The Open Collections website will be unavailable July 27 from 2100-2200 PST ahead of planned usability and performance enhancements on July 28. More information here.

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Crustal structure near Explorer Ridge : ocean-bottom seismometer results parallel to Revere-Dellwood… Cheung, Henry P. Y. 1978

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

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

Item Metadata

Download

Media
831-UBC_1978_A6_7 C34.pdf [ 5.85MB ]
Metadata
JSON: 831-1.0052919.json
JSON-LD: 831-1.0052919-ld.json
RDF/XML (Pretty): 831-1.0052919-rdf.xml
RDF/JSON: 831-1.0052919-rdf.json
Turtle: 831-1.0052919-turtle.txt
N-Triples: 831-1.0052919-rdf-ntriples.txt
Original Record: 831-1.0052919-source.json
Full Text
831-1.0052919-fulltext.txt
Citation
831-1.0052919.ris

Full Text

No Title PageIn presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British-Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of (jJLApJWSlLS AtAfiAAJtUj The University of Brit ish Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 ^ t e /jwptf U,<ni i i ABSTRACT An 80 km seismic r e f r a c t i o n l i n e was recorded on an array of three ocean bottom seismometers located 5 km west of the northern t i p of Explorer ridge and p a r a l l e l to Revere-Dellwood fracture zone on the P a c i f i c plate., One reversed and two split-spread p r o f i l e s have been obtained. The combined use of rotated SV component and polarization f i l t e r e d record sections enabled i d e n t i f i c a t i o n and timing of the refracted S-wave on most sections. The t r a v e l time - distance r e l a t i o n for both P and S waves i s interpreted i n the intercept time (tau) and ray parameter domain using the technigue introduced by Bessoncva et a l . (1974). This enables application of tau inversion to give extremal bounds f o r velocity-depth curves. A l i n e a r i z e d inversion technigue i s applied to give the smoothest velocity-depth p r o f i l e s consistent with the t r a v e l time data. Amplitude analysis using disk ray theory synthetic seismograms further refine the P-wave velocity-depth models. The f i n a l P- and S-«ave velocity-depth p r o f i l e s show a general increase of velocity with depth and no d i s t i n c t s t r u c t u r a l d i s c o n t i n u i t i e s . A normal oceanic c r u s t a l thickness of approximately 6.5 km and an anomalously low Pn velocity of 7.3 km s e c - 1 are interpreted. The existence of an abnormally thick crust (8-10 km}., on the opposite side of the ridge i n Explorer plate, determined i n other s t u d i e s , contrasts markedly with the res u l t s of t h i s research. Such a contrast lends support to the proposal that the complex structure and thick crust are tfce result of compressive interaction between the ycung, small Explorer plate and the older, larger North America plate. Values of Poisson's r a t i o i n the range of 0.25 to 0.32 are determined for the c r u s t a l material but better resolution of the velocity-depth p r o f i l e s i s reguired before a meaningful geological interpretation can be made. iv TABLE CF CONTENTS Abstract i i Table cf Contents i v L i s t of Tables v L i s t of Figures v i Acknowledgements v i i i I J L IMlfiDOCTigN — — • 1 1. 1 Tectonic setting of the study area — — — 3 1.2 Outline of study -~ — •.—• 7 2JL PATA ACQUISITION AND PBELIHIliASY PROCESSING — 9 2.1 Instrumentation • —• 9 2.2 F i e l d data c o l l e c t i o n — 11 2.3 D i g i t i z a t i o n and demultiplexing — 12 2.4 Time determination and correction — — - ; — 14 2.5 Amplitude correction ~ -v— 16 2.6 Shot to OBS distances — 18 3. DATA ANALYSIS AND RECORD SECTIONS — 20 3.1 Reverberation problem — — - 20 3.2 EEflODE f i l t e r — — — — • * ~ ~ — 23 3.3 P r i n c i p a l component analysis -; •• — 25 3.4 Application to data —------ - — 27 3.5 fiecord sections — — — — • • — 34 iU TRAVEL TIME. INVERSION AND SYNTHETIC SETS HOGBAMS -- 45 4.1 Tau inversion •—— • 45 4.2 Linearized inversion • < — - — 49 4.3 Synthetic seismograas • : 53 5JL liSCJUSSION • • — 6 2 j§i SOJH AS Y — ~— — •—r —: 7 1 References — . . . 73 V LIST OF TABLES Table Page 3.1 Eeverberati.cn freguency as a function cf sediment thickness •—• : 20 3.2 Angles alpha and beta from p a r t i c l e motion analysis • — ~ 30 5.1 Average S-wave sediment v e l o c i t i e s and Poisson's r a t i o s from time differences i n PPP and PPS phases 70 v i Figure Page 1.1 A r t i s t i c conception of a general tectonic picture f o r Canada's west coast — — — 2 1.2 Major plate boundaries i n northeastern P a c i f i c :—• • --- 5 1.3 Location of study area • •—•—-—— 6 2.1 OBS system .-• : 10 2.2 Examples of some i n i t i a l data ; 15 2.3 Sediment structure along r e f r a c t i o n p r o f i l e — • — •—•- 17 3.1 A t y p i c a l power spectrum of 0BS1 • — — 21 3.2 Seismogram rotation diagram •— ;- 29 3.3 Seismic signals before and a f t e r f i l t e r i n g 31 3.4 Portions of CGMA2D f i l t e r e d record section of OBS1 (SV component) — 33 3.5 V e r t i c a l component bandpass f i l t e r e d record section of 0BS1 — • — 35 3.6 V e r t i c a l component bandpass f i l t e r e d record section of 0BS2 leg 1 — — : — * — — • 36 3.7 V e r t i c a l component bandpass f i l t e r e d record section of 0BS2 leg 2 — — — — 37 3.8 V e r t i c a l component bandpass f i l t e r e d record section of 0BS3 38 v i i 3.9 Rotated SV component bandpass f i l t e r e d record section of 0 B S 1 — • • ~—------•— 39 3.10 Rotated SV component bandpass f i l t e r e d record section of OBS2 leg 1 • « 40 3.11 Rotated SV component bandpass f i l t e r e d record section of OBS2 leg 2 — - ~ 41 3.12 Rotated SV component bandpass f i l t e r e d record section of 0BS3 ' - - - r — - —.• 42 4.1 Extremal bounds of velocity-depth p r o f i l e s • — — — — 50 4.2 Velocity-depth p r o f i l e s from l i n e a r i z e d inversions -- •— — • 50 4.3 DRT synthetics of OBS1 — i — •— 57 4.4 DRT synthetics of 0BS2 leg 1 • 58 4.5 DRT synthetics of OBS2 leg 2 — 59 4.6 Final P-wave velocity-depth p r o f i l e s • •- 60 5.1 Poisson's r a t i o as a function of depth — — — 6 8 v i i i ACKNOWLEDGEMENTS I wish to take t h i s opportunity to express my deepest appreciation to Dr. R.M.Clowes; without his i n i t i a t i o n , encouragement, guidance and long hours of discussion, t h i s project could not have been completed so smoothly. F r u i t f u l ideas from Dr. R.D.Hyndman were invaluable. Helpful discussions with Dr. D.H.Oldenburg and Dr. T.J.Ulyrch are gr a t e f u l l y acknowledged. Discussions with Mr. G.A.McMechan and Mr. A.J.Thorleifson were useful and stimulating. I also wish to thank Dr. E.E.Davis, Dr. /T.Lewis, Dr. C.R.B.Lister, Mr. D.L.Barrett, Mr. M.N.Bone, Mr. S.Lynch, Mr. G.C.Rogers, Mr. D.Seemann, Mr. 0. S. Wade and again to Dr. B.p.Hyndman for th e i r invaluable contribution i n c o l l e c t i n g the data. Further, I wish to express my gratitude to the o f f i c e r s and crew on board C.F.A.V, ENDEAVOUR for their assistance during the cruise., Explosive experts provided by the Fleet Diving o n i t . P a c i f i c Maritime Command, Esguimalt B.C. are g r a t e f u l l y acknowledged. Dr. R.A.Wiggins provided a copy of his computer routines HRGLTZ and MDLPLT. Financial support for t h i s project was provided by a research contract and research agreement from the Earth Physics Branch, Department of Energy, Mines and Resources and by the National Research Council of Canada operating grant A7707. Additional funding was obtained from Mobil O i l Canada Ltd., S h e l l Canada Resources Ltd., and chevron Standard Ltd., 1 l i I N I S O D O C T I O N Because of i t s i n t e r e s t i n g and complex tectonic environment, the region o f f Canada*s west coast has continuously attracted the attention of many earth s c i e n t i s t s {Vine and Matthews,1963; Vine and Wilson,1965; Vine,1966; McKenzie and Parker,1967; Tofcin and Sykes,1968; &twater,1S70; Silver,1971; Crosson,1972; Srivastava,1973; Stacey,1973; Barr and Chase,1974; Chase et al.,1975; Riddihough and Hyndman,1976; Dickinson,1976; Riddihough, 1977) . An inf e r r e d trench, spreading centres and transform faults are a l l present; these outline the boundaries of the four active plates (see Figures 1.1 and 1.2)., The two smaller plates. Explorer and Juan de Fuca, wedged between the two larger plates. P a c i f i c and North America, have reoriented themselves in the l a s t 10 m.y. (Riddihough,1977). Of p a r t i c u l a r importance i n t h i s study i s the adjustment of the en-echelon spreading ridges, and probably also the small plates, to the approaching trench., Previous deep seismic sounding surveys i n the region have revealed complex c r u s t a l structure i n Winona basin to the northeast of Explorer ridge (Lynch,1977; Thcrleifson,1978) and on Explorer plate east of the ridge (Malecek and Clowes,1978). The complex structures are inferred to be the r e s u l t of interaction with the North America plate., The guestion of whether or not the western flank of Explorer ridge has a normal cr u s t a l structure led to the 1976 deep seismic sounding survey using ocean-bottom 2 Figure 1.1 A r t i s t i c conception of a general tectonic picture for Canada's west coast (after Riddihough and Carnes, unpublished work). 3 seismometers (OBS's).,. •1. 1 Tectonic setting of the study area : According to Atwater (1970), the tectonic setting of the North P a c i f i c 80 a, y. . ago included a large ridge/transform f a u l t system s i m i l a r to that i n the present South A t l a n t i c . Two ancient oceanic plates, the Kula and Farallon, and part of the P a c i f i c plate, a l l of which were associated with the ridge system, have been overridden by the North American plate. Unstable t r i p l e point junctions (McKenzie and Morgan,1969) have led to both the activation and a n n i h i l a t i o n of trenches and transform f a u l t s with the accompanying reorientation of plate motions and spreading directions. The Juan de Fuca plate i s now believed to be a remnant of the subducted Farallon plate. Recent detailed re-examination of the magnetic anomaly patterns in the north-eastern P a c i f i c by Riddihough (1977) has demonstrated that the northern end of Juan de Fuca plate has been moving independently with respect to i t s southern associate for the l a s t 7 m.y. Consequently, the name Explorer plate i s designated to distinguish i t from the o r i g i n a l Juan de Fuca plate. In fact, the presence of the Nootka f a u l t zone as the boundary of the two plates was confirmed during a recent geophysical cruise (R. D. Hyndman, personal communication, 1977). The major tectonic features off Canada*s west coast are summarized in Figure 1.2. Based on various geophysical 4 findings, Riddihough and Hyndman (1976) have convincingly argued that subduction i s presently taking place along the west coast of B r i t i s h Columbia south of 50°N. The Queen Charlotte transform f a u l t between the P a c i f i c and the North America plates i s presently active and well defined north of 51°N. The plate boundary between 50°N and 51$N probably i s associated with obligue subduction (Riddihough,1977; Thorleifson,1978). To the west, the boundary of Explorer plate i s well defined by a s e r i e s of fracture zones and spreading centres (see Figure 1.2). The transform f a u l t extending northwest from Dellwood knolls and the associated Tuzo Wilson seamounts (formerly J . Tuzo Wilson knolls) are seismically active. As well, the fresh volcanic rock dredged frcm Tuzo Wilson seamounts also indicates that t h i s feature i s an active spreading centre (Chase,1977). Explorer ridge and Revere-Dellwood fracture zone form the major part of the western boundary of Explorer plate. Bathymetrically, Explorer ridge i s poorly outlined; the locus of the spreading centre was o r i g i n a l l y defined by i t s magnetic signature. Figure 1.3 shows the interpreted ridges as the stip p l e d region overlain on a bathymetric map. The southern end of the ridge is c l e a r l y traceable, whereas the northern end bifurcates into two poorly defined segments which have a trough-like character. Based on a recent seismicity study, Hyndman et a l . (1978) conclude that both segments of the ridge are presently active although the northwest segment i s believed to be responsible for most of the spreading. 1 3 2 ° 1 2 8 ° 124° Figure 1.2 Major p l a t e boundaries o f f Canada's west coast. Arrows sbow d i r e c t i o n s of pl a t e motions r e l a t i v e to the America p l a t e ( a f t e r Riddihough,1977). The square block outlines the study area shown i n d e t a i l i n Figure 1.3. 6 Figure 1.3 Location of the three OBS's and the seismic r e f r a c t i o n p r o f i l e off Explorer ridge. The Interpreted ridge areas are stippled. Bathy-metric contours are i n meters (from T i f f i n and Seemann,1975). 7 The Revere-Dellwocd fracture zone i s a topographic low extending northwest from the ridqe. The seismic re f r a c t i o n p r o f i l e which forms the basis of t h i s study i s located approximately p a r a l l e l to t h i s fracture zone, on the P a c i f i c plate, and i s separated from the transform f a u l t by a series of seamounts. Figure 1.3 shows the locations of the three OBS*s and the 80 km l i n e along which explosives were detonated. Other than the work of Malecek and Clowes (1 978) , centred at 49030* and 130°W on Explorer plate, no detailed geophysical study i n the region of Explorer ridge has been published. The present OBS experiment extends seismic re f r a c t i o n coverage to the northwest from the ridge. Hyndman et a l . r (1978) have given a preliminary int e r p r e t a t i o n of these data. 1.2 Cutline of study : The geophysical experiments c a r r i e d out on the cruise i n the summer of 1976 can be divided into four aspects; namely deep c r u s t a l seismic sounding, a seismicity study, heat f l u x measurements and continuous seismic p r o f i l i n g . The concern of t h i s study i s the deep c r u s t a l seismic sounding survey. Hyndman et a l . (1978) have presented a simple int e r p r e t a t i o n based on the conventional layered model using f i r s t a r r i v a l t r a v e l times. Current and more sophisticated interpretation technigues (Kennett,1977) use t r a v e l times of a l l observable phases as well as amplitude information., Such methods y i e l d a 8 mere r e a l i s t i c velocity-depth model of the oceanic crust. As well, one of the main reasons for using OBS»s instead of the more t r a d i t i o n a l hydrophone or sonbbuoy systems i s the desire to obtain accurate information on both the P-wave and the S-wave vel o c i t y structures by using 3-component recordings. Then, Poisson* s r a t i o can be estimated and t h i s leads to speculation concerning the geological properties and o r i g i n of the oceanic crust. , The following i s a brief description of procedures used in processing and interpreting the data., Details are given i n subseguent chapters. The o r i g i n a l seismograms recorded on slow speed, d i r e c t mode tape recorders had to be transcribed to FM tape for subseguent d i g i t i z a t i o n . , Time and amplitude corrections are applied to account for water and sediment thickness variations and di f f e r e n t charge size s . Travel time information i s derived from record sections and then inverted, using the tau inversion technigue of Bessonova et a l . 41974,1976), to give an envelope of a l l possible v e l o c i t y -depth models consistent with the t r a v e l times for both P and S waves. A l i n e a r i z e d inversion technique described by Johnson and G i l b e r t (1972) i s further employed to give a best f i t model for the t r a v e l time data.. F i n a l l y the t r a v e l time and amplitude content of the data are incorporated into an interpretation using disk ray theory synthetic seismograms (Wiggins,1976).„ With t h i s procedure, the aim i s to obtain a detailed seismic model of the crust and upper mantle i n the region immediately northwest of Explorer ridge. 9 2 A , BAT A ACQUISTION ASP JPHEL1II-INA.BY PROCESSING 2,1 Instrumentation ; Four OBS*s of two diff e r e n t types (two units of each type) were used i n t h i s study. Hyndman et a l . (1978) have already described the f i e l d procedure for t h i s experiment. For completeness, I w i l l outline b r i e f l y the instruments and f i e l d operations.. Both types of instruments used are of the f r e e - f a l l , pop-up class. The f i r s t type, designed by the University of Washington (Lister and Lewis,1976; Johnson et al.,1977) hereafter i s referred to as the PGC instrument; the second type, developed by the Hawaii I n s t i t u t e of Geophysics (designated as POBS by Sutton et al.,1977) has a release mechanism modified by Heffler and Locke (1977) and i s referred to as the AGC instrument. Each design d i f f e r e d completely i n shape, size and instrumentation. Only the recording aspect i s of immediate concern i n t h i s study. Both instruments record i n continuous direct AM mode using a 4 channel tape head. The PGC instrument (see Figure 2.1) records signals at a tape speed of 1 mm s e c - 1 from one v e r t i c a l and two horizontal 4-1/2 hz geopnones with a 20 hz time code i n the f i r s t channel., The o v e r a l l frequency response i s bandlimited between 2 hz and 100 hz. Signal compression by a bipolar square-rooting c i r c u i t increases the dynamic range to 80 db,, The AGC instrument records signals at a tape speed of 0.46 mm s e c - 1 , and has a dynamic range of 1 0 ' r Figure 2.1 OBS system (PGC instrument) (a) Photograph of recording electronics i n the lower hemisphere of the spherical housing. The tape transport and electronics are i n the upper part. Batteries and the three geophones are below. (b) Launching of an OBS. The sphere containing the instrumentation i s attached to a s p e c i a l l y designed concrete anchor f o r deployment. At a predetermined time, the sphere having pos i t i v e buoyancy releases from the anchor and f l o a t s to the surface. 11 40 db and a bandwidth of 2 hz to 20 hz. To detect the direct water wave a r r i v a l , the high frequency output of the hydrophone i s r e c t i f i e d into the time channel. / Signals are recorded through one hydrophone and one v e r t i c a l and one horizontal 4-1/2 hz geophones. 2.2 Fi e l d data c o l l e c t i o n : The four OBS»s were deployed along an 80 km p r o f i l e perpendicular to the northern end of Explorer ridge and p a r a l l e l to Revere-Dellwocd fracture 2one. Position accuracy of about 1 km i s attained by the use of s a t e l l i t e and Loran A navigation. Two PGC instruments (designated OBS1 and OBS2) were launched about 5 km and 35 km from the ridge axis, i n crust around 0.2 m.y., and 1.2 m.y. old. The two AGC instruments (designated 0BS3 and 0BS4) were deployed close together about 75 km and 79 km from the ridge axis, i n crust arcund 2.5 m.y. old. Unfortunately QBS4 f a i l e d to return and was l o s t . Figure 1.3 shows the locations of the three recovered OBS's and the seismic re f r a c t i o n p r o f i l e . while crui s i n g at a speed of 4.32 m s e c - 1 (8.4 k t ) , 2300 kg of high velocity explosives were detonated i n 81 alternating large and small shots ranging from 4.5 kg to 180 kg i n s i z e . This shooting pattern was designed with the expectation that the large charges provide s u f f i c i e n t energy to distant OBS*s whereas the smaller charges would not overload the OBS's nearby. The charges were a l l time-fuse detonated at predetermined optimal depths (Shor,1963) assuming 12 a constant sinking rate of 1 m seer 1. , The ditch time {time frcm dropping of charge overboard u n t i l detonation) was noted. Direct water wave a r r i v a l s , detected with a towed hydrophone and a low s e n s i t i v i t y gecphone on the ship's deck, and the HHVJB radio time code were recorded on FM tape. As well, a high speed two-channel chart recorder was used to monitor the dire c t water wave a r r i v a l s and SWVB. A l l shots were recorded on the three OBS*s, re s u l t i n g i n one reversed p r o f i l e {OBS1 and GBS3) and two sp l i t - s p r e a d p r o f i l e s (0BS2, l e g l and leg2) which overlapped the longer ones. The p r o f i l e 0BS2 l e g l i s the northwest trending l i n e to 0BS3 and the p r o f i l e 0BS2 leg2 i s the corresponding southeast trending l i n e to 0BS1. In addition to the explosive charges, a 16 l i t r e {1000 cu in) airgun was employed to run 30 km p r o f i l e s over each OBS s i t e in d i r e c t i o n along and across the l i n e . Interpretation of these data i s beyond the scope of this study. 2.3 D i g i t i z a t i o n and demultiplexing : The data set f o r t h i s project was o r i g i n a l l y recorded i n direct continuous AM mode (see section 2.1). A method to transcribe the data into d i g i t a l form i s e s s e n t i a l and the treatment used depends on the i n i t i a l recording instruments (types PGC and AGC) since the recording speed of each i s d i f f e r e n t . The only analog playback system currently available was a Hewlett-Packard Model 3960 tape recorder. I t s slowest playback speed of 15/16 ips, representing a r e a l time speed up of about 24 times, gives a time base which i s beyond 13 the present analog-to-digital conversion c a p a b i l i t y of the marine d i g i t a l acquisition system i n the department (Clowes,1977). In order to circumvent t h i s d i f f i c u l t y , a two-stage procedure of playback and recording was followed. For the PGC instruments, the data i s f i r s t transcribed from AM to FM at a playback speed of 15/16 ips (approximately 24 times r e a l time) and a recording speed of 15 i p s . The FM tape was then replayed at a speed of 15/16 i p s , r e s u l t i n g in an o v e r a l l change of 1.5 times re a l time, for d i g i t i z a t i o n at a rate of 312.5 samples sec- 1.. This gives a Nyguist frequency of 104.5 hz i n re a l time. The equivalent r e a l time sampling rate of about 0.0048 sec just barely avoids possible a l i a s i n g problems associated with the 80 hz high cut (24 db o c t _ l r o l l o f f ) analog f i l t e r employed. , Thus, i t i s recommended that a lower corner freguency for the high cut analog f i l t e r should be used, but the real-time d i g i t i z a t i o n rate of 209 samples s e c - 1 i s adequate for inte r p r e t i v e purposes. The data tape of the AGC instrument was transcribed to FM mode at a playback speed of 16 times real time and a recording speed of 15 ips by Mr. D. L. Barrett of the A t l a n t i c Gecscience Centre. Replaying the FM tape at a speed of 15/16 i p s , resulting i n a f i n a l rate of 1.0 times r e a l time, and d i g i t i z i n g at a rate of 312.5 samples seer - 1 gives a Nyguist frequency of 156 hz i n r e a l time. No analog a l i a s i n g f i l t e r i s applied as the data are o r i g i n a l l y bandlimited between 2 hz and 20 hz. A l l the data tapes from the PGC and AGC instruments are 14 edited before d i g i t i z a t i o n . Overall, the data recorded fay the P€C instruments are of higher quality than the AGC instruments both in terms of lower ambient noise and a les s severe reverberation problem. The records are d i g i t i z e d from roughly 5 seconds before the f i r s t a r r i v a l to about 5 seconds a f t e r the d i r e c t water waves. The d i g i t a l tapes are then demultiplexed and written on new tapes. Figure 2.2 shows some examples of the raw data. I t displays s i g n i f i c a n t reverberation after the onset of the f i r s t a r r i v a l , a phenomenon common to many OBS seismograms. This particular problem w i l l be considered further i n Chapter 3. 2.4 Time determination and correction ; Once the data are edited, d i g i t i z e d and demultiplexed, the next l o g i c a l procedure i s to compile seismograms into a record section. Before proceeding to t h i s stage, accurate orig i n time determination are reguired. As mentioned i n section 2.1, d i r e c t water wave a r r i v a l s are recorded along with the WWVB signal aboard tbe ship. Assuming a charge sicking rate of 1 m s e c - 1 f o r the measured ditch time, a mean ship speed of 4.32 m sec-* (8.4 kt), and a water velocity of 1.49 km s e c - 1 , the correct o r i g i n time of each shot can be extrapolated using the slant path approximation. By assuming a v e r t i c a l ray travel path, a l l shots are time corrected to be at the sea's surface., The estimated error of t h i s s t a t i c correction i s about 15 msec., Corrections f o r seabottcm topography and sediment 15 (3) 0BS2— 80 A = 35.7Km 6MM Z i. Direct Wolff Worn MUM H2 HI l.^i» * l f . > f ^ V » « W < ( ' < « S « < I ^ V T * > > » ' ' * * V ^ (b) OIS1-4S A 43 6K1* *|lsecjf Direct Water Wave Figure 2.2 Examples of demultiplexed d i g i t a l data plotted on a Calcomp p l o t t e r . Note the reverberations inherent i n the OBS coda after the onset of the f i r s t a r r i v a l . Ca) PGC instrument. (b) AGC instrument. 16 thicknesses are made by taking a reference water depth of 1.74 sec (1-way time), a sediment thickness of 0.18 sec (1-way time) and assuming a deep refr a c t o r {layer 3) of 6.3 km s e c - 1 . Differences from these reference values are replaced by 4.5 km s e c - 1 material {layer 2) using a refracted path approximation. The 4.5 km sec— * material and the 6.3 km s e c - 1 refractor are chosen from preliminary r e s u l t s of Hyndman et a l . (1978). Hater depth and sediment thickness are estimated from a high resolution pulser p r o f i l e along the refraction l i n e (see Figure 2.3) using a constant water velocity of 1.49 km s e c - 1 and an assumed average P-wave sediment velocity of 1.8 km see-* (Peterson et al.,1S74). This correction contributes an additional error of up to 10 msec. / Since the OBS clocks sere a l l synchronized with a standarc clock before deployment, an estimate of the clock d r i f t can be made by comparison with the same standard clock after recovery. Error for this clock d r i f t correction i s less than 25 msec over the recording period for the refr a c t i o n program. Thus, the o v e r a l l t o t a l t r a v e l time error i s estimated to be a maximum cf 50 msec. 2.5 Amplitude correction : Amplitude information i s useful as a discriminant for r e f i n i n g velocity models, since t r a v e l time data alone, as w i l l be seen i n Chapter 4, w i l l give r i s e to many possible velccity-depth curves. Thus, i t i s necessary to recover the o r i g i n a l amplitude information and compensate for the e f f e c t of instrument response, varying charge size and geometrical 17 Figure 2.3 Topography and sediment structure along the refrac t i o n p r o f i l e . 18 spreading e f f e c t s . In section 2.1, i t was mentioned that the PGC instruments use a bipolar sguare-rooting c i r c u i t to improve the dynamic range of the recording system. The f i r s t step undertaken to recover the o r i g i n a l signals of-OBS1 and 0BS2 i s simply t c apply a squaring f i l t e r . This poses no d i f f i c u l t y as the data are stored i n d i g i t a l form and the unity l e v e l f or squaring has a low value (~15mv) as determined by c a l i b r a t i o n signals. For data from OBS3, t h i s step i s not necessary. To compensate for charge size variations, the W 2 / 3 relationship between charge size (H i n lbs) and seisttic amplitudes determined experimentally by O fBrien (1967) i s applied. That i s , every trace of the OBS»s i s multiplied by a factor H-*/3. Compensation f o r the geometrical spreading e f f e c t i s based on the t h e o r e t i c a l r e s u l t s of £erveny and Ravindra (1971, p147) that head wave amplitudes decrease as 1/r 2 at large distance r . Although wide angle r e f l e c t i o n amplitudes vary as 1/r (Braile and Smith,1975) and amplitude decreases near the c r i t i c a l distance i s of the order of 1/r 3 or 1/r*, a factor of r 2 i s used on a l l r e a l and synthetic seismograms. This provides good amplitude normalization along the p r o f i l e , enabling a r r i v a l s at a l l distances to be seen c l e a r l y . 2.6 Shot to OBS distances : The f i n a l information needed i n compiling a record section i s the determination of the shot-to-OBS distances. This information was compiled for the preliminary 19 interpretation cf Hyndman et a l . (1978). ; Dr. Hyndman provided a l l distances used here. For completeness, I w i l l give a summary cf the procedures employed.. A l l d i r e c t water wave a r r i v a l s and the i r multiples, corresponding to d i f f e r e n t t r a v e l path segments through the water, are i d e n t i f i e d . Then, ray tracing and synthetic seismograms as described by McMechan (1S78) are employed to give a t r a v e l time - distance r e l a t i o n for each path using an empirically determined water velocity versus depth p r o f i l e i n the area. F i n a l l y , the constraint that the computed distances between d i f f e r e n t shots and different OBS*s should be consistent was used as a check cn these distances, giving an accuracy of about 0.3 km. 20 3^ DATA ANALYSIS AND BECOBD SECTIONS 3.1 Beverberaticn problem : Reverberations and bubble pulses are inherent problems associated with many marine seismic studies. In par t i c u l a r , SV waves trapped in the sediment layer have proven to be the dominant factor a f f e c t i n g seismograms recorded by OBS's (Lewis and HcClain,1977}. This makes picking secondary a r r i v a l s extremely d i f f i c u l t . In t h i s s ection, I w i l l discuss some i n i t i a l attempts made toward dereverberation of the OBS seismograms. Figure 3.3a on page 31 shows a t y p i c a l trace of recorded data. The reverberation stands out c l e a r l y . Note that the n c r l i n e a r i t i e s i n the waveforms are due solely to the bipolar sguare-root f i l t e r i n the OBS i t s e l f . After being processed through an inverse squaring operator to recover the o r i g i n a l signal l e v e l s , a d i g i t a l bandpass f i l t e r was applied to the r 1 : 1 • — i I L o c a t i o n | Sediment J BE observed | I | t h i c k n e s s (m) | (hz) j | , -I : - ~ | I 0BS1 | 99 ± 9 J 6.98 ± 0 . 6 0 | I I I I 1 OBS2 j 225 ± 9 J 7.38 ± 1 . 6 9 | I I I ) I OBS3 | 315 ± 9 | 7.20 ± 1.06 J I L 1__ 1 TABiE 3.1 fieverberation freguency as a function of sediment thickness. 21 OBS1-74 0.0 4.0 8.0 12.0 16.0 20.0 FREQUENCY (HZ) Figure 3.1 A t y p i c a l power spectrum of a seismic trace from 0BS1. This spectrum i s computed for the time i n t e r v a l from 0.25 sec before to 9 sec a f t e r the f i r s t a r r i v a l . The shot was d i r e c t l y over the OBS C6.=0 km). 22 data as a f i r s t step i n data enhancement. Numerical experiments using d i f f e r e n t bandwidths were tested. Corner freguencies at 5 hz and 20 hz were found to y i e l d the best re s u l t s even though considerable ringing s t i l l persisted after f i l t e r i n g . (see Figure 3.3b), Inspection of the i n d i v i d u a l power spectrum of a number of seismograms at each s i t e shows that most of the energy i s concentrated near one frequency,. Figure 3.1 shows a t y p i c a l power spectrum from OBS1. The frequency at which t h i s peak occurs s i l l henceforth be referred to as the reverberation frequency (fiF). A s t a t i s t i c a l analysis of these RF«s has been done on a l l of the recorded seismograms. As well, the inferred sediment thickness corresponding to each OBS s i t e was derived from the pulser p r o f i l e (Figure 2.3) by assuming a constant P-wave sediment v e l o c i t y of 1.8 km s e c - 1 . A comparison between the KF's and the sediment thickness under each OBS location i s made i n Table 3.1. Since the average RF's corresponding to the three OBS locations have standard deviations which overlap, no dir e c t r e l a t i o n s h i p i s i n f e r r e d . However, t h i s analysis does show that the reverberation associated with many OBS seismograms probably has a more complex or i g i n than the suggestion of an SV wave trapped i n a sediment layer. The monochromatic signature of the power spectrum suggests that dereverberation might be accomplished by using a zero phase s h i f t , frequency rejec t i o n f i l t e r (Kanasewich,1975)•, The f i l t e r i s tuned at a frequency 23 corresponding to the RF of each seismogram with adjustments on the degree of notch or sharpness. Figure 3,3b,c displays a t y p i c a l trace before and after i t has been processed by the notch f i l t e r . S i g n i f i c a n t decrease i n the reverberation i s observed and the seismic character of the trace i s improved., But the i d e n t i f i c a t i o n of secondary a r r i v a l s remains extremely d i f f i c u l t . 3.2 BEMODE f i l t e r : Another approach to enhancing S-wave a r r i v a l s and Bi n i f l i i z i n g the reverberations i s to use a po l a r i z a t i o n f i l t e r . Two types of such f i l t e r s are reported in the l i t e r a t u r e . However, they are both based on the pr i n c i p l e that P and SV motions are r e c t i l i n e a r l y polarized and thus may be enhanced r e l a t i v e to the signal generated noise and random background noise. As well, both types are time-varying, non-linear f i l t e r s . The f i r s t class of polarization f i l t e r reported i n t h i s section i s the REMODE (BEctilinear Motion DEtector) type o r i g i n a l l y developed by the Teledyne research group (Sax and Mims,1965; G r i f f i n , 1966a,b; Sax,1966). Consider the v e r t i c a l z(t) and the r a d i a l r(t) seismograms as simple unit-amplitude sinusoids with no phase difference (ie. pure r e c t i l i n e a r motion). The cross co r r e l a t i o n function G(T) with a lag T between z(t) and r(t) for a window length w centered at time t i s expressed as sin(t)sVn& + T) At (3.2-1) t - vo/2 24 which aft e r integration reduces to G(T) = ± {sinC^)sihC2t)}sin(T) (3.2-2) + 2 " { w " sin(w^cosC2t)| cos(T) If w i s large compared to unity, then G(T) ft £ c ^ C T > (3.2-3) which i s an even function. By the same token, i f 2 ( t ) and r{t) have a phase lag of 90° (ie. .pure c i r c u l a r motion), and again i f w i s large compared to unity, one arrives at the r e l a t i o n Q(T) * -f sftiCT) (3.2-4) which i s an odd function. Thus, the even part of 6{T) can be used as a f i l t e r function to convolve with the o r i g i n a l time series (Sax and Hims,1965). The even part i s used because generally the seismograms involved are contaminated with random noise and so r e c t i l i n e a r detection w i l l be a better discriminator. This type of f i l t e r i s both amplitude and phase dependent with respect to the recorded s i g n a l . To eliminate the amplitude dependence, the f i l t e r function i s multiplied by a normalisation factor defined as the square root of the product between the autocorrelations of z(t) and r ( t ) . The phase dependence i s inherent and cannot be removed. The resultant f i l t e r i s referred to as BEMODE 3 by G r i f f i n (1966a). For pure compressional motion, the p a r t i c l e t r a j e c t o r y 25 would follow the E1 component of figure 3.2c and for pure transverse (SV) motion, the p a r t i c l e motion would follow the corresponding E2 component. Thus, the BEMODE 3 f i l t e r can be shaped to a P-detection f i l t e r (designated BEMODE 3A by Griffin,1966a) or an S-detection f i l t e r (designated BEMODE 3B here) by setting the product of z ( t ) r ( t ) to zero using t h i s •particle-motion' c r i t e r i a such that z ( t ) r ( t ) > 0 for P-detection (BEMODE 3A) f i l t e r and z ( t ) r ( t ) < 0 for S-detection (BEMODE 3B) f i l t e r . Basham (1967) successfully applied a BEMODE 3A f i l t e r to land-recorded, P-wave codas as an aid i n the i d e n t i f i c a t i o n of dif f e r e n t phases of P-arrivals. l e t to my knowledge, no attempt at using a BEMODE-3B f i l t e r on marine seismic data as an aid in the i d e n t i f i c a t i o n of S-wave a r r i v a l s has been published. This fact and the iuportance of i d e n t i f y i n g S-»ave a r r i v a l s provided much of the impetus to explore the f e a s i b i l i t y of using BEMODE 3B f i l t e r s as an int e r p r e t a t i o n aid for OBS data (see section 3.4). 3.3 P r i n c i p a l component analysis : In t h i s section, I w i l l discuss the type of time-varying f i l t e r introduced by Flinn (1965) and modified by Montalbetti and Kanasewich (1970). Following Souriau and Veinante {1975), a two-dimensional analysis i s given here. For convenience, t h i s type of f i l t e r i s hereafter referred to as C0MA2B (2-Dimensional COMponent Analysis) f i l t e r . Assume for s i m p l i c i t y that two orthogonal seismometers 26 are oriented i n a fashion such that one i s p a r a l l e l to the SV p a r t i c l e motion and the other i s p a r a l l e l to the P motion. In t h i s way, the covariance matrix DP CX CX DS (3.3-1) (33-2) for a set of N observations taken over the two seismograms w i l l have a simple geometrical interpretation i n Euclidian vector space. In equation 3.3-1, DP and DS are the corresponding variances of the seismogram-recorded P and SV motion, while CX i s the covariance between BP and DS. The eigenvalues of a are 11 = (DP • DS + d)/2 and 12 = (DP * .DS - d}/2 in which d = {(DP-DS)* • 4CX 2}*/2 such that 11 > 12 > 0. The corresponding normalized eigenvectors are x1 = (cos a,cos b) x2 •= (-cos b,cos a) The components are simply d i r e c t i o n cosines between the vectors and the direction of a rotated seismogram such that cosz a = (DP - DS «• d)/2d c o s 2 b = (DS - DP • d)/2d In terms of Euclidian geometric space, x1 and x2 are in the dir e c t i o n of the major and the minor axis of an e l l i p s e with lengths equal to the square-roots of the eiqenvalues. For pure r e c t i l i n e a r motion, 12 i s close to zero while for pure (3.3-5) (35 -4 ) 27 c i r c u l a r motion 12 approaches the value of 11. Thus the f i l t e r function F1|t) = 1 .-• 12/11 would enhance a l l r e c t i l i n e a r motion. But one can further shape the C0MA2D f i l t e r into a P-detection or a S-detection f i l t e r by cascading the f i l t e r function F1(t) with another direction-oriented function F2(t) = cos b f o r P-detection f i l t e r or F3{t) = cos a for S-detection f i l t e r . The f i l t e r functions F1 (t) , F2(t) or F3(t) are a l l taken over a s p e c i f i c time window w, s i m i l a r to the REflODE f i l t e r , and these are then used as point by point gain controls on the o r i g i n a l time series. But f i r s t , these f i l t e r functions are a l l smoothed i n a smaller window in order to subdue any anomalous spikes (Hontalbetti and Kanasewich,1976) . This kind of time-varying, nonlinear f i l t e r has been applied on land seismic data (Flinn,1965; Montalbetti and Kanasewich,1970; Souriau and Veinante,1975; Spence,1976) with premising r e s u l t s but no work has yet been published on marine seismic data. 3. >i Application to data : The seismograms have to be rotated before the application of the REflODE and the C0MA2D f i l t e r s . Since a l l 0BS»s used (both PGC and AGC instruments) are of the f r e e - f a l l pep-up class, orientation of the horizontal seismometers i s not known and must be determined. F i r s t , the mean azimuthal angles of 28 approach («) of the seismic energy with respect to one of the horizontal components are determined using the r e l a t i v e amplitudes of the two horizontal seismograms over one and a half cycles cf the di r e c t P-wave a r r i v a l s . Second, the two horizontal seismograms are then resolved into the r a d i a l and the transverse components {see Figure 3.2a). , Third, the angle of incidence of the seismic ray (j3) i s found from a p a r t i c l e motion plot (between the r a d i a l and the v e r t i c a l components) using one and a half cycles of the d i r e c t P-wave a r r i v a l s . Fourth, the r a d i a l and the v e r t i c a l components are further rotated into two orthogonal components C1 and C2 or E1 and E2 (see Figure 3.2b,c). The rotation to C1 and C2 i s reguired before the application of the REMODE f i l t e r since C1 and G2 are both 45<> apart from the angle of incidence and so share approximately equal energy., Thus the even part of the cross c o r r e l a t i o n function G (T) between C1 and G2 would be maximized (see section 3.2). On the other hand, the rotation to El and E2 f a c i l i t a t e s the application of the C0MA2D f i l t e r as they are separately p a r a l l e l to the di r e c t i o n of the P and SV p a r t i c l e motions. Table 3.2 tabulates the angles « and |3 used for rotation of a l l the seismograms. The values shown are based on a s t a t i s t i c a l analysis done on a l l the p a r t i c l e motion plots from each i n d i v i d u a l seismogram. since the AGC instrument has only one horizontal geophone, i t i s taken as the r a d i a l component to be used in determining the angle (3. Both tabulated angles « and p are believed to be as accurate as (b) about transverse component (for REMODE f i l t e r ) : 6 i s the. angle of incidence. / +C1 (c) about transverse component (for C0MA2D f i l t e r ) : Q i s the angle of incidence. I+Z Figure 3.2 Seismogram rotation diagrams. 30 possible with the given data and are consistent with the r e l a t i v e amplitudes of the 3-component seismograms. However, an analysis using the preliminary layered velocity-depth model of Hyndman et a l . (1978) shows that (3 should be i n the range of 130 to 2U<>. .. Only 0BS2 has |3 l y i n g within t h i s range and even allowing for large variations in (3 determined from a layered model, the (3s of OBS1 and OBS3 are anomalous with respect to the expected value. ;. Part of the discrepancy may be due to the data analysis procedures, especially for OBS3 i n which the horizontal component was taken as the r a d i a l component. Another reason for the differences probably involves near-surface inhomogeneities which would d i s t o r t the upgoing ray paths. , After the seismograms have been rotated, the optimal parameters for the f i l t e r s must be determined. After some numerical experiments, a window length of half a second was chosen for both the BEMODE and the COMA2D f i l t e r s . The time f -J Location - 1 — 1 1 Alpha Caeg) ! Beta (deg) - i 1 j | OBS1 l 42*6 ± 2.99 1 51.3 ± 3.08 T | 0BS2 ! 44.3 ± 3.05 1 17.0 ± 2.22 J OBS3 t _J 63.4 ± 3.29 J TABLE 3.2 Angles alpha and beta from p a r t i c l e motion analysis., 31 O B S 1 - 0 1 A = 74.2Km (3) i j ORIGINAL RECORDED SIGNAL i (b) (c) (d) (e) BIPOLRR-SOLBRE RND BflNDPfiSS FILTERED I I | NOTCH FILTERED i I ROTATED SV r ! i I CtWGO FILTERED | • i I : i ( f ) ! I REMODE FILTERED ! i ppp PPS i jpss ; . — I — 1 S I C O N D Figure 3.3 Seismic signals before and after f i l t e r i n g . Amplitude for each trace i s scaled arb i t r a r i l y for display purposes. Ca) Original recorded signal. (b) Bipolar-square and bandpass f i l t e r e d . (c) Notch f i l t e r e d with a notch at 6.3 hz. (d) Rotated SV component. Ce) C0MA2D fi l t e r e d . (f) REMODE fi l t e r e d . 32 lag used for the cross c o r r e l a t i o n i n the REMODE f i l t e r and the smoothing window used for the f i l t e r functions (see section 3.3) i n the COMA2D f i l t e r are found to yie l d best r e s u l t s i f they are set at one-third of a second. Figure 3.3 displays a t y p i c a l trace before and after i t has been processed by the BEMODE (3.3f) and the COMA2D (3.3e) f i l t e r s . Both f i l t e r s have reduced the reverberation considerably, with an accompanying tradeoff of lower signal-tc-noise r a t i o . However, the enhancement of the PSS phase (refracted S-wave) i s quite d i s t i n c t . The reason for the phase difference of 180° observed between the different f i l t e r e d traces i s uncertain but may be due to the analysis program., In general, based on a number of t r i a l s on d i f f e r e n t seismograms at diff e r e n t epicentral distances, results (not shown) from the CGMA2D f i l t e r are more encouraginq than the BEMODE f i l t e r . Fiqure 3.4 shows portions cf a COMA2D f i l t e r e d record section of OBS1 data. This section can be compared with the rotated S7 component record section as shown i n Fiqure 3.9. A sinqle trace comparison i s given in Figure 3.3d,e. In qeceral, the rotated SV component data display a larger reverberation c h a r a c t e r i s t i c while the COMA2D f i l t e r e d section i s more spiky. Because of t h i s , the signal-to-noise r a t i o on the rotated SV data i s considerably higher.. For OBS2, the rotated SV record sections showed clear PSS a r r i v a l s so additional f i l t e r i n g was unnecessary., However, fo r OBS1, timing of the PSS phase required the combined use of both types of sections, especially for the larqer distances., CD CD O or o O O ' CD r-»-0 B 5 1 t_> LU CO o CDO \ • (—in-to a CD -CD CO o CD 47 48 « 44 43 42 41 40 38 SB 3? 3B 33 34 K si so 2S a n a a 24 a 22 21 a ta is r 11 is it n 11 11 10 n R gi oa 0$ 04 as 02 "20.0 24.0 28.0 32.0 36.0 40.0 44.0 48.0.52.0 56.0 60.0 64.0 68.0 72.0 76 DISTANCE (KM) Figure 3.4 C0MA2D filtered record section of 6BS1 (SV component) over the distance range 25-75 km. The data are also bandpass filtered from 5 to 20 hz. PPP line is transposed from Figure 3.5.. PSS picks shown by triangles are made using this section and that df Figure 3.9. 34 3.5 fiecord sections : Figures 3.5 to 3.8 are v e r t i c a l component, bandpass f i l t e r e d record sections compiled from the seismograms recorded by each OBS. Both the PPP phase (refracted P-wave) , and the PPS phase (converted S~wave at the sediment-basement interface) are shown. The PSS phase (refracted S-wave) appears in the rotated SV component, bandpass f i l t e r e d record sections displayed in Figures 3.9 to 3. 12. Note that the PPP a r r i v a l s shown i n Figures 3.5 to 3.8 are simply overlain onto the SV component sections. A l l seismic traces shown i n these figures are both s t a t i c s corrected and amplitude compensated as discussed i n sections 2.4 and 2.5. Timing errors cf the PPP and the PSS phases are estimated to be 10 and 50 msec, respectively. These errors, plus the error of 50 msec associated with the timing corrections (see section 2.4), are used in the tau and l i n e a r i z e d inversion schemes. The onset cf a r r i v a l s in OBS3 (Figure 3.8) could be picked confidently only to a distance of 40 km. Times fo r distances greater than 40 km are determined using the constant, observed i n t e r v a l time of 1.15 sec between the PPP and the PPS phases, and the c a p a b i l i t y of i d e n t i f y i n g the l a t t e r a r r i v a l s . Due to the more severe inherent instrumental reverberation problem of OBS3 (AGC instrument), the t r a v e l time branch of the refracted S-wave i s masked (see Figure 3.12). The i r r e g u l a r trend of the onset of the PPP a r r i v a l s i n 0BS1 Vertical component bandpass filtered (5-20 hz) record section of OBS1. Amplitude corrections.outlined in section 2.5, have been applied. Solid triangles show first arrival picks made,from this section and from an unfiltered section (not shown). The line shows the PPS phase identified on the SV component section (Figure 3.9). QB52-1 38 37 36 35 34 32 31 10 29 28 27 26 25 24 23 . 22 21 20 19 IB 17 16 15 14 13 12 11 10 09 08 07 06 05 04 03 02 ^ . 0 4 . 0 g - g 12 0 16 .0 2 0 . 0 2 4 . 0 2 8 . 0 3 2 . 0 3 6 . 0 4 0 . 0 4 4 . 0 4 8 . 0 ' " • DISTANCE (KM) Figure 3.6 Vertical component bandpass f i l t e r e d (5-^ 20 hz) record section of OBS2 leg 1. Amplitude corrections, outlined in section 2.5, have been applied. Solid triangles show f i r s t a r r i v a l picks made from this section and from an unfiltered section (not shown). The line shows the PPS phase identified on the SV component section (Figure 3.10). O N 0 B 5 2 - 2 =8 49 50 51 52 53 54 55 56 57 58 59 6061 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 .0 4 . 0 8 . 0 1 2 . 0 1 6 . 0 2 0 . 0 DISTANCE (KM) 2 4 . 0 2 8 . 0 3 2 . 0 3 6 . 0 Figure 3.7 Vertical component bandpass filtered (5-20 hz) record section of OBS2 leg 2. Amplitude corrections, outlined in section 2.5, have been applied. Solid triangles show first arrival picks made from this section and from an unfiltered section (not shown). The line shows the PPS phase identified on the SV component section (Figure 3.11). _ H | I* » M 15 18 IT 1» 18 » 11 a tt 14 O a 17 a a 3P 31 M 34 33 3» 3T * 4141 <fl < 4 ) 30 SI 3? 53 34 3B 37 38 38 61 H 63 M 63 tt G8 7 1 1 3 7 3 7 4 1 3 « 19 M 'OTO 47o O 12.0 16.0 20.0 24.0 28.0 32.0 36.0 40.0 44.0 48.0 52.0 56.0 60.0 64.0 68.0 72.0 76 U DISTANCE (KM) Figure 3.8 Vertical component bandpass f i l t e r e d C5-2Q hz) record section of OBS3. Amplitude corrections, outlined in section 2.5, have been applied. Solid triangles show f i r s t a r r i v a l picks made from this section to a distance of 40 km. Beyond 40 km, f i r s t arrival?are determined from the constant observed time interval between the PPP and the PPS phases as shown in Figure 3.12. Co ' oo X u 4T0 O 1'2.0 16.0 20.0 24.0 28.0 32.0 3G".0 40.0 44.0 48.0 52.0 56.0 60.0 64.0 68.0 12.0 76.0 DISTRNCE (KM) Figure 3.9 Rotated SV component bandpass fil t e r e d (5-20 hz) record section of OBS1. A l l traces 1 have been amplitude corrected. The PPP phase shown in Figure 3.5 has been super-imposed. PSS phase is shown by the triangles and timed from this section and that of Figure 3.4. to VD 40 ^0 4T0 O 12.0 16.0 20.0 24.0 28.0 32.0 36.0 40.0 44.0 48.0 DISTANCE (KM) Figure 3.10 -Rotated SV-component- bandpass filtered-(5-20 hz) record section of OBS2 leg 1. A l l traces have been amplitude corrected. The PPP phase shown in Figure 3.6 has been superimposed. _PPS and PSS phases are identified from this section. 0352-2 1 2 . 0 1 6 . 0 2 0 . 0 2 4 . 0 2 8 . 0 3 2 . 0 36 D I S T A N C E (KM) Figure 3.11 Rotated SV component bandpass fi l t e r e d (5-20 hz) record section of OBS2 leg 2. A l l traces have been amplitude corrected. The PPP phase._shown^  in Figure 3.6 has been superimposed. PPS and PSS phases are identified from this section. 'oTO 475 O 12.0 16.0 20.0 24.0 28.0 32.0 36.0 40.0 44.0 48.0 52.0 56.0 60.0 64.0 68.0 72.0 76.0 DISTRNCE (KM) Figure 3.12 Rotated SV component bandpass f i l t e r e d (5-20 hz) record section of 0BS3. A l l traces have been amplitude corrected. The PPP phase shown i n Figure 3.8 has been super-i > imposed. The PSS phase could not be i d e n t i f i e d . 43 0.BS1 and 0BS2 leg 2 are believed to be caused by near-surface l a t e r a l velocity inhomogeneities or str u c t u r a l changes. Such a suggestion i s consistent with the uppermost c r u s t a l structure shewn i n the pulser p r o f i l e of Figure 2.3. There are some indications of the same trends for the PSS phase but the i d e n t i f i c a t i o n of the onset i s so d i f f i c u l t that minor perturbations are smoothed over. Besides the three i d e n t i f i e d phases, no c l e a r secondary refracted or re f l e c t e d a r r i v a l s could be seen on any of the record sections. This i s important i n terms of the interpreted velocity-depth curves., Except for 0BS3, s i g n i f i c a n t increases in amplitude i n pa r t i c u l a r distance ranges are observed f o r both the PPP and the PSS phases. From the record sections, these are i n the range of 30 to 35 km f o r 0BS1, 34 to 42 km for 0BS2 leg 1 and 26 to at least 36 km f o r 0BS2 leg 2, respectively, fiowever, the amplitude increase of OBS2 leg 2 i s less pronounced than on the other two sections. Apart from the distance ranges just mentioned, amplitude variations with distance are r e l a t i v e l y smooth and simple, except for OBS3 i n which the reverberation i s p a r t i c u l a r l y severe and guite unpredictable. The sudden fade-out of amplitude beyond 55 km, and at t r a v e l times greater than the PPS l i n e , i s not v i s i b l e on the reverse l i n e of 0BS1 or the partly reversed p r o f i l e of OBS2 leg 1. No apparent reason could be found except i t may due to the i n s t a b i l i t y of instrumental response, as the hydrophone was found to operate intermittently. Therefore, no amplitude analysis has been done for the data from OBS3., 45 HJL IBAVEL TI BE^INV EESI^ 4.1 Tau inversion : The tau method (Bessonova et al.,1974), a powerful t r a v e l tiase inversion technique, i s p a r t i c u l a r l y applicable to marine seismic data. Often, the assumption of a l a t e r a l l y homogeneous velocity d i s t r i b u t i o n , which i s also a piecewise continuous function of depth with a physical lower bound velo c i t y within a low velocity zone, i s well s a t i s f i e d . / Therefore, t h i s inversion method has been incorporated by several s c i e n t i s t s (Kennett and Orcutt,1976; 0rcutt,1976; Kennett et a l . , 1977) into a standard processing procedure f o r interpretation of marine r e f r a c t i o n p r o f i l e s . The tau method i s based on the t h e o r e t i c a l r e s u l t s of Gerver and Markushevich (1966), i n which the t r a v e l time data are i n t e r r e l a t e d with a velocity depth curve V(y) through a function T(P) » where T and X are the t r a v e l time and distance from source to receiver, p i s the ray parameter, u i s the wave slowness V _ 1 ( y ) , and Y i s the depth at which the ray with parameter p turns. In equation 4.1-1, Y(p) can be expressed (Gerver and Markushevich,1966) as Yfp) (4 .1-1) o Y(p) = 4>(j) + >|i(f) (4 .1 -2 ) 46 such that 4>(p) = A- C x ( <0 „ d 9 and ^ ) = I i arc ta n.1 Ik i ; - r" The function i s the usual Berqlotz-Wiechert i n t e g r a l . The addition of the term \jf(p) i s a generalization which includes low v e l o c i t y zones; yt and y( are the depths to the top and bottom of the i - t h low velocity zone respectively. q v (>p) i s the ray parameter for the i - t h low vel o c i t y zone which produces a shadow i n the t r a v e l time data. However, i t can be shown (Eessonova et al.,1976) that equation 4.1-1 i s just Abel's equation. By a modification of equation 4.1-2, the the corresponding inverse formula can be determined to obtain to) Y(r> = ± { w i l - e - r " i T ( 4 i - 3 ) such that T(p) and 0(1) are mutually inverse functions of each other. Usinq equation 4.1-3, the extremal bounds f o r the velocity function can be constructed usinq the technique introduced by Bessonova et a l . (1974,1976). A b r i e f 47 d e s c r i p t i o n f o l l o w s . R e c o g n i t i o n o f t h e f u n c t i o n Tip) a s a s i n g u l a r s o l u t i o n o f C l a i r a u t ' s e q u a t i o n , w i t h t h e t r a v e l t i m e c u r v e T ( X ) a s a f r e e t e r m , i s i m p o r t a n t f o r t h e s t a b l e e s t i m a t i o n o f TCp) f r o m a d i s c r e t e number o f p o i n t s on t h e o b s e r v e d t r a v e l t i m e c u r v e . Prom e q u a t i o n 4 . 1 - 1 , T v e r s u s X c u r v e s f o r a s p e c i f i e d r a n g e c f p v a l u e s c a n b e c o n s t r u c t e d ( s i n c e T ( p ) i s k n o w n ) . The a b o v e - m e n t i o n e d m a t h e m a t i c a l p r o p e r t y i m p l i e s t h a t T ( X , p ) h a s a s i n g l e s t a t i o n a r y p o i n t T(p 0 ) f o r e v e r y c o n s t a n t r a y p a r a m e t e r p 0 . T h u s , by f i n d i n g a l l t h e e x t r e m a l p o i n t s o f t h e f u n c t i o n T ( X , p ) o v e r an i n t e r v a l o f p , t h e t r a v e l t i m e d a t a a r e mapped o n t o t h e T - p p l a n e . T h e a d v a n t a g e o f w o r k i n g i n t h e X~P d o m a i n r a t h e r t h a n t h e p - X p l a n e (HcMechan a n d B i g g i n s , 1 9 7 2 ) i s t h a t T i s a s i n g l e - v a l u e d , m o n o t o n i c a l l y d e c r e a s i n g f u n c t i o n o f p , w h i l e p may be a m u l t i - v a l u e d f u n c t i o n o f X . A l s o , d e r i v a t i o n o f T i n t h i s way i s n o t s e n s i t i v e t o e r r o r s i n p s i n c e a f i r s t o r d e r e r r o r i n p o n l y p r o p a g a t e s a s e c o n d o r d e r e r r o r i n X ( J o h n s o n a n d G i l b e r t , 1 9 7 2 ) a n d t h i s i s i m p o r t a n t f o r t h e c o n s t r u c t i o n o f t h e f u n c t i o n Hp) f r o m a g i v e n s e t o f t r a v e l t i m e d a t a . T h i s makes t h e e r r o r i n T(P) o f t h e same o r d e r a s t h e e r r o r i n T ( X ) , s o t h a t t h e two e x t r e m a l b o u n d s o f r ( p ) c a n be d e t e r m i n e d . A s w e l l , t h e b o u n d s o f t h e i n v e r s e f u n c t i o n 0(T) a r e c a l c u l a t e d f r o m t h e b o u n d s o f Tip) • E s t i m a t i o n o f t h e e x t r e m a l b o u n d s o f t h e v e l o c i t y - d e p t h f u n c t i o n V ( y ) i s o b t a i n e d f r o m t h e i n t e g r a t i o n o f e q u a t i o n 4 . 1 - 3 a n d c o n v e r s i o n f r o m Y(p) t o V ( y ) . F i n a l l y t h e e x t r e m a l b o u n d s a r e 48 i n t e r p o l a t e d with a c u b i c s p l i n e , these bounds, however, only l i m i t a l l p o s s i b l e v e l o c i t y s t r u c t u r e s t h a t f i t the observed t r a v e l time data w i t h i n i t s e r r o r t o l e r a n c e . Thus, any v e l o c i t y - d e p t h curve t h a t l i e s w ithin the extremal bounds does not n e c e s s a r i l y r e p r e s e n t a t r u e p r o f i l e i n the r e g i o n of i n v e s t i g a t i o n . The above f o r m u l a t i o n i s completely g e n e r a l and the presence of one or more low v e l o c i t y zones i s allowed. A low v e l o c i t y zone i n the v e l o c i t y - d e p t h curve i s r e p r e s e n t e d by a sharp jump of T a t constant p i n the T-p plane, j u s t as t h e r e i s a sharp d i s c o n t i n u i t y of X at a constant p i n the X(p) domain. The t i m e - d i s t a n c e r e l a t i o n s shown i n F i g u r e s 3.5 t o 3.12 are mapped onto i n d i v i d u a l T-p planes as d e s c r i b e d above. The a l t e r n a t e s p l i n e smoothing technigue (Kennett,1976) or the •polynomial f i t * method (Bates and Kanasewich,1976) are not used, s i n c e f o r the data o f t h i s study, the mapping to the T(p) plane i s n u m e r i c a l l y s t a b l e . No d i s t i n c t jump of T values i s observed i n any of the data; consequently, no low v e l o c i t y zone i s i n f e r r e d . The extremal bounds of the f u n c t i o n TCP) are then i n v e r t e d to give the corresponding bounds i n the f u n c t i o n Y(p) and then V ( y ) . F i g u r e U,1 shows the extremal bounds of the v e l o c i t y -depth p r o f i l e s f o r both P- and S-wave data recorded on the G£S*s, Si n c e the s e i s m i c r e f r a c t i o n experiment was not designed to sample the s u r f i c i a l sedimentary s t r u c t u r e (layer 1 and p o s s i b l y l a y e r 2), the bounds i n F i g u r e 4,1 f o r the 49 uppermost kilometer are not based on the data set. They are an extrapolation from velocity bounds at a lower depth to the water velocity at the ocean bottom. The extrapolation i s essential to the inversion scheme so as to preserve the t o t a l t r a v e l time behaviour from source to receiver., Analysis of the r e f l e c t i o n and shallow r e f r a c t i o n data recorded over the same array of OBS*s using an airgun source (see section 2.2) should provide more information on the structure of the uppermost crust. However, such a study i s outside the scope of t h i s t h e s i s . 4.2 Linearized inversion : The l i n e a r i z e d inversion method of interpreting t r a v e l time data (Johnson and Gilbert,1872) also i s very applicable to marine seismic data because the assumption of a l a t e r a l l y homogeneous velocity d i s t r i b u t i o n i s often well s a t i s f i e d . However, th i s technigue has suffered from the c r i t i c i s m that i t requires a good sta r t i n g model. With the ccuplementary tau inversion method, there i s no d i f f i c u l t y in picking such a model. In f a c t , Kennett (1976) has shown that the f i n a l model i s only weakly dependent on a l l i n i t i a l models that are within the extremal bounds. Thus, the mean of the extremal bounds i s chosen as a s t a r t i n g model for the l i n e a r i z e d inversion. Following Johnson and G i l b e r t (1972), the function T<P) i s used as the datum for the l i n e a r i z e d inversion scheme. To f i r s t order of approximation, %TiP) can be treated as a l i n e a r functional of Su (y) (Julian and Anderson, 19 68) i n equation 50 (a) VELOCITY (KM/SEC) 0.0 2.0 4.0 6.0 8.0 _i i VELOCITY 1.0 2.0 (KM/SEC) 3.0 4.0 5.0 Figure 4.1 Extremal bounds of velocity-depth, profiles for travel time data in Figures 3.5 to 3.12. Ca) P-wave. (b) S-wave. (a) VELOCITY (KM/SEC) 0.0 2.0 4.0 6.0 8.0 (b) 0.0 VELOCITY 1.0 2.0 (KM/SEC) 3.0 4.0 5.0 Figure 4.2 Velocity-depth profiles from linearized inversion of the travel time data shown in Figures 3.5 to 3.12. Ca) P-wave. Cb) S-wave. 51 4.1-1 such that ^ - 2 j c ; y > (4.2-1) Thus, iX can be expressed as a combination of kernels G.(y) and velocity functions so that Ui - J Q^Wy) ^ (4.2-2) o i n which and In equation 4.2-2, I^CP) i s the maximum depth to be considered in any velocity model. Dsing the technique described i n section 4.1, a set cf N values TiiPi) with variances 0.2 can be constructed from the observed t r a v e l time data. Further, i f 9;, are the differences between the observed t-APi ) f o r the r e a l earth and the Xtfpc) calculated from a part i c u l a r t r i a l model, and i f 9t are small 52 with respect to T^, then $Tt in equation 4.2-2 can he replaced by )L i n the hope that a minor perturbation of m(y) could bring the t r i a l model into agreement with the observations. Therefore, the condition (Johnson and Gilbert,1972) Up <^pmCu)d3 « V i * 8 t > ( 4 . 2 -3 ) i s reguired. In order to s a t i s f y eguation 4.2-3 in a le a s t square sense and simultaneously avoid the s i n g u l a r i t y i n Gi (y)* eguation 4.2-3 i s f i r s t integrated by parts. Thus Vp) \ - ® L $ J J i C ^ t n ' ^ ) ^ + J ^ o W o ) * 9i -f 6c > ( 4 2 - 4 ) such that i s a set of new kernels and the prime denotes d i f f e r e n t i a t i o n with respect to y. J£(y) can be interpreted physically as the tra v e l time from the l e v e l y to the depth at which the ray with parameter p turns., A discontinuity i n the velocity-depth curve at a depth y=c could be accounted for by the introduction of an additional term J t-(c) [ m (c+)-m (c-) ]. Thus, J• could be calculated e a s i l y for any t r i a l model. Then, using the ' f l a t t e s t * perturbation c r i t e r i a of Backus and 53 Gilbert (1969), i . e . the search for the minimum of the function subject to the constraints of equation 4.2^4, a f i n a l model i s achieved after successive i t e r a t i o n s . This f i n a l model, which has the tendency to minimize velocity gradients i s more r e a l i s t i c than the conventional layered model. Also, the f i n a l model i s more consistent with frequently observed results from recent seismic r e f r a c t i o n studies than i s the layered model; namely the presence of prominent f i r s t a r r i v a l s and the general absence of d i s t i n c t secondary a r r i v a l s , except in some instances for wide-angle r e f l e c t i o n s from the H-disccntinuity (Grcutt et al.,1976; Hhitmarsh,1S78). Velocity-depth curves derived from the application of t h i s l i n e a r inversion method to the t r a v e l time data for both P and S waves are shown i n Figure 4.2. Again for the reason given i n section 4.1, a l l velocity-depth structures i n the uppermost kilometer are not derived from the data set. The curves shown represent extrapolation from v e l o c i t i e s at a lower depth to the water ve l o c i t y at the ocean bottom. 4.3 Synthetic seismograms : The two inversion procedures discussed above are based on tr a v e l time data only. Interpretation using synthetic seismograms makes use of the additional information contained (4 .2 -5) 54 in the data, namely amplitude changes along the p r o f i l e . Such information puts a d d i t i o n a l constraints on acceptable velocity-depth models. No j o i n t inversion technique using both t r a v e l times and amplitudes has been published, but t h e o r e t i c a l work i n t h i s d i r e c t i o n i s being pursued by some researchers. In t h i s study, synthetic seismograms are calculated by the disk ray theory (DRT) approach described by Wiggins (1976), and for which t i e t h e o r e t i c a l derivation i s provided by Chapman (1976a,b). The advantages and disadvantages of DRT calc u l a t i o n s have been discussed by Malecek and Clowes (1S78) in the f i r s t published interpretation of marine seismic data using DRT synthetic seismograms. The most important aspect of the DRT method i s that i t i s computationally e f f i c i e n t and thus inexpensive. Many models can be t r i e d , f a c i l i t a t i n g the desire to obtain the best f i t of computed t r a v e l times and amplitudes with the observed seismograms., The use of the computer algorithm, HRGLTZ, (coded by R.A.Wiggins), i n which the DST computations are made can be related to the basic equation 4.1-1. In th i s equation, the T term can also be expressed {Biggins and Madrid, 1974) as I r (4.3-1) such that the equation can be re-written as 55 where p i s the maximum p value. Equation 4.3-2 relates the tr a v e l time of an a r r i v a l to the area under the X (p) curve. S i m i l a r l y , the amplitude of an a r r i v a l can be expressed as a function of p and X through the r e l a t i o n In equation 4.3-3, F(p,X) i s a slowly varying complex function cf p and X compared to the second factor, |dp/dXj 1 / 2 (Bullen,1965). Thus to a close approximation, eguation 4.3-3 relates the amplitude of an a r r i v a l to the sguare root of the slope cf the p-X curve. Because of t h i s direct r e l a t i o n of the X(p) curve to t r a v e l times and amplitudes, synthetic seismogram calculations using DfiT are best done with input i n the p-X plane rather than in the velocity-depth domain {Wiggins and Madrid,1974; McMechan,1976). To obtain i n i t i a l models, E-wave velocity-depth p r o f i l e s from the li n e a r i z e d inversion (see Figure 4.2a) are used i n the program MDLPLT also written by E.A.Biggins. This program interpolates the input velocity-depth p r o f i l e s and generates the corresponding p-x curves. These form the input data for HRG1TZ to compute the DfiT theore t i c a l seismograms for comparison with the observed OBS records. Since the input p-X curve has s a t i s f i e d the t r a v e l time r e l a t i o n i n a best f i t sense, simultaneously changing the slope of the p-X curve and keeping the area under the curve constant w i l l produce synthetic record sections with d i f f e r e n t amplitude 56 c h a r a c t e r i s t i c s but i d e n t i c a l t r a v e l times. This t r i a l - a n d -error procedure i s repeated u n t i l an acceptable f i t of the synthetic seismograms to the observed data i s attained. Figures 4.3 to 4.5 show the DBT synthetic seismograms for p r o f i l e s OBS1 and 0BS2. The source wavelet which was convolved with the calculated DBT impulse response i s given on the upper ri g h t . I t was derived from a clear f i r s t a r r i v a l on the OBS2 p r o f i l e . In Figure 4.6, the velocity-depth p r o f i l e s corresponding to the synthetic sections are displayed. Synthetic seismogram matching was not done for 0BS3 (AGC instrument) since the amplitude information was considered unreliable, as discussed in section 3.5. Also, no attempt was made toward the modelling of the PSS phase at a l l since no suitable computer algorithm could be implemented., Figures 4.3 to 4.5 compare well with the observed data (PPP phase) shown i n Figures 3.5, 3.6 and 3.7. Note p a r t i c u l a r l y that the i r r e g u l a r trend of a r r i v a l s probably caused by surface v e l o c i t y inhomogeneities cannot be modelled by DBT. As well, because the s t a r t i n g model i s chosen from the l i n e a r i z e d inversion i n t e r p r e t a t i o n , preference i s given to velocity-depth models that possess the least changes i n velocity gradient rather than st e p - l i k e increases i n ve l o c i t y (see the sketch below). SMOOTH GRW>.et*T STtsf-UKE Irtcpfeftse SOURCE VRVELET ^ . 0 8 . 0 1 6 . 0 2 4 . 0 3 2 . 0 4 0 . 0 4 8 . 0 5 6 . 0 6 4 . 0 7 2 . 0 8 0 . 0 DISTANCE (KM) Figure 4.3 Computed DRT synthetic seismograms and travel times for data of OBS1. The source wavelet convolved with the impulse response from the synthetic calculation i s also shown. Compare with..Figure 3.5. SOURCE VBVELET 12.0 16 .0 20 . 0 24 .0 28 . 0 32 . 0 36 . 0 40 . 0 44 0 48 0 DISTANCE (KM) Figure 4.4 Computed DRT synthetic seismograms and t r a v e l times f o r data of OBS2 l e g 1. The source wavelet convolved with the impulse response from the synthetic c a l c u l a t i o n i s also shown. Compare with Figure 3.6. Cn OO SOURCE VRVELET o in ^ C D CD CD I— CD CD Q C V I L L J 0 CD T o " 0 B S 2 - 2 1 S E C O N D 1 2 . 0 1 6 . 0 2 0 . 0 2 4 . 0 2 8 . 0 3 2 . 0 -36 .0 D I S T R N C E (KM) Figure 4.5 Computed DRT synthetic seismograms and travel times for data of 0BS2 leg 2. The source wavelet convolved with the impulse response from the synthetic calculation i s also shown. Compare with Figure 3.7. VELOCITY (KM/SEC) 0.0 2.0 4.0 6.0 8.0 O - l ^ 1 I I ft O ro o Figure 4.6 P-wave vel o c i t y - d e p t h curves obtained from the DRT sy n t h e t i c seismogram c a l c u l a t i o n s shown i n Figures 4.3 to 4.5. 61 The preferred model does net produce prominent secondary a r r i v a l s , consistent with the observed data. But, depending on the gradient between * layers*, the smoothly layered model may produce such phases.. However, in testing t h i s point, i t has been possible to generate a set of synthetic seismograms, s i u i l a r to those shown in Figures 4 . 3 to 4 . 5 , from smoothly layered models with small v e l o c i t y gradients. Thus, fine s t r u c t u r a l d e t a i l s of velocity-depth p r o f i l e s cannot be resolved with the e x i s t i n g data set. The P-wave velocity-depth p r o f i l e s shown in Figure 4 . 6 represent the most l i x e l y models consistent with both t r a v e l times and amplitudes of the observed seismograms. 62 £ J L D I S C I S S I O N Velocity-depth curves (Figures 4.1 and 4*2) derived from the t r a v e l time inversion technigues discussed in sections U,1 and 4.2 are based on refracted f i r s t a r r i v a l times (PPP and PSS phases) only. Thus, d e t a i l s of any s t r u c t u r a l t r a n s i t i o n s are concealed and cannot be resolved (Kennett and Orcutt, 1976; Kennett et a l . , 1977). Hith the aid of DRT synthetic seismogram modelling procedures, the P-wave velocity-depth models from l i n e a r i z e d inversion are further refined (Figure 4.6) to s a t i s f y both t r a v e l time and amplitude c h a r a c t e r i s t i c s . Since no amplitude analysis has been done to refine the S-wave velocity-depth curves af t e r l i n e a r i z e d inversion, those displayed in Figure 4.2b are considered f i n a l , though by no means unigue. Unless otherwise stated, these S-wave velocity-depth p r o f i l e s together with the DRT P-wave velocity-depth models (Figure 4.6) are used for interpretation purposes. Both the P- and S-wave models display a very high velocity gradient i n the uppermost 0.5 km of the crust with a sharp decrease at a depth of 1 to 2 km below the ocean f l o o r . As discussed before, the data do not define well the velocity structure i n the uppermost kilometer, but to reach the velocity values at greater depths, a substantial gradient must ex i s t . Recall from section 2.4 that an average sediment thickness of 324 km (0.18 sec 1-way time at a velocity of 1.8 km sec -*) i s assumed. In terms of conventional 63 interpretations which subdivide the oceanic crust into layers (e.g. Houtz,1976; Hhitmarsh,1978), the change i n v e l c c i t y gradient i s correlated with the t r a n s i t i o n from layer 2 tc layer 3. The t r a n s i t i o n i s believed to represent a compositional change from metabasalt or diabase to gabbro (Peterson et a l . , 1974) or a s t r u c t u r a l change from brecciated dikes to sheeted dikes (Salisbury and Christensen,1978). The boundary between layer 3 and the upper mantle (the H-discontinuity) i s not c l e a r l y defined on the velocity-depth curves for either P or S waves. But a close examination of figures 4.2b and 4.6 shows that a s l i g h t change i n gradient occurs at depths between 8.8 and 10 km. In pa r t i c u l a r , the P-wave model, which i s better constrained, shows a v i s i b l e gradient change at depths of 9.2 km for OBS1 and 8.8 km for OBS2 leg 2 although OBS2 leg 1 reveals no comparable change. These changes i n gradient are interpreted as representing the tr a n s i t i o n from crust to mantle, giving a sub-bottom c r u s t a l thickness i n the range of 6.5 km. On the basis of t h i s interpretation, upper mantle apparent v e l o c i t i e s l i e i n the range from 6.9 to 7.8 km s e c - 1 f o r P waves and 3.9 to 4.6 km s e c - 1 for S waves where the lower values are from OBS1 and the higher ones from OBS2 leg 2. This asymmetry i n upper mantle v e l o c i t i e s could be seen in a l l velocity-depth p r o f i l e s (Figure 4.1, 4.2 and 4.6) derived in chapter 4., In fact, the asymmetries e x i s t not only between 0BS1 and OBS2 leg 2 but between 0BS1 and a l l the other 0BS*s at depths greater than 6 km. As the analyses are based 64 on a l a t e r a l l y homogeneous model, the asymmetry i n v e l o c i t i e s suggests that the crust-mantle interface dips away from the ridge. By assuming reversed upper mantle apparent P-wave v e l o c i t i e s of 6.9 and 7.8 km s e c - 1 and a dipping M-discontinuity underlying an assumed homogeneous c r u s t a l layer of 6 km s e c - 1 , a P-wave velocity cf 7.3 km s e c - 1 for the upper mantle i s determined. The calculated dip i s approximately 10° away from the ridge. Changing the assumed velocity cf the overlying crust does not a l t e r the end r e s u l t s i g n i f i c a n t l y . A similar c a l c u l a t i o n using S-wave values y i e l d s an upper mantle velocity of 4.2 km s e c - 1 , assuming an average S-wave cru s t a l velocity of 3.3 km sec—* corresponding to a Pcisscn•s r a t i o cf 0.29 (Christensen and Salisbury,1975). Since the partly reversed p r o f i l e between OBS3 and 0BS2 leg 1 displays only a small asymmetry in v e l o c i t i e s (Figure 4.1a and 4,2a), the dipping crust-mantle interface probably i s confined p r i n c i p a l l y to the region (35 km i n length) between OES2 and the ridge axis., This suggests that the M-discontinuity dips more steeply near Explorer ridge than i t does farther from i t . The foregoing i n t e r p r e t a t i o n of c r u s t a l thickness (-v6.5 km) and upper mantle P-wave ve l o c i t y (7,3 km sec -*) l i e s within the range reported for other studies near a x i a l zones (le Pichon et al.,1965; Keen and Tramontini,1970; Talwani et al.,1971; Poehls,1974). In contrast, a recent seismic r e f r a c t i o n study over Juan de Fuca ridge (Davis et a l . * 1976) suggested an abnormal c r u s t a l thickness of 10 to 11 km. .. The c r u s t a l thickness as interpreted was due to a l o c a l i z e d low 65 velocity mantle near the spreading ridge and d i r e c t l y under the OBS's. This caused a time residual to be observed even though the t r a v e l time branch corresponding to mantle a r r i v a l s would be c o r r e c t l y representative of the apparent v e l o c i t y along the ridge flank. As a r e s u l t , the delay in intercept time would be interpreted erroneously as a deep i n t e r f a c e . They also recommended that the above explanation could be checked e a s i l y by reversing the cross p r o f i l e or simply by doing another r e f r a c t i o n l i n e on the flank. Malecek and Clowes (1978) have done both i n a seismic r e f r a c t i o n experiment over the southern Explorer ridge and on Explorer plate. However, t h e i r r e s u l t s s t i l l reguired an anomalously thick crust of 8 to 10 km. As well, Pn v e l o c i t i e s ranging from 7.8 km s e c - 1 perpendicular to the ridge to 7.3 km s e c - 1 p a r a l l e l to i t were interpreted from the reversed p r o f i l e s . These authors suggest that the abnormal c r u s t a l thickness may be the general tectonic r e s u l t of two small plates jammed between two large plates. In p a r t i c u l a r , they postulate that the cause may be due to the *bunching-up* e f f e c t of the ycung crust (<3.5 m.y.) of Explorer plate on • c o l l i s i o n * with the giant North America plate. Whether the proposed hypothesis i s correct oi not i s a subject of conjecture and future research. Bowever, the common occurrence of an anomalously thick crust in Explorer plate and possibly Juan de Fuca plate i s reguired to s a t i s f y a l l existing seismic r e f r a c t i o n studies, the two mentioned above plus those of Lynch (1977) and Thorleifscr. (1978) i n Winona basin. 66 No further attempt i s made here to explain the existence of an anomalously thick crust (8-10 km) i n Explorer plate. But the normal c r u s t a l thickness (~6.5 km) and anomalously low P-wave upper mantle v e l o c i t y (7.3 km s e c - 1 ) derived from this study on the P a c i f i c plate contrast markedly with the equivalent values on Explorer plate. This leads one to question whether the B-discontinuity has been interpreted corre c t l y i n t h i s study. If the crust-mantle interface actually occurs at a greater depth than 6,5 km and has a r e l a t i v e l y rapid t r a n s i t i o n to a normal northeastern P a c i f i c upper mantle ve l o c i t y of 8.1 km s e c - 1 (Keen and Barrett, 1971) , a prominent p o s t - c r i t i c a l l y r e f l e c t e d phase should appear as a secondary a r r i v a l i n the compiled record sections. No such phase i s v i s i b l e so the inter p r e t a t i o n as discussed i s the one most consistent with the data. Therefore, the small change i n velo c i t y gradient between 6.2 to 7.4 km below the ocean bottom may be a r e a l i s t i c representation for a gradual t r a n s i t i o n from lower crust to upper mantle. „• Also, i t i s worthwhile to point out that a l l P-wave velocity-depth models except OBS1 are within the extremal bounds derived from seismic r e f r a c t i o n studies near the East P a c i f i c Rise f o r crust of les s than 5 m.y. old (Kennett et a l . , 1977) . Besides the usual interpretation i n terms of the velocity-depth p r o f i l e s , a further attempt was made to estimate Poisson's r a t i o (<r) in the region of study. Such estimates are based on the standard r e l a t i o n (Bullen, 1965) 67 where Vf and Vs are the P- and S-wave v e l o c i t i e s . Figure 5.1 shews the calculated Poisson*s r a t i o values as a function of depth. The four d i f f e r e n t curves marked LO, HI, LN and DL represent differences i n the basic data from which they axe calculated. LO and HI symbolize Poisson l ,s r a t i o s derived from the two extremal bounds (lower and upper) of v e l o c i t i e s (see Figure 4.1); LN stands for Poisscn»s ratios calculated from results of l i n e a r i z e d inversion (see Figure 4.2); DL refers to Poisson*s r a t i o s computed from tbe P-wave velocity-depth structure derived from DBT synthetic seismograms (see Figure 4.6) and S-wave velocity-depth structure from l i n e a r i z e d inversion (see Figure 4.2b)..- Since the velocity-depth structures on which the cal c u l a t i o n of Poisson*s r a t i o was based have a low resolving power (see discussion e a r l i e r ) cn st r u c t u r a l d e t a i l s , only general discussions are warranted. For-OBS1 and 0BS2 leg 1, a l l four curves scattered widely with the l i n e LN having an upper bound c h a r a c t e r i s t i c , whereas for 0BS2 leg 2, the curves have a tendency to c l u s t e r together. While the r e l i a b i l i t y of the Poisson»s r a t i o data i s poor, they indicate a tendency to be between values of 0.25 and 0.32 i n the deeper crust and somewhat greater in the upper crust (although 0BS2 leg 1 i s an exception). Only the curve DL in 0BS1 compares favourably with laboratory data on Poisson's r a t i o derived from sampled rocks i n the Bay of 68 Figure 5.1 Poisson's r a t i o as a f u n c t i o n of depth f o r (a) 0BS1, (b) 0BS2 l e g 1 and Cc) 0BS2 l e g 2. The four curves represent c a l c u l a t i o n s from d i f f e r e n t P- and S-wave v e l o c i t y - d e p t h p r o f i l e s . LO and HI represent r a t i o s d e r i v e d from lower and upper extremal bounds of the v e l o c i t i e s ; LN from v e l o -c i t i e s determined by l i n e a r i z e d i n v e r s i o n ; DL from P-wave v e l o c i t y - d e p t h s t r u c t u r e s based on DRT s y n t h e t i c s and S-wave v e l o c i t y - d e p t h p r o f i l e s from l i n e a r i z e d i n v e r s i o n . The dashed l i n e at 2.6 km i s the water bottom. For add-i t i o n a l d i s c u s s i o n , see t e x t . 69 Islands o p h i c l i t e complex i n Newfoundland (Salisbury and Christensen, 1978). The data set presented here i s believed to be a f i r s t attempt to derive Poisson ,s r a t i o as a function of depth from OBS seismograms. Because of the s e n s i t i v i t y of £cisson*s r a t i o to changes i n the r a t i o of P- and S-wave v e l o c i t i e s , much better constrained velocity-depth curves are required before meaningful geological interpretations can be made. In Figure 5.1, no Poisson»s r a t i o was calculated f o r the uppermost 1.5 km below sea bottom since the corresponding velocity i n t e r v a l i s based on extrapolation from the data set (see section 4.2). However, through the estimation of an average S-wave sediment velocity over each OBS by using the observed constant i n t e r v a l time between the PPP and the PPS phases, and assuming a constant P-wave sediment velocity of 1.8 km sec-* (Talwani et al.,1971; Peterson et al.,1974), an average Poisson*s r a t i o can be derived. Table 5.J summarizes the r e s u l t s . The sediment thickness i s derived from the pulser p r o f i l e shown i n Figure 2.3, as discussed i n section 3.1. Using a lower average P-wave sediment velocity would decrease the tabulated Poisson's r a t i o s s l i g h t l y but not s i g n i f i c a n t l y . Noting that <f=0.5 for a l i q u i d , the high values of Poisson's r a t i o suggest a very poorly consolidated sediment. This interpretation i s consistent with conclusions made i n comparable studies (Davis et al.,1976; Lewis and McClain,1S77). More detailed information on the physical properties of the sediments (thickness and degree of 70 compaction) could prove useful for inte r p r e t i n g heat flow data and understanding the tectonic a c t i v i t y i n the area. | Site | Thickness PPP-PPS | S-wave velocity (sec) | (km s e c - 1 ) Poisson *s r a t i o J OBS1 | I i J CBS2 | I I I OBS3 j 99 225 315 0 . 60 1. 10 1. 15 -JL-0 . 1 7 0 . 2 0 0 . 2 7 0.495 0.490 0.488 TABLE 5. 1 Average S-wave sediment v e l o c i t i e s and Poisson 1s r a t i o s from time differences i n PPP and PPS phases., 71 fix SOHHAfiY Through the use of t r a v e l time inversion techniques and amplitude analyses, three-component seismograms recorded cn an array of three GBS's deployed on the P a c i f i c plate immediately west of Explorer ridge have been interpreted. The refracted S-wave a r r i v a l s are i d e n t i f i e d and timed on most sections by the combined use of the rotated SV component record sections and the p o l a r i z a t i o n f i l t e r e d data. This enables two t r a v e l tine inversion technigues, tau and l i n e a r i z e d inversions, to be applied on both P- and S-wave t r a v e l times., The tau inversion gives extremal velocity-depth bounds such that a l l models which s a t i s f y the t r a v e l time data must l i e i n s i d e . The l i n e a r i z e d inversion seeks a best f i t model within the extremal bounds under the constraint of least change i n ve l o c i t y gradient. The use of DBT synthetic seismograms further refine the resultant velocity-depth models to s a t i s f y the observed amplitude c h a r a c t e r i s t i c s . A l l f i n a l models, as interpreted, show a pronounced velocity gradient i n the upper crust, a somewhat lesser gradient i n the lower crust, and a gradual t r a n s i t i o n to the upper mantle. No v e l o c i t y d i s c o n t i n u i t i e s are required tc s a t i s f y the data. ,; A normal oceanic c r u s t a l thickness of 6.5 km and a reversed Pn velocity of 7.3 km s e c - 1 are interpreted. The l a t t e r i s r e l a t i v e l y low but i s consistent with Pn v e l o c i t i e s derived from comparable studies near a x i a l zones where young, immature lithosphere i s present. 72 From the P- and S-wave velocity-depth curves, Poisson*s r a t i o s are calculated to be i n the ranqe of 0.25 to 0.32, at a sub-bottom depth of 4 to 10 km, with the hiqher values found in the upper crust. However, resolution of the velocity-depth p r o f i l e s i s not adequate to determine Poisson*s r a t i o s such that a meaningful qeoloqical interpretation can be made., In terms of the p r i n c i p a l tectonic r e s u l t s f o r the reqion west of Explorer ridge on the P a c i f i c plate, the normal oceanic c r u s t a l thickness (and low Pn velocity) contrast markedly with the thick crust of Explorer and Juan de Fuca plates, located east of the ridge system (Halecek and Clowes,1978; Davis et al,,1976). Assuming the c l a s s i c a l theory of sea-floor spreading at a ridge, the disparate results on either side of Explorer ridge, give support to the suggestion by Malecek and Clowes (1978) that the thick crust i s the r e s u l t of compressive interaction between the young, small Explorer plate and the older, larger North America plate., I t i s hoped that i n the future, more geophysical and geological studies w i l l provide an integrated explanation of the uncertain geodynamical features i n t h i s complex tectonic region of the north-eastern P a c i f i c . REFEBE^CES Atsater,T. (1970) Implications of plate tectonics for the Cenozoic tectonic evolution of western North America. B u l l . Geol. , Soc. Amer,, V81, pp.3513-3536. Backus,G. 5 Gilb e r t , F . (1969) Construction of P-velocity models tc f i t r e s t r i c t e d sets of t r a v e l time data., B u l l . Seism., Soc. Am., 759, pp.1407-1414. Barr,S.M. 8 Chase,B.L. (1974) Geology of the northern end of Juan de Fuca Ridge and sea-floor spreading. Can. J• Earth. S c i . , V11, pp.1384-1406. Basham,P.i. (1967) Time domain studies of short period teleseismic P phases. Unpublished HSc thesis, university of B r i t i s h Columbia. 145pp. Bates,A.C. S Kanasewich,E.B. (1976) Inversion of seismic t r a v e l times using the tau method. Geophys. J . Boy. astron. S o c , V47, pp. 59-72. Bessonova,E.N., Fishman,V.M., Byaboyi,V.Z. £ Sitnikcva,G.A. (1974) The tau method for inversion of travel times I. Deep seismic sounding data. Geophys. J . Boy. astron. / S o c , V36, p p . 3 7 7 - 3 9 8 . / Fishman,V.fi., Shniraan,H,G., Sitnikova ,G.A. 6 Johnson,!.8. (1976) The tau method for inversion of t r a v e l times I I . Earthquake data. Geophys., J. Boy. astron. S o c , V46, pp.87-108. Braile,L.W. 6 Smith, R. B. , (1975) Guide to the inte r p r e t a t i o n of c r u s t a l r e f r a c t i o n p r o f i l e s . Geophys. J . Boy. astron. S o c , V40, pp. 145-176. Bullen,K.E. (1965) An introduction to the theory of seismology. Cambridge University Press. 381pp. £ervenf,V. & fiavindra,B. (1971) Theory cf seismic head waves. University of Toronto Press. 312pp. Chapman,CH. (1976a) A first-motion alternative to geometrical ray theory. Geophys. Bes. Lett., V3, pp. 153-156. , (1976b) Exact and approximate generalized ray theory i n v e r t i c a l l y inhomogeneous media., Geophys. J . Boy. astron. S o c , V46, pp.217-233. Chase,B.L. ( 1977) J. Tuzo Hilson knolls: Canadian hotspot. Nature, V266, pp.344-346. T i f f i n , D. L. & Murray,J.H. (1975) The western Canadian continental margin, i n Canada's Continental Margins and Offshore Petroleum Exploration, edited by C.J.Yorath, E.B.Parker, and D.J,Glass, Can,,Soc. Petrol. Geol. Mem. 4, pp.701-721. Christensen,N.I. S Salisbury,M.H. (1975) Structure and constitution of the lower crust. Bev. Geophys., Sp. Phys., V13, pp.57-86. Clowes,B.M. (1977) A marine deep seismic sounding system. Can. J . Earth S c i . , V14, pp. 1276-1285. Crosson,B.S. <1972) Small earthguakes, structure and tectonics of the Puget-Sound region. B u l l . Seism. Soc. Am., V62, pp. 1133-1171. Davis,I.E., L i s t e r , C . B. B. 8 Lewis,B.T. R. (1976) Seismic structure on Juan de Fuca ridge: ocean bottom seismometer r e s u l t s from the Median Valley, J. Geophys. Res., V81, pp.3541-3555. Dickinson,H.B. (1976) Sedimentary basins developed during evolution of Mesozoic-Cenozoic arc-trench system i n western North America. Can. J. Earth S c i . , V13, pp.1266-1287. Flinn,E.A. (1965) Signal analysis using r e c t i l i n e a r i t y and dir e c t i o n of p a r t i c l e motion. Proc. I.E.E.E., 753, pp. 1874-1876. Gerver,M. S Markushevich, V. (1966) Determination of seismic wave velocity from the t r a v e l time curve. Geophys. J. Boy. astron, Soc,V11, pp. 165-173, G r i f f i n , J.H. (1966a) Applications and development of polarization (BEHODE) f i l t e r s . ," Seismic data lab. report #141, Earth Science Division, Teledyne Inc. (1966b) Bemode signal/noise tests i n polarized noise. Seismic data lab. report #162, Earth Science Division, Teledyne Inc. Heffler,D.£. & Locke,D.B. (1977) A deep ocean release mechanism. Mar. Geophys. Bes,, V3, pp.229-232. Houtz,B.E. (1976) Seismic properties of layer 2A i n the P a c i f i c . J . Geophys. Bes., V81, pp.6321-6340. Hyndman,B.D., Rogers,G.C., Lister,C.B.B., Wade,U.S., Barrett,D.L., Davis,E.E.# Lewis,!., Lynch,S. S Seeman,D. (1978) Geophysical measurements i n the region of the Explorer ridge o f f western Canada. Submitted to Can. J. Earth S c i . Johnson,L.E. & Gilbert,?. (1972) Inversion and inference for teleseismic ray data, i n Methods i n Computational Physics, V12, edited by B.A.Bolt, pp.231-266, Academic, New York., Johnson 11,8.7., Lister,C.B.B. S Lewis,B.T.B, (1977) A di r e c t recording ocean bottom seismometer. Mar. Geophys. Res., V3, pp.65-85. Julian,B.B. 6 Anderson,D.L. (1968) Travel times, apparent v e l o c i t i e s and amplitudes of body waves. B u l l . Seism. Soc. Am., V58, pp.339-366. Kanasewich,E.R. (1975) Time sequence analysis i n geophysics (second e d i t i o n ) . University of Alberta press. 364pp. Keen,C. E. & Barrett,D.L. (1971) A measurement of seismic anisotropy in the northeast P a c i f i c . Can. J . Earth S c i . , V8, pp.1056-1064. — g Tramontini,C. (1970) A seismic r e f r a c t i o n survey on the mid-A t l a n t i c ridge. Geophys. J. Roy. astron. S o c , V20, pp. 473-491. Kennett,B.L.N. (1976) A comparison of t r a v e l time inversions. Geophys. J . Roy. astron. S o c , ¥44, pp.517-536. , — (1977) Towards a more detailed seismic picture of the oceanic crust and mantle. Mar. Geophys. Res., V3, pp. 7-42. Bunch,A.».H., Orcutt,J.A. 5 Saitt,B.H. (1977) Variations in c r u s t a l structure on the East P a c i f i c Rise c r e s t : a t r a v e l time inversion approach. Earth Planet. S c i . Lett., V34, pp.439-444., & Orcutt,J.A. (1976) A comparison of t r a v e l time inversions for marine r e f r a c t i o n p r o f i l e s . J . Geophys. Res., V81, pp.4061-4040. Le JPichon,X. , Houtz,R.E., Drake,CL. 6 Nafe,J.E. (1965) Crustal structure of the mid-ocean ridges, 1, seismic r e f r a c t i o n measurements. J. Geophys. Res., V70, pp.319-339. Lewis,B.T.B. & McClain,J. ., (1977) Converted shear waves as seen by ocean bottom seismometers and surface buoys. B u l l . Seism. Soc. Am. , V67, pp. 1291-1302. Lister,C.B.B. 5 Lewis,B.T.B. (1976) An ocean-bottom seismometer suitable f o r arrays. Deep Sea Bes., V23, pp.113-124. Lynch,S. (1977) The c r u s t a l structure of Winona Basin as determined by deep s e i s n i c sounding. Unpublished MSc t h e s i s . University of B r i t i s h Columbia. 122pp. Malecek,S.J. 8 Clowes,B.H. (1978) Crustal structure near Explorer ridge from a marine deep seismic sounding survey. J. Geophys. Bes. (in press) McKenzie,D.P. S Morgan,H.J. (1969) Evolution of t r i p l e junctions. Nature, V224, pp. 125-133. £ Parker,B.L. (1967) The North P a c i f i c : an example of tectonics on a sphere. Nature, V216, pp.1276-1280. . McMechan,G.A. (1976) Generalized p-A curves. Geophys. J . Boy. astron. Soc., V47, pp.9-18. (1978) Multiple water r e f l e c t i o n s recorded at the ocean bottom. Can. J . Earth S c i . , V15, pp.78-85. & Wiggins,B.A. (1972) Depth l i m i t s i n body wave inversion, Geophys. J . Boy. astron, Soc., V28, pp.459-473. Montalbetti, J. ?• , 8 Kanasewich, E. B. (1970) Enchancement of teleseistaic body phases with a p o l a r i z a t i o n f i l t e r . , Geophys. J . Boy. astron. Soc., V21, pp.119-129. O'Brien,P.N.S. (1967) The e f f i c i e n t use of large charges, i n Seismic Refraction Prospecting, edited by fl.W.Musgrave. Society of Exploration Geophysicists. pp.152-170. Orcutt,J.A., Kennett,B.L.N. & Dorman,l.M. (1976) Structure of the East P a c i f i c Bise from an ocean bott o i seismometer survey. Geophys. J. Boy. astron. S o c , V45, pp. 305-32 0. Peterson,J.J., Fox,P.J.,& Schreiber,E. (1974) Newfoundland op h i o l i t e s and the geology of the oceanic layer. Nature, V247, pp.194-196. Poehls,K., (1974) Seismic refraction on the mid-AItantic ridge at 370N. J . Geophys. Bes., V79, pp.3370-3372. Biddihough,fi.P. (1977) A model f o r recent plate interactions off Canada's west coast. Can, J . Earth S c i . , V14, pp.384-396. 6 Hyndman,B.D. (1976) Canada's active western margin - the case for subduction. Geosci. Canada, V3, pp.269-278. Salisbury,M.H. e Christensen,N.I. *1978) The seismic v e l o c i t y structure of a traverse through the Bay of Islands Opbiolite Complex, Newfoundland, an exposure of oceanic crust and upper mantle., J. Geophys. Bes. , V83, pp. 805-817,• y Sax,B.L. (1966) F e a s i b i l i t y of linear p o l a r i z a t i o n measurements for detecting and measuring seismic body waves. Seismic data lab. report #163, Earth Science Divi s i o n , Teledyne Inc. & Mims,C. H. , (1965) B e c t i l i n e a r motion detection (BEMODE), Seismic data lab. report #118, Earth Science Division, Teledyne Inc. Shcr,G.G.Jr. , (1963) Befraction and r e f l e c t i o n technigues and procedure, in The Sea, edited by M.N.Hill., Interscience. pp.30-3 8. Silver,E.A. (1971) Small plate tectonics on the north-eastern P a c i f i c . B u l l . Geol. Sec. Amer., V82, pp.3491-3496. Souriau,M. S Veinante,J.L. (1975) Three adaptive f i l t e r s for the detection of body waves with application to deep seismic soundings. B u l l . Seism. Soc. Am. / V65, pp.995-1012. Spence,G.D. (1976) Gravity and seismic studies i n the Southern Bocky Mountain Trench. Unpublished MSc thesis, University of B r i t i s h Columbia. 84 pp. Srivastava,S,P. (1973) Interpretation of gravity and magnetic measurements across the continental margin of B r i t i s h Columbia, Canada. / Can. J. Earth S c i . , V10, pp.1664-1677. Stacey,B.A., (1S73) Gravity anomalies, cr u s t a l structure, and plate tectonics i n the Canadian C o r d i l l e r a . Can. J. Earth S c i . , V10, pp.615-628. Sutton,G.H., Kasabara,J., Ichinose,S.N, 6 Byrne,D.k. (1977) Ocean bottom seismograph development at Hawaii Inst i t u t e of Geophysics. Mar. Geophys. Res., V3, pp.153-177. Talwani,M., Windisch,C.C. 6 Langseth,M.G. (1971) Beykjanes ridge crest : a detailed geophysical study, J . Geophys. Bes., V76, pp473-517. Thorleifson,A.J. (1978) k marine deep seismic sounding survey over Hinona Basin. Unpublished MSc thesis. University of B r i t i s h Columbia, 70pp. T i f f i n , D . I . S Seemann,D. (1975) Bathymetric map of the continental margin of western Canada, open f i l e map, Geol. Surv. Canada. Tobin,D.G. & Sykes,L.E. (1968) Seismicity and the tectonics of the northern P a c i f i c Ocean. J. Geophys. Bes., V73, pp.3821-3845. Vine,F.J. (1966) Spreading of the ocean f l o o r , new evidence Science, V154, pp.1405-1415. & Wilson,J.T. (1965) Magnetic anomalies over a young oceanic ridge o f f Vancouver Island. Science, V150 pp.485-489. & Matthews,D.H. (1963) Magnetic anomalies over ocean ridges. Nature, V199, pp.947-949. Bhitmarsh, B.B. , (1978) Seismic r e f r a c t i o n studies of the upper igneous crust i n the North A t l a n t i c and porosity estimates for layer 2. Earth Planet. S c i . L e t t . , V37, pp.451-464. Wiggins,B.A. (1976) Body wave amplitude calculations I I. Geophys. J . Boy. astron., S o c , V46, pp.1-10. & Madrid, J . A- (1974) Body wave amplitude ca l c u l a t i o n s . Geophys J. Boy. astron. S o c , V37, pp.423-433. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0052919/manifest

Comment

Related Items