o n e n t , the a m p l i t u d e s of the a r r i va l s , c, are more c lear ly seen even though the s igna l is at the noise leve l . There fore a g rad ien t of 0.70 k m / s / k m was chosen to ma in ta i n a ba lance between the a m p l i t u d e observed on the v e r t i c a l componen t a n d those observed on h y d r o p h o n e c o m p o n e n t . A t ime delay can be observed for the a r r i va ls m a t c h e d by curve d . T h e low ve loc i ty layer m a y be the U p p e r C re taceous sed iments tha t were depos i ted across the suture zone def ined by Y o r a t h and C h a s e (1981) as the post suture assemblage. A s ment ioned in sec t ion 4.5.1, a s im i l i a r low ve loc i t y layer was requ i red in a m o d e l f r om the C h a r l o t t e 100 -24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 Source/Receiver Distance (km) Model Distance (km) Figure 4.17 Comparison of data for OBS 4 reverse profile (A) and synthetic seismograms (B) computed for ray tracing diagram in figure 4.1 IB (see figure 4.3 for explanation). 101 s u b - b a s i n and has been i n te rp re ted as the post suture assemblage (C lowes and G e n s -L e n a r t o w i c z , 1985). T h e t heo re t i ca l t r ave l t i m e , cu rve d, was c o m p u t e d by t rac ing rays t h r o u g h the b lock i n te rp re ted as the A l e x a n d e r te r rane. P o o r s igna l - to -no ise ra t ios p revented an accurate d e t e r m i n a t i o n of the f i rst b reaks for th is b r a n c h . T h e hyd rophone c o m p o n e n t of the d a t a w a s b a n d p a s s filtered between 5 and 12 H z a n d succeeded in b r i n g i n g out the p r i m a r y a r r i v a l b r a n c h ( F i g u r e 11.12, A p p e n d i x II) w h i c h a ided in the final de te rm ina t i on of i ts p o s i t i o n (curve d). A s shown on the theore t i ca l se ismograms ( F i g u r e 4 . 1 7 B ) , the s l ight increase in a m p l i t u d e has been m o d e l l e d fa i r l y successfu l ly by the a d d i t i o n of ref lect ions f r o m the top of the 7.2 k m / s layer (curve e on F i g u r e 4 .16A) T h e d a t a i n F i g u r e s 4 . 1 6 A and 4 . 1 7 A show a s t r ong set of secondary a r r i va ls , event D, w h i c h have not been mode l l ed exp l i c i t l y . T h i s large a m p l i t u d e event ranges f rom 13 to 24 k m SRD at a reduced t ime of 2.5 seconds. To exp la in the nature of th is event mo re eas i ly , the order in w h i c h the inves t iga t ion proceeded w i l l be desc r i bed . I n i t i a l l y the event was t h o u g h t to represent a s t rong set o f p r i m a r y a r r i va ls and were m o d e l l e d as s u c h . H o w e v e r , the resu l t i ng m o d e l was f o u n d to be incons is ten t w i t h the i n t e rp re ta t i on of the fo rward prof i le . T h e general s t ruc tu re of the un i ts be low the s e d i m e n t s were requ i red t o have an eas tward d i p , wh i ch is not i n accord w i th the a c c e p t e d reg iona l geology of the area. U p o n a closer e x a m i n a t i o n of the d a t a , weak a r r i va l s pa ra l l e l l i ng event D w i t h an a r r i va l t ime app rox ima te l y 0.8 seconds ear l ier were no ted a n d subsequent l y m o d e l l e d as d e s c r i b e d . T h i s raised the q u e s t i o n — w h a t s i tua t ion w i l l g ive r ise to weak p r i m a r y a r r i va ls fo l lowed by large a m p l i t u d e secondary a r r i va ls? T h e first a t t e m p t to solve th is p r o b l e m was to m o d e l the ear l ier events as head waves a n d the secondary event as ref lect ions f r o m the base of the layer generat ing the head 102 waves . T h i s so lu t ion fa i led to generate the obse rved a m p l i t u d e s and the para l le l na ture of the two events. A so lu t i on w h i c h adequate ly exp la ined the obse rva t i ons evo lved fo l low ing e x a m i n a -t i o n of the three c o m p o n e n t s for th is d a t a set. F i g u r e 4.18 shows the three componen ts of d a t a for O B S 4 w i t h i n the range and t ime w i n d o w of in terest . T h e ver t i ca l c o m p o -nen t , F i g u r e 4.18c, and the h o r i z o n t a l c o m p o n e n t , F i g u r e 4.18b, were p lo t ted w i t h the same scale fac tor w h i c h is three t imes greater t h a n the fac tor used for the hydrophone c o m p o n e n t , F i g u r e 4.18a. C o m p a r i n g F i g u r e 4.18 b a n d c, the large a m p l i t u d e event D is s t ronger and more coheren t on the h o r i z o n t a l c o m p o n e n t . C o m p a r i n g F i g u r e 4.18a a n d 4.18e the weak p r i m a r y a r r i va l s , d are observed to ex tend across b o t h record sec-t i ons . E v e n t d is not seen on the ho r i zon ta l c o m p o n e n t , whi le event D is not present on the h y d r o p h o n e c o m p o n e n t . F i n a l l y the pa ra l l e l na tu re of the two events suggests a re-l a t i o n s h i p between t h e m a n d the t r ave l - t ime di f ference between t h e m is consis tent w i th wave convers ion at the basement sed imen t basement . These observa t ions are d iagnost ic of conver ted S-wave a r r i va l s . T h e r e is a larger h o r i z o n t a l c o m p o n e n t of the i n c o m i n g wave and i t is not t r a n s m i t t e d th rough the wa te r . A rev iew of the l i t e ra tu re suppor ts the premise t ha t the la te r , s t ronger a m p l i t u d e a r r i va l s cou ld be conver ted phases. C o n -ve r ted S - w a v e a r r i va ls have been shown to be charac ter is t i c of a n u m b e r of mar ine se ismic re f rac t ion su rveys ( W h i t e et al., m a n u s c r i p t in p repara t ion 1986; C h e u n g a n d C l o w e s , 1981; A u and C l o w e s , 1984), a l though a l l o f these were in deep water env i ron -men ts . W h i t e and Stephens (1980) rev iewed the p roper t ies for shear wave convers ion. M o d e convers ion between P and S waves occurs w h e n the P or S wave encounters an in te r face w i t h large ve loc i t y con t ras t . In m a r i n e e n v i r o n m e n t s there are two such inter-faces, the w a t e r - s e d i m e n t in ter face a n d the b a s e m e n t - s e d i m e n t in ter face. T h e P to S 103 - 2 4 - 2 2 - 2 0 - 1 8 - 1 6 - 1 4 O B S 4 H y d r o p h o n e ( 5 - 1 2 ) - 1 2 - 1 0 - 2 4 - 2 2 - 2 0 - 1 8 - 1 6 - 1 4 O B S 4 H o r i z o n t a l ( 5 - 1 2 H z ) - 2 0 - 1 8 - 1 6 - 1 4 O B S 4 V e r t i c a l ( 5 - 1 2 H z ) F,gure 4.18 Comparison of filtered hydrophone (a), horizontal (b), and vertical (c) components for OBS 4. The source/receiver distance is plotted along the horizontal axis. The data is plotted from 1 to 3 seconds m reduced time format. Primary arrival events are labelled with d and secondary P to S converted phases are labelled by D (see text for explanation). 104 conversions at the basement-sediment interface have been observed to yield S-wave ar-rivals that are much larger than the P-wave arrivals ( Cheung and Clowes, 1981; White et al., manuscript in preparation 1986). Cheung and Clowes used these large amplitude arrivals to determine the position of the weak P-wave arrivals. White et al. modelled these conversions using 1-D WKBJ synthetic seismograms which included an option for phase conversions (Chapman, 1978). Where conditions enhance the conversion of P to S waves for the basement-sediment interface, they reduce the efficiency for S to P wave conversions at the water-sediment interface (White and Stephens, 1980). The lack of a doubly converted phase (P to S to P) on the hydrophone may be due to such an effect. 4 . 5 . 3 Summary Figure 4.19 shows the velocity cube display for this sub-model along with the geo-logical interpretation for the major units. The Tertiary sediments thin substantially as the Tertiary volcanic basement structure rises. This factor may account for the pres-ence of the P to S wave arrivals seen only on OBS 4. The low velocity unit has been interpreted as Upper Cretaceous sediments. This unit pinches out at 5.0 km MD. This pinch-out was required as there was no evidence for a low velocity layer in the adjoining sub-model for OBS 3-OBS 2. The Alexander Terrane is nearly truncated by the low-ermost unidentified unit, for which only the position of its upper surface is even partly constrained by the data. 106 C H A P T E R V DISCUSSION A N D CONCLUSIONS 5.1 Discussion of the Final Composite Model T h e f ina l ve loc i t y m o d e l was comp i l ed f r om the three s u b - m o d e l segments of C h a p t e r I V . F i g u r e 5.1 shows the compos i t e m o d e l w i t h a 5:1 v e r t i c a l exaggerat ion and the same m o d e l w i t h no ve r t i ca l exaggera t ion . A legend , re la t ing the var ious un i ts to the ve loc i t i es , appea rs be low the m o d e l . D u r i n g the comp i l a t i on of the f inal m o d e l f r o m the three s u b m o d e l s , m i n o r d i sc repanc ies occu r red at thei r c o m m o n boundar ies beneath O B S 2 a n d O B S 3 . T h e compos i te m o d e l was ad jus ted to make the s u b - m o d e l bounda r i es a l l cons is ten t . T h e var ious i n d i v i d u a l ve loc i t y b l ocks , shown in F i g u r e 4 .1 , have been rep laced by average ve loc i t ies and g rad ien ts represen t ing the m a j o r un i ts . T h e i n te rp re ta t i on is n o n - u n i q u e , bu t is based on a carefu l eva lua t ion of the c o m -par ison of observed and theore t ica l t rave l t imes and a m p l i t u d e s for reversed prof i les. S m a l l l a te ra l var ia t ions in ve loc i t y and g rad ien t appea r to be necessary a l t hough the i r r ep resen ta t i on by d iscrete b o u n d a r i e s in F i g u r e 4.1 is an a r t i f ac t of the m o d e l i n p u t for the a s y m p t o t i c ray theory m o d e l l i n g p r o g r a m . T h e t h i c k e n i n g of un i t D at a b o u t 30 k m d is tance is a feature i n t r oduced to accoun t for a m p l i t u d e s and appa ren t ve loc i t ies of p r i m a r y and secondary a r r i va ls on b o t h the f o r w a r d and reverse prof i les for O B S 2 - O B S 3. T h i s seemed to be a necessary feature of the m o d e l as d id the i n t r o d u c t i o n of large ve loc i t y g rad ien ts (up to 0.5 k m / s / k m ) . C l o w e s a n d G e n s - L e n a r t o w i c z (1985) used s im i l i a r g rad ien ts for a lower sed imen ta ry un i t for one of t he i r mode ls f r om Q u e e n C h a r l o t t e S o u n d and the sonic logs (sect ion 4.3.3) s u p p o r t the ex is tence of large a n d v a r y i n g g rad ien ts for the sed imen ts . S i m i l i a r l y the rise of un i t E to the east and the ex is tence of a low ve loc i t y zone (un i t F) be low are features 0 10 20 30 40 50 60 VE 1:1 OBS 1 OBS 2 OBS 3 OBS 4 f \ Distance (km) f f 0 5 10 15 20 25 30 35 40 45 50 55 60 VE 5:1 VELOCITY (km/s) ; GRADIENT (km/s/km) • 1.49 ; 0.001 A CEI 4.80 ; 0.500 E IZZi 2.00 ; 0.250 B 3.50 ; 0.100 F a 2.27 ; 0.300 C KS 5.90 ; 0.230 G IZZD 2.72 ; 0.400 D WM 7.70 ; 0.300 H ? Figure 6.1 Final composite structural velocity model. The model is displayed with no vertical exagger-ation (top insert) and a vertical exaggeration of 5:1 . The legend at the bottom of the figure related the symbols for the structural units to the velocities and gradients for those units. Features in this model below 5.0km are either poorly constrained or completely unconstrained (see text for explanation). 108 of the m o d e l w h i c h were i n t r o d u c e d to accoun t for pa r t i cu la r d a t a cha rac te r i s t i cs . T h e ve loc i t y for un i t E was m o d e l l e d as 4.8 k m / s w i t h an average g rad ien t of 0.5 k m / s / k m . T h i s g rad ien t decreases f r o m large values in the east (0.7 k m / s / k m ) to sma l le r values (0.2 k m / s / k m ) in the west . T h e ve loc i t y for the u p p e r sur face has been ind ica ted as 4.8 k m / s , bu t increases to 5.0 k m / s w i t h i nc reas ing dep th of bu r i a l . T h e i n t r o d u c t i o n of low ve loc i t y layers in to m o d e l s deve loped f r om se ismic re f ract ion i n te rp re ta t i ons presents the in te rp re te r w i t h a greater degree of f reedom in d e t e r m i n i n g the final m o d e l . T h i s ar ises because low ve loc i t y zones represent a h i d d e n p r o b l e m for se ismic re f rac t ion m e t h o d s . T h e same de layed t rave l - t ime can be genera ted for many ve loc i t y a n d th ickness c o m b i n a t i o n s . T o achieve a c o m m o n basel ine for the compa r i son of these resu l ts w i t h o thers a ve loc i t y for the low ve loc i t y layer was chosen to m a t c h t h a t used by C l o w e s a n d G e n s - L e n a r t o w i c z (1985) in the i r m o d e l . B y us ing s im i l i a r ve loc i t i es , the c o m p a r i s o n of the m o d e l desc r i bed here w i t h tha t for Q u e e n C h a r l o t t e S o u n d w i l l not be d i m i n i s h e d by the non -un iqueness assoc ia ted w i t h the inc lus ion of the low ve loc i t y zone. O B S 3 - O B S 4 reverse prof i le (F igu re 4.16 a n d 4.17) shows an u n e q u i v o c a l a r r i va l w i t h an appa ren t ve loc i t y of 6.0 k m / s . A s th is is the on ly prof i le where th is phase ex is ts , un i t G is poo r l y de f ined elsewhere and essent ia l ly in fer red. U n i t H is not def ined be tween 0 and 40 k m a n d on ly poo r l y def ined beyond tha t range. Re f lec t i ons f rom the t o p of the layer beyond 40 k m were i n t r o d u c e d to sat is fy par t i cu la r events for O B S 3 - O B S 4 i n te rp re ta t i on but the un i t i tsel f was never s a m p l e d . Ve loc i t i es were chosen to p roduce a ref lect ion coef f ic ient necessary to generate reasonable a m p l i t u d e s to m a t c h the energy burs t d i scussed in sec t ion 4.5.1 . In s u m m a r y , the lower 3 to 4 k m of the m o d e l shown in F i g u r e 5.1 are e i ther poo r l y cons t ra ined or u n c o n s t r a i n e d . 109 UNIT STRATIGRAPHIC INTERPRETATION A Water B Pleistocene Sediments C Pleistocene and/or Pliocene Sediments D Tertiary Skonun Sediments E Tertiary Masset Volcanics F Upper Cretaceous Sediments G Paleozoic Alexander Terrane H Plutons ? T a b l e 5.1 S u m m a r y of s t r a t i g raph i c i n te rp re ta t i on . Tab le 5.1 s u m m a r i z e s the s t r a t i g raph i c i n t e rp re ta t i on o f the var ious un i t s . U n i t s B a n d C can be cons idered co l lec t i ve ly , espec ia l ly in l igh t of the loss of a ve loc i t y con -t ras t be tween 50 k m and 58 k m where the lower one pinches out . These have been in te rp re ted as P l e i s t o c e n e a n d / o r P l i o c e n e sed imen ts la id down in a nearshore depo-s i t i o n a l e n v i r o n m e n t . T h e m a x i m u m th ickness reaches a p p r o x i m a t e l y 1.0 k m between 30 and 40 k m . T h e ve loc i t i es and the c o m b i n e d th i ckness of these un i ts are s im i l a r to m o d e l s f r o m Queen C h a r l o t t e S o u n d (C lowes a n d G e n s - L e n a r t o w i c z , 1985). U n i t D has been i n te rp re ted as the Te r t i a r y S k o n u n F o r m a t i o n (Su the r l and B r o w n , 1968; Shou ld iee , 1971, 1973). These are U p p e r M i o c e n e sed imen ts depos i ted in near shore ma r i ne and n o n - m a r i n e e n v i r o n m e n t s . T h e Te r t i a r y sed imen ts represent the th ickest sed imenta ry un i t for the Heca te S t ra i t m o d e l , r each ing a m a x i m u m th ickness of a p p r o x i m a t e l y 3.0 k m . T h e comb ined m a x i m u m th ickness for a l l the sed imenta ry un i ts is a b o u t 4.0 k m , s i m i l i a r to tha t def ined by Shou ld iee (1971, 1973) fo r Heca te sub-bas in a n d C lowes and G e n s - L e n a r t o w i c z (1985) for C h a r l o t t e s u b - b a s i n . U n i t E has been i n te rp re ted as the Ter t ia ry M a s s e t fo rmat ion wh i ch cons is ts of sub-aer ia l l y e rup ted vo lcan ics . T h i s un i t a lso e x h i b i t s features o f a bur ied e ros iona l surface. U s i n g i n fo rma t i on co l lec ted f r o m the wel ls in H e c a t e S t ra i t by Shou ld iee (1971, 1973), 110 Y o r a t h a n d C h a s e (1981) suggest t ha t th is f o r m a t i o n had been p rev ious ly up l i f t ed and e roded between U p p e r a n d L o w e r M i o c e n e t imes . T h e Tyee wel l (see F i g u r e l . l ) showed no Te r t i a r y M a s s e t vo l can i cs ; however i t d i d pene t ra te a h igh ve loc i t y ma te r i a l w h i c h has been i n te rp re ted by Shou ld i ce (1971, 1973) as a Pa leozo ic i n t rus i ve . T h i s is cons is ten t w i t h the m o d e l p resen ted here since Shou ld i ce (1973) i nd i ca ted tha t e ros iona l chan -nels had removed the ove r l y i ng m a t e r i a l . Te r t i a ry M a s s e t vo lcan ics have been logged in o the r wel ls and have been in fer red f r o m re f lec t ion se ismic d a t a (Shou ld i ce , 1971, 1973). C l o w e s and G e n s - L e n a r t o w i c z (1985) have in fer red Ter t i a ry v o l c a n i c s beneath Q u e e n C h a r l o t t e S o u n d based on a s im i la r re f rac t i on survey. These range in th ickness be tween 1.5 and 3.5 k m as c o m p a r e d w i t h 0.2 and 1.8 k m for the v o l c a n i c s beneath H e c a t e S t ra i t . T h e ve loc i t ies ass igned to th is un i t (4.8-5.0 k m / s ) c o m p a r e favourab ly w i t h the 5.2 k m / s ve loc i t y for the Te r t i a r y vo l can i cs benea th Queen C h a r l o t t e S o u n d . T h e i sopach for H e c a t e S t ra i t (F igu re 1.5) shows t h i n n i n g of the sed imen ts i n the west to less than 0.1 k m . T h e i sopach is not we l l de f ined in th is area a n d the resu l ts here w o u l d favour an eas tward shi f t for the basin edge. U n i t F , w h i c h on ly appea rs beneath the Te r t i a r y vo lcan ics at the eas te rn end of the m o d e l , was in te rp re ted as the U p p e r C r e t a c e o u s Queen C h a r l o t t e G r o u p of the pos t su tu re assemblage ( Y o r a t h a n d C h a s e , 1981). T h e y recognized U p p e r C re taceous sed imen ts ( the i r post su ture assemblage) in the Tyee wel l w h i c h lies close to the a i r -g u n / O B S l ine . T h e y also observe tha t no Te r t i a r y vo lcan ics were pene t ra ted by th is we l l . T h e present s tudy d id not requi re the ex tens ion of the low ve loc i ty zone f rom the eas tern end to the p o r t i o n of the m o d e l near the we l l , bu t the ve loc i ty un i t in te rpre ted as the Te r t i a r y vo l can i cs is requ i red to ex tend across the bas in . T h e Te r t i a r y vo l can i cs , where present i n the Heca te s u b - b a s i n , u n c o n f o r m a b l y over ly the U p p e r C re taceous I l l s e d i m e n t s (Su the r l and B r o w n , 1968; Shou ld i ee , 1971, 1973) and i t is not unusua l for the l a t t e r un i t to be absent . U n i t G has been i n te rp re ted as the Pa leozo ic rocks of the A l e x a n d e r Te r rane wh i ch are be l i eved to under l y the H e c a t e s u b - b a s i n . T h e ve loc i t y and g rad ien t for th is un i t were m o d e l l e d as 5.9 k m / s and 0.23 k m / s / k m , respect ive ly . T h e ve loc i t y compares f a v o u r a b l y w i t h the va lue of 6.0 k m / s f r om the sonic log of the Tyee we l l (see F igu re 4 .2 ) . C l o w e s and G e n s - L e n a r t o w i c z (1985) def ine a ve loc i t y of 6.0 k m / s for the Mesozo ic W r a n g e l l i a Ter rane benea th Q u e e n C h a r l o t t e S o u n d . B a s e d upon these resu l t s , i t wou ld a p p e a r t h a t the ve loc i t ies for the W r a n g e l l i a and A l e x a n d e r Terranes are c o m p a r a b l e . U n i t H has been ten ta t i ve l y i n te rp re ted as represen t ing p lu tons , but as no ted ear l ier , wh i l e i t s presence is i n d i c a t e d , l i t t l e can be in fe r red abou t i ts p roper t ies . T h e uppe r b o u n d a r y of th is un i t beyond 40 k m appears to have a comp lex s t ruc tu re wh i ch rises a b r u p t l y a t the eastern end of th i s m o d e l . T h i s is cons is ten t w i t h the resul ts of g rav i ty f r o m a prof i le co i nc i den t w i t h the a i r g u n / O B S l ine (Stacey and S tephens , 1969). T h e i r i n te rp re ta t i on is shown in F i g u r e 5.2 for c o m p a r i s o n w i th the gross charac ter is t i cs of the ve loc i t y m o d e l . 5.2 Conclusion A se ismic s t r u c t u r a l ve loc i t y m o d e l has been deve loped for the Heca te s u b - b a s i n . T h i s m o d e l was f o u n d to be cons is ten t w i th prev ious s tud ies f rom Heca te S t ra i t and Q u e e n C h a r l o t t e S o u n d . T h e i n t e r p r e t a t o n is also cons is ten t w i t h the geology expec ted benea th H e c a t e S t ra i t based on geological s t ruc tu res ex t rapo la ted f r om the Queen C h a r -l o t te Is lands . T h e nature of the sur face of the Te r t i a r y vo lcan ics suggests an erosion al sur face s im i l i a r to t ha t for the T e r t i a r y M a s s e t vo lcan ics on G r a h a m Is land and in the 112 -301 kn Or OBSERVED BOUGUER ANOMALY REGIONAL ANOMALY CALCULATED ANOMALY RESIDUAL ANOMALY INSULAR TECTONIC BELT I 50 HECATE DEPRESSION WO COAST CRYSTALLINE BELT 200 km UEST DENSITY CONTRAST RELATIVE TO SURROUNOING ROCKS EAST -0.5 g / c ^ 0.3 g / c r Fignre 5.2 Proposed geological structure for Hecate depression based on gravity profile coincident with airgun/OBS line (after Stacey and Stephens, 1969). 113 She l l C a n a d a L t d . wel ls . T h e ex t reme t h i n n i n g of the vo lcan ics be low a nar row depres-s ion filled w i t h sed imen ts (30 k m d is tance on F i g u r e 5.1) also suppo r t s th is premise. T e r t i a r y vo l can i cs are observed on the Q u e e n C h a r l o t t e Is lands to uncon fo rmab l y over l ie C r e t a c e o u s sed iments . T h e Cre taceous sed imen ts and the Ter t ia ry vo lcan ics m a y a p p e a r together or w i t h e i ther one absent . There fo re the low ve loc i t y C re taceous sed imen ts observed to occu r on ly at the eastern end of the profi le is not incons is tent w i t h ava i lab le i n f o r m a t i o n . T h e l owe rmos t un i t can only be desc r i bed in te rms of i ts presence as a comp lex r i s ing s t ruc tu re at the eastern edge of the bas in . T h e g rav i t y i n te rp re ta t i on (Stacey and S tephens , 1969) is very s im i l i a r to the gross s t r uc tu re for the ve loc i t y m o d e l . T h e g rav i t y i n te rp re ta t i on y i e l ded a s imple two b lock m o d e l for the Heca te depress ion . T h e bas in in f i l l was mode l l ed w i t h a dens i ty con t ras t o f -0.5 g / c m 3 wh i le the wedge- l ike b lock has been m o d e l l e d as h a v i n g a dens i ty con t ras t of +0.3g/cm3 w i t h the su r round ing rocks . T h e resu l ts of th is g rav i t y survey comb ined w i t h those f rom the re f rac t ion mode l l i ng m a y be i n d i c a t i n g geolog ica l features w h i c h a r r i ve f r om the co l l is ion of A l e x a n d e r and W r a n g e l l i a w i t h the con t i nen ta l m a r g i n . O n the basis of th is s u p p o s i t i o n , the lowermost un i t of the m o d e l has ten ta t i ve l y been desc r i bed as p lu tons . T h e low ve loc i t y layer may be represen ta t i ve of the post suture assemblage of Y o r a t h a n d Chase (1981). T h i s assemblage is d iscussed in C h a p t e r I as be ing compr i sed of U p p e r C r e t a c e o u s sed iments of the Q u e e n C h a r l o t t e G r o u p . T h i s group conta ins g o o d reservo i r r o c k s , the H o n n a F o r m a t i o n , and t r a p p i n g mechan i sms for hyd roca rbons w i t h i n these un i ts themse lves and also above t h e m . T h e low ve loc i ty layer , as def ined, is l oca ted in a favou rab le pos i t ion for the a c c u m u l a t i o n of hyd roca rbons . However , the th i ckness of the o v e r l y i n g Ter t i a ry vo lcan ics cou ld make exp lo ra t ion a cost ly venture . F u r t h e r m o r e , the t h e r m a l h is tory does not seem to favour the generat ion of o i l bu t on ly 114 gas and un fo r tuna te l y no source rocks are k n o w n to ex is t in the P a l e o z o i c A l e x a n d e r Te r rane ( Y o r a t h and C a m e r o n , 1982). A re f rac t i on survey f rom the C h a r l o t t e sub-bas in ( C l o w e s and G e n s - L e n a r t o w i c z , 1985) also f o u n d ev idence for a low ve loc i ty zone i n te rp re ted as u p p e r C r e t a c e o u s sed imen ts . T h e Heca te and C h a r l o t t e sub -bas ins a p p e a r to be s im i l a r in the i r m a k e u p for the u p p e r un i ts . T h e M a s s e t vo lcan ics w h i c h o c c u r i n b o t h areas s u p p o r t the be l ie f t ha t th is vo l can i c ep isode was very w idesp read . T h e U p p e r C re taceous sed iments of the pos t - su tu re assemblage va r i ous l y appea r a n d d i sappea r i r r egu la r l y t h r o u g h o u t the Q u e e n C h a r l o t t e b a s i n . In genera l the f o rma t i ons appea r to be th i cke r in the C h a r l o t t e s u b - b a s i n , sou th of the present s tudy a rea . In C h a p t e r III, an i n d e p e n d e n t s tudy i n v o l v i n g the invers ion of re f rac t ion d a t a by the m e t h o d of wavef ie ld c o n t i n u a t i o n was u n d e r t a k e n . A segment of d a t a f rom the a i r -g u n / O B S survey , under s tudy as the ma jo r pa r t of th is thes is , was inver ted to ob ta in the o n e - d i m e n s i o n a l v e l o c i t y - d e p t h s t ruc tu re . T h e m e t h o d was found to p roduce rea-sonab le 1-D v e l o c i t y - d e p t h s t r uc tu res for these e x a m p l e s . These f ind ings wou ld suppo r t the app l i ca t i on of th is i nve rs ion m e t h o d to o b t a i n i n i t i a l ve loc i t y versus dep th est imates for use by 2 -D m o d e l l i n g schemes . F o r re f rac t ion l ines a long s t r i ke , where la tera l het-e rogene i ty is less p r o n o u n c e d , the resu l t ing ve loc i t y dep th mode l s wou ld more closely represent the subsur face s t r u c t u r e . 115 References Aki, K. and Richards, P.G., 1980, Quantitative Seismology, Theory and Methods, 1: W.H. Freeman and Company, San Francisco, 535 pp. Berg, B.C., Jones, D.L. and Coney, P.J.,. 1978, Pre-Cenozoie teetonostratigraphie ter-ranes of southeastern Alaska and adjacent areas, United! States Geological Survey Open File Report 78-1085. Berg, H.C., Jones, D.L. and Riehter, D.H., 1972, Gravina- Nutzotin belt-tectonic sig-nificance of an upper Mesozoic sedimentary and volcanic sequence in southern and southeastern Alaska, United States Geological Survey, Professional Paper 800-D, pp. D1-D24. Bessonova, E.N., Fishman, V.M., Ryaboyi, V.Z., and Sitnikova, G.A., 1974, The Tau method for inversion of traveltimes-I. Deep seismic sounding data, Geophysical Journal of the Royal Astronomical Society 3 6 , pp. 377-398. Bessonova, E.N., Fishman, V.M., Shnirman, M.G., Sitnikova, G.A. and Johnson, L.R., 1976, The Tau method for inversion of traveltimes-I I. Earthquake data, Geophysical Journal of the Royal Astronomical Society, 4 6 , pp. 87-108. Carrion, P.M., Kuo, J.T., and Stoffa, P.L., 1984, Inversion method in the slant stack domain using amplitudes of reflection arrivals, Geophysical Prospecting 3 2 , pp. 375-391 Chapman, C.H., 1978, A new method for computing synthetic seismograms, Geophysical Journal of the Royal Astronomical Society, 5 4 , pp.481-518. Chapman, C.H., 1981, Generalized Radon transform and slant stacks. Geophysical Journal of the Royal Astronomical Society, 6 6 , pp.445-453. 116 C l a e r b o u t , J . F . , 1976, Fundamentals of Geophysical Data Processing, N e w Y o r k , M c G r a w - H i l l B o o k C o . Inc., 274 pp . C l a y t o n , R . W . and M c M e c h a n , G . A . , 1981, Invers ion of re f ract ion d a t a by wave f ield c o n t i n u a t i o n , G e o p h y s i c s 46, pp . 860-868. C l o w e s , R . M . , 1985, S t u d y o f u p p e r c rus ta l s t ruc tu re below Heca te S t ra i t f rom A i r -g u n / O c e a n B o t t o m S e i s m o g r a p h d a t a , F i n a l R e p o r t , D S S C o n t r a c t file numbe r 0 6 S B . 2 3 445-4-1170, 16 pp. C l o w e s , R . M . a n d G e n s - L e n a r t o w i c z , E . , 1985, U p p e r c rus ta l s t r uc tu re of sou thern Q u e e n C h a r l o t t e B a s i n f r om sonobuoy re f ract ion s tud ies , C a n a d i a n J o u r n a l of E a r t h Sc iences 22, pp . 1696-1710. D e h l e r , S. A . , 1986, A se ismic re f rac t i on s tudy of the Queen C h a r l o t t e fau l t zone, M . S c . thes is , U n i v e r s i t y of B r i t i s h C o l u m b i a , V a n c o u v e r , B . C , 112 pp . G a r m a n y , J . , O r c u t t , J . A . , a n d P a r k e r , R . L . , 1979, T rave l t ime inve rs ion : a geomet r ica l a p p r o a c h , J o u r n a l of G e o p h y s i c a l R e s e a r c h , 84, pp 3615-3622. G a z d a g , J . , 1978, Wave equa t ion m ig ra t i on w i t h the phase shi f t m e t h o d , G e o p h y s i c s , 43, pp .1342-1351. Hef f ler , P . E . and B a r r e t t , D . L . , 1979, O B S deve lopmen t at B e d f o r d Ins t i tue of O c e a n o g -raphy , M a r i n e G e o p h y s i c a l R e s e a r c h , 4, pp 227-245. H i l l h o u s e , J . W . , 1977, P a l e o m a g n e t i s m of the T r iass ic N i k o l a i G reens tone , M c C a r t h y quad rang le , A l a s k a . C a n a d i a n J o u r n a l of E a r t h Sciences, 14, pp . 2578-2598. J o n e s , D . L . , I r w i n , W . P . and H i l l h o u s e J . , 1977., Wrange l l i a — a d isp laced terrane i n no r thwes te rn N o r t h A m e r i c a . C a n a d i a n J o u r n a l o f E a r t h Sc iences , 15, pp . 2565-2575. 117 M a c k i e , D . , 1985., S u b d u c t i o n benea th the Q u e e n C h a r l o t t e Is lands? T h e resul ts of a seismic re f ract ion survey. M . S c . thes is , U n i v e r s i t y of B r i t i s h C o l u m b i a , V a n c o u v e r , B . C . , 130 pp . M c M e c h a n , G . A . and W i g g i n s , R . A . , 1972, D e p t h l i m i t s in b o d y wave inve rs ions , Geo -physieal i J o u r n a l of the R o y a l A s t r o n o m i c a l Soc ie ty , 283, pp . 459 -473 . M c M e c h a n , G . A . and O t t o l i n i , R . , 1980, D i r e c t observa t ion of a p - r cu rve in a s lant s tacked wave f ie ld , B u l l e t i n of the Se ismo log iea l Soc ie t y of A m e r i c a 70, pp . 775-789. P h i n n e y , R . A . , C h o w d h u r y , K . R . and F rase r , L . N . , 1981, T r a n s f o r m a t i o n and ana lys is of reco rd sect ions, J o u r n a l of G e o p h y s i c a l R e s e a r c h , 86, no. B I , pp . 359-377. R o n a , P . A . , 1980, G l o b a l p late m o t i o n a n d m i n e r a l resources. In T h e con t i nen ta l crust and i ts m ine ra l depos i ts . Edited by D . W . S t rangway . G e o l o g i c a l A s s o c i a t i o n of C a n a d a , Spec ia l P a p e r 20, pp . 607-622. S h o u l d i c e , D . H . , 1971, Geo logy of the western C a n a d i a n con t i nen ta l shelf. B u l l e t i n of the C a n a d i a n A s s o c i a t i o n of P e t r o l e u m G e o l o g y , 19, pp . 405-436. 1973, Wes te rn C a n a d i a n con t i nen ta l shelf. In Fu tu re p e t r o l e u m prov inces of C a n a d a . Edited by R . G . M c C r o s s a n . C a n a d i a n Soc ie t y of P e t r o l e u m Geo log i s t s , M e m o i r 1, pp . 7-35. S p e n c e , G . D . , W h i t t a l l , K . P . and C l o w e s , R . M . , 1984, P r a c t i c a l syn the t i c se ismograms for la te ra l l y v a r y i n g m e d i a ca lcu la ted by a s y m p t o t i c ray theory. B u l l e t i n of the Se ismolog iea l Soc ie ty of A m e r i c a , 74, pp . 1209-1223. S tacey , R . A . , 1975, S t r u c t u r e o f the Queen C h a r l o t t e B a s i n . In C a n a d a ' s con t inen ta l marg ins and offshore p e t r o l e u m e x p l o r a t i o n . Edited by C . J . Y o r a t h , E . R . P a r k e r , and D . J . G l a s s . C a n a d i a n Soc ie ty of P e t r o l e u m G e o l o g i s t s , M e m o i r 4, pp . 723-741. 118 S t a c e y , R . A . and S tephens , L . E . , 1969, A n i n te rp re ta t i on of g rav i t y measuremen ts on the west coast of C a n a d a , C a n a d i a n J o u r n a l of E a r t h Sc iences , 6, pp . 463-474. S u t h e r l a n d B r o w n , A . 1968. Geo logy of the Queen C h a r l o t t e Is lands. B r i t i s h C o l u m b i a D e p a r t m e n t o f M i n e s and P e t r o l e u m resources, B u l l e t i n 54, 226 pp; T i p p e r , H . W . , a m d C a m e r o n , B . E . B . , 1980, S t r a t i g r a p h y a n d pa leonto logy of the U p -per Y u k o n F o r m a t i o n ( J u r a s s i c ) i n A l l i f o r d B a y syne l ine , Queen C h a r l o t t e Is lands, B r i t i s h C o l u m b i a , in C u r r e n t R e s e a r c h , par t C . Geo log i ca l Su rvey of C a n a d a , P a p e r 80- IC , pp . 37-44. V a n der V o o , R . a n d C h a n n e l , J . E . T . , 1980, P a l e o m a g n e t i s m in oregenic be l ts . Rev iews of G e o p h y s i c s and Space P h y s i c s , 18(2), pp . 455-481. V a n der V o o , R. , Jones , M . , G r o m m e , C . S . , Iber le in , G . D . a n d C h u r k e n , J r . , M . , 1980; P a l e o m a g n e t i s m and n o r t h w a r d dr i f t o f A l e x a n d e r te r rane, southeastern A l a s k a . J o u r n a l of G e o p h y s i c a l R e s e a r c h , 8 5 , pp . 5281-5296. W e n z e l , F . , S to f fa , P . L . , and B u h l , P . , 1982, Se i sm ic m o d e l l i n g in the d o m a i n of inter-cep t t i m e a n d ray pa rame te r , I E E E T ransac t i ons on A c o u s t i c s , Speech and S igna l P r o c e s s i n g , A S S P - 3 0 ( 3 ) , p p . 406-422. Y o r a t h , C . J . a n d C a m e r o n , B . E . B . , 1982, O i l off the west coast? G E O S , 1 1 , pp . 13-15. Y o r a t h , C . J . and C h a s e , R . L . , 1981. Tec ton ic h is tory of the Q u e e n C h a r l o t t e Is lands and ad jacen t areas — a m o d e l . C a n a d i a n J o u r n a l of E a r t h Sc iences, 1 8 , pp . 1717-1739. Y o r a t h , C . J . and H y n d m a n , R . D . , 1983, Subs idence and t h e r m a l h istory of the Queen C h a r l o t t e B a s i n . C a n a d i a n J o u r n a l o f E a r t h Sc iences, 2 0 , p p 135-158. Y o u n g , I .F., 1981, G e o l o g i c a l deve lopmen t of the western marg in of the Q u e e n C h a r l o t t e B a s i n . M . S c . thes is , U n i v e r s i t y of B r i t i s h C o l u m b i a , V a n c o u v e r , B . C . 119 Appendix I Summary of the Formations for the Queen Charlotte Region 120 PERIOD EPOCH/ STAGE GROUP OR FORMATION LITHOLOGY .RAX I MUM H1CKNES! (METERS? TECTONIC OR DEPOSITIONAL ENVIRONMENT INTRUSIVE ROCKS QUATERNARY RECENT Alluviu »— •az P1FIST0CENF Cao« Ball Formation T i l l , sand, s i l t , cl4y_. PLEISTOCENE or PLIOCENE PLIOCENE ? Hl-Ul MIOCENE LOWER' MIOCENE OLIGOCENE EOCENE PALEOCENE ? Tow Hill Sil ls Fn. Masset Fin. Olivine basalt . Calcareous Ss; Dana; Facies Kootenay Fades Sandstone, slltstone] 1800* Conqlomerate: Pyroclastic brecclasTisoQ^ volcanic Ss,pomhvr> Rhyolite tuffs»flows" ows' dacite; basalt flows 1200+ Basalt Mbr. Basalt flows, pyro-clasdics, andesite 1500+ Rhyolite Mbr Rhyolite, ash flows basalt flows 2100 Mixed Mbr. Skldegate Formation Basalt breccias & flows 2000 Slltstone, sandstone 600+ Near-shore, marine and! non-marine: Mantle plume ?' Rifting ? Divergent wrenching ? Shallow marine BAJOCIAN TOARCIAN PLEINSBACHIAN S1NEMURIAN HETTANGIAN NORIAH KARNIAN OR POWSYLVAHfAH Yakoun Fm. C Mbr. Andesitic: agglome-rates and tuffs 290 B. Mbr. Shale. Ss, tuffs A Mbr. Calcareous anu lapi l l i tuffs 30+ 200 Maude Formation Shale,, sandstone 225' Kunga Fm. Black Argil-l i te Mbr-. ArgUllte.sltstone, shale, Ls . , Ss. 580 Black Lime-stone Mbr. Carbonaceous lime-stone, argill ite 270 'Grey Llme-:. stone Mbr. Limestone- ISO Karmutsen Formation Sicker Group ? Basalt flows & pi l -lows, tuffs, minor Limestone, shale, basalt, diabase 4.300 inot ex-posed ?) Non-marine and near-shore marine Glacial marine and non-marine Post Tectonic Plutons Middle Miocene -Upper Eocene, Ridge, jubduction V Divergent wrenching 7 Syntectonic Bathollths Lower Cretaceous (?)-Upper Jurassic Collision event Volcanic arc Marine Marine Marine Ocean crust ? Volcanic are ? Injterarcbasln Shallow marine Volcanic arc ? Table I'.l Table of formations for Queen Charlotte Islands (Young, 1981). 121 A p p e n d i x I I H o r i z o n t a l a n d H y d r o p h o n e C o m p o n e n t D a t a for 1983 A i r g u n / O B S S u r v e y Figures II.I - 11 .12 R e c o r d sect ions for the h o r i z o n t a l and hyd rophone componen ts for O B S s 1 to 4 p l o t t ed us ing the same p a r a m e t e r s as those in C h a p t e r I V (see C h a p t e r I V for e x p l a n a t i o n ) . S o u r c e / R e c e i v e r D i s t a n c e ( k m ) Figure II.1 20 - 1 8 - 1 6 - 1 4 - 1 2 - 1 0 - 8 - 6 - 4 - 2 0 S o u r c e / R e c e i v e r D i s t a n c e ( k m ) Figure II.2 Figure II.4 26 -24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 Source/Receiver Distance (km) F i g u r e II.6 S o u r c e / R e c e i v e r D i s t a n c e ( k m ) Figure II. 7 S o u r c e / R e c e i v e r D i s t a n c e ( k m ) Figure II..8 S o u r c e / R e c e i v e r D i s t a n c e ( k m ) Figure II.9 Source /Rece iver Distance (km) Figure 11.10 S o u r c e / R e c e i v e r D i s t a n c e ( k m ) Figure 11.11 Source/Receiver Distance (km) F i g u r e 11,12