UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Seismicity along the continental margin of Vancouver Island Dimate Castellanos, Maria Cristina 1991

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

Item Metadata

Download

Media
831-UBC_1991_A6_7 D55.pdf [ 5.23MB ]
Metadata
JSON: 831-1.0052922.json
JSON-LD: 831-1.0052922-ld.json
RDF/XML (Pretty): 831-1.0052922-rdf.xml
RDF/JSON: 831-1.0052922-rdf.json
Turtle: 831-1.0052922-turtle.txt
N-Triples: 831-1.0052922-rdf-ntriples.txt
Original Record: 831-1.0052922-source.json
Full Text
831-1.0052922-fulltext.txt
Citation
831-1.0052922.ris

Full Text

SEISMICITY ALONG T H E CONTINENTAL MARGIN OF VANCOUVER ISLAND  by MARIA CRISTINA DIMATE CASTELLANOS B. Sc., Universidad Nacional de Colombia, 1980 M. Sc., Universidad Nacional de Colombia, 1988  A THESIS SUBMITTED IN PARTIAL FULLFTLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE  in THE FACULTY OF GRADUATE STUDIES  Department of Geophysics and Astronomy  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA July 1991 ©Maria Cristina Dimate Castellanos, 1991  In  presenting  degree freely  at  this  the  thesis  in  partial  University  of  British Columbia, I agree  available for reference  copying  of  department publication  this or of  thesis for by  his  fulfilment  and study.  I further  of  or  this thesis for  her  a  t  e  representatives.  npnpriy^irg anrl AgtT-nnnmy  July. 5,1001  DE-6 (2/88)  requirements that the  agree  It  financial gain shall not  The University of British Columbia Vancouver, Canada  D  the  scholarly purposes may be  permission.  Department  of  is  for  an  Library shall make  that permission for granted  advanced  by the  understood  be allowed  extensive  head  that  without  it  of  copying  my or  my written  ABSTRACT  The oceanic Juan de Fuca and Explorer plates are subducting beneath the continental North America plate west of Vancouver Island at rates of 47 mm/yr and 21 mm/yr, respectively. The Nootka fault zone, which is the boundary between the two subducting plates, experiences left-lateral shear due to their different rate of subduction. The Juan de Fuca/North America interaction vector is normal to the boundary beneath central-south Vancouver Island and oblique beneath western Washington. During the last 15 years, seismograph station coverage in western Canada has been adequate to record a good number of small magnitude earthquakes lying mainly in three regions: the Nootka fault zone, the Georgia Strait-Puget Sound region, and along the west coast of Vancouver Island and immediate offshore region. A detailed study of seismicity data in the Vancouver Island and offshore region, within the depth range of 25-55 km and recorded in the Canadian Earthquake Catalogue during the period 1976-1989, has been undertaken. The objectives of this study are to determine the frequency of occurrence and b-value, to improve the hypocentral locations and to interpret them in terms of both structure and the subductions process, and to examine the focal mechanisms to gain understanding of the stress field. A total of 127 earthquakes satisfied the depth range criteria and were analyzed in detail. Earthquake depths, recalculated with a recently determined velocity structure from refraction-reflection data and using the best station distribution and the more reliable arrival times, delineate a zone of seismicity beneath central-south Vancouver Island. The hypocenters show a distinct band of activity which extends ~60 km down-dip at a distance range of 35 to ii  95 km from the shelf edge and at a depth of 25 to 45 km. The structure is about 9 km thick and dips at an angle of 14° at N29°E. Most of these deep earthquakes are located in the subducting oceanic crust, a conclusion consistent with the position of the oceanic plate from other geophysical studies. The 6-value of the magnitude-frequency distribution calculated as 0.79 also is consistent with the conclusion of the seismicity occurring within the slab, on the basis of similar studies in western Washington and in other subduction zones. Areally, the new results confirm previous suggestions that the subducting plate is arched below the Georgia Strait-Puget Sound region. Composite focal mechanism solutions constructed for five groups of earthquakes show strike-slip and normal faulting mechanisms, but no thrust events. This important result confirms, at the small magnitude level, previous observations of the absence of thrust faulting earthquakes. Normal faulting may be associated with plate bending, an expected result, but the strike-slip mechanisms which suggest lateral shearing within the subducting plate, are an unexpected result. The composite of P (pressure) and T (tension) axes does not show a preferred orientation; the only noticeable feature is that the T axes indicate in-plane tensions. The diversity in the orientation of the P and T axes suggests a more complex state of stresses in the subducted slab than expected from the application of a normal subduction model. Such a complexity may be related to the arched plate geometry, to stresses generated by phase changes in the plate and to plate interactions at their boundaries.  iii  TABLE OF CONTENTS  ABSTRACT. . .  ii  LIST OF TABLES  vi  LIST OF FIGURES  vii  ACKNOWLEDGEMENTS  ix  CHAPTER 1 INTRODUCTION  -. 1  1.1 Tectonic setting  1  1.2 Seismicity  3  1.3 Thesis outline  .11  1.4 The data set  11  CHAPTER 2 SEISMICITY RATE AND b-VALUE CHAPTER 3 HYPOCENTRAL LOCATIONS  13 ,  20  3.1 Introduction  20  3.2 Velocity structure  21  3.3 Earthquake locations  28  3.3.1 Epicentral distribution  32  3.3.2 Hypocentral locations  32  3.3.3 Relocation of some earthquakes offshore  41  3.4 Plate geometry  43  3.5 Discussion  46  CHAPTER 4 FOCAL MECHANISM SOLUTIONS  51  4.1 Introduction  51  4.2 Composite focal mechanism solutions  52  4.3 Discussion  64 iv  CHAPTER 5 SUMMARY AND CONCLUSIONS  68  REFERENCES  72  "  APPENDIX A  77  v  LIST OF TABLES  1.1  Significant earthquakes in central-south Vancouver Island and north Puget Sound region  5  3.1  Velocity models used in hypocentral locations  22  4.1  Detailed composite focal solutions  54  4.2  Focal solutions of significant earthquakes in the Juan de Fuca plate beneath British Columbia and Washington  61  vi  LIST OF FIGURES  1.1  The continental margin of southwestern Canada and  ..2  1.2  Seismicity of southwestern British Columbia from 1899 to 1975  1.3  Location and focal mechanisms of significant earthquakes in central-south Vancouver  4  Island-north Puget Sound region 1.4  6  Depth distribution of seismicity in central-south Vancouver Island during 1975-1988 and 1982-1988  1.5  9  Epicentral distribution of seismicity in southwestern British Columbia for 1982-1988  *  10  2.1  Western Canada Telemetered Network  2.2  Magnirade-frequency distribution for deep earthquakes beneath central-south  15  Vancouver Island 2.3  16  b-value for deep earthquakes beneath central-south Vancouver Island and beneath Puget Sound  17  2.4  Scaled magnitude frequency distribution in central-south Vancouver Island.  3.1  Comparison of hypocentral locations with three crustal models  25  3.2  P-wave velocity to S-wave velocity ratio  26  3.3  Comparison between the Canadian Earthquake Catalogue and the model 3 hypocentral locations  3.4  . . . 19  29  Epicenters in the 25—45 km depth range beneath central-south Vancouver Island  33  3.5  Depth intervals for earthquakes in the 25-45 km range.  34  3.6  Areas plotted in cross-sections  37 vii  3.7  Cross-sections parallel to the subduction direction of seismicity in the depth range 25 to 55 km  3.8  38  Cross-sections parallel to the shelf edge line of seismicity in the depth range 25 to 55 km  3.9  .40  RMS residuals of the hypocentral solutions as a function of depth  42  3.10 Contour depths to the top of the brittle zone inferred from seismicity  44  3.11 Contour depths to the top of the oceanic crust from reflection data  45  3.12 Plate geometry beneath western Washington and Vancouver Island inferred from seismicity  47  3.13 Cross section of deep seismicity including earthquakes beneath the Georgia Strait  48  3.14 Overlay of deep seismicity with structural models  49  4.1  Composite focal mechanism solutions for five groups of earthquakes.  53  4.2  Details offirstmotion observations for composite focal solutions.  58  4.3  Focal plots for significant earthquakes in the Juan de Fuca plate beneath British  4.4  .  Columbia and Washington  60  Composite of P and T axes  63  viii  ACKNOWLEDGEMENTS  I would like to thank Dr. Robert M. Ellis for his supervision and guidance during the course of this study. His comments and suggestions have contributed immensely to this thesis. I am also grateful to Dr. Garry Rogers who suggested the subject, provided the basic data and also contributed with many helpful comments and discussions which improved the results of this study. I thank Dr. Ron M. Clowes, Dr. Mattew Yedlin and Dr. William Slawson who reviewed the manuscript and made useful suggestions. Financial support for this study was provided by the Canadian International Development Agency through a training agreement with the Colombian Government. Additional funding was obtained from the International Peace Scholarship Fundation.  ix  CHAPTER 1 INTRODUCTION  1.1 Tectonic setting The tectonic regime of western Canada is dominated by the relative motion of three lithospheric plates: the large Pacific and America plates and the intervening Juan de Fuca plate (Figure 1.1). The boundary between the Pacific and North America plates is the Queen Charlotte transform fault which has right-lateral motion at a rate of about 55 mm/yr with a small amount of oblique convergence (Yorath and Hyndman, 1983). The Juan de Fuca plate is made up of two segments separated by the Nootka fault: a small northern segment, the Explorer plate, and the main lithospheric plate to the south. The boundary between the Juan de Fuca plate and the Pacific plate is defined by a series of spreading ridges offset by transform segments (see Figure 1.1). The northern ridge system (Tuzo Wilson Knolls, Delwood and Explorer ridges), with a full spreading of about 40 mm/yr, defines the boundary between the Explorer and Pacific plates, and the Juan de Fuca ridge, diverging at about 60 mm/yr (full rate), defines the main Juan de Fuca-Pacific boundary (Riddihough, 1977). Along the continental margin, at the boundary between the Explorer-Juan de Fuca and North American plates, there is a zone of slow convergence. The Explorer plate has possibly ceased to subduct into the mande and is being overridden by the continent at a rate of 21 mm/yr at N50°E. South of the Nootka fault the rate of convergence is larger, about 47 mm/yr at N56°E (Riddihough and Hyndman, 1990). The Juan de Fuca-America interaction is  l  CHAPTER 1. INTRODUCTION  Figure 1.1 The continental margin of southwestern Canada and northwestern United States showing the plate-tectonics regime and main tectonic elements (after Hyndman et al., 1990). The box encloses the area of this study. Arrows indicate directions and rate of movement. 2  CHAPTER 1. INTRODUCTION  primarily oblique subduction (Rogers, 1983), except for a short section on Vancouver Island between the Nootka fault and Georgia Strait where the convergence vector is orthogonal to the boundary between the two plates. Two triple junctions, one at the northern end of the Explorer plate and the other at the southern end of the south Gorda plate, mark the ends of the subduction zone.  1.2 Seismicity In a review of the seismicity of western Canada up to 1975 Milne et al. (1978) pointed out that most of the seismicity of the margin appeared to be associated with the major plate boundaries: the Queen Charlotte fault system (Pacific and America plates), the offshore ridgefracture zone system (Pacific-Juan de Fuca plates, and the Vancouver Island-Puget sound region (Juan de Fuca-America plates). More recent studies (e.g. Rogers, 1983; Cassidy et al., 1988) have clarified the relationship of seismicity to the tectonic model and refined it. Seismicity of southwestern Canada between 1899-1975 is shown in Figure 1.2. The largest earthquakes in the recording period have been on the Queen Charlotte fault and in central Vancouver Island. The Queen Charlotte fault is a very active earthquake area and has been the location of three large earthquakes in the recording period: 1929, M=7; 1949, M =8.1; 1970, M =1A. . Focal mechanisms for the 1949 and 1970 earthquakes indicate strike s  S  slip with some thrust component (Rogers, 1983). In central Vancouver Island six significant earthquakes (M > 5.2) have occurred since 1918: 1918, 1946, 1957,1972, 1975a, 1986, "see Table 1.1 and Figure 1.3 for details. Magnitude abbreviations in this section represent M or M local, M surfave wave and m body wave magnitudes, respectively, as given by L  s  b  authors. These events all have mechanisms which are predominantly strike-slip with a similar orientation of nodal planes. The closest earthquakes to the Nootka fault (1957, 1972, 1986) 3  CHAPTER 1. INTRODUCTION  Figure 1.2 Seismicity of southwestern British Columbia from 1899 to 1975. Dot sizes are proportional to the earthquake magnitude. Earthquakes with magnitude less than 3.0 are marked by V . Uncertainties in epicenters of offshore events may approach 50 km (from Keen and Hyndman, 1979).  4  CHAPTER 1. INTRODUCTION  T A B L E 1.1 Significant earthquakes in central Vancouver Island and north Puget Sound Region  DATE  LOCATION  Jan. 11, 1909 < > Dec. 6, 1918 < > Dec. 4, 1926 <» Feb. 9, 1928 ^ June 23 1946 « Dec. 16, 1957 » July 5, 1972 <» Mar. 31, 1975a « Nov. 30, 1975b <" May 16, 1976<» June 16, 1986 « J  2  48.7°N 49.4°N 48.5°N 49.0°N 49.8°N 49.6°N 49.5°N 49.3°N 49.2°N 48.8°N 49.4°N  122.8°W 126.2°W 123.0°W 125.3°W 125.3°W 127.0°W 127.2°W 126.0°W 123.6°W 123.3°W 127.0°W  <» Rogers, 1983. ® Cassidy et al., 1988. Rogers and Hasegawa, 1978. P)  5  DEPTH (km) MAGNITUDE —  15  M =6.0 L  M =e.9 s  —  M =5.0  —  M =5.S  30 30 25 18 10 62 • 35.2  L  L  M =7.2 s  M =5.9 s  m =5.7 b  M =5A L  M =4.9 L  M =5A L  M =5.2 L  CHAPTER 1. INTRODUCTION  130°  128°  126°  Figure 1.3 Location and focal mechanisms of significant earthquakes in central-south Vancouver Island-north Puget Sound region. Larger dots indicate unknown focal mechanisms. Locations correspond to list in Table 1.1 (adapted from Cassidy et al., 1988).  6  CHAPTER 1. INTRODUCTION  have locations and source parameters which are consistent with slip along the fault (Cassidy et al., 1988). The 1918 and 1946 earthquakes which occurred in the continental crust and further inland and the 1975a earthquake, also a crustal event, are likely due to the stress regime in the continental crust which results from the coupling of the obliquely subducting Juan de Fuca and Explorer plates with the America plate (Rogers, 1983; Cassidy et al., 1988). In south Vancouver-Island and north Puget Sound region several moderate magnitude earthquakes have occurred: 1909, M =6; 1926, M =5.0; 1928, M =5.8; 19756, M =49; L  L  L  L  1976, M =5A). The 1909 earthquake was apparently a comparatively deep event (Rogers, L  1983). The depths of the 1926 and 1928 earthquakes have not been determined. The 19756 is a shallow thrust earthquake and the 1976 has been interpreted as a subcrustal event associated to phase changes in the subducting plate (Rogers, 1983). Location and focal mechanisms of seismicity in this region contrast markedly with the distribution of seismicity typical of most subduction zones. It lacks a deep (> 70 km) Benioff zone of earthquakes and large thrust earthquakes at the interface between the two plates have not been observed (Rogers, 1983). This distribution is also characteristic of the OregonWashington portion of the subduction zone (e.g.  Ludwin et al., preprint). Whether the  oceanic and continental plates are normally locked together or whether they move relatively to each other aseismically has been a subject of debate (Ando and Balazs, 1979; Heaton and Kanamori, 1984). Comparison with other subduction zones and recent evidence from strain measurements and paleoseismicity studies indicate that the former situation is likely in the Cascadia subduction zone and that the possibility of a great subduction zone earthquake should be considered (Heaton and Kanamori, 1984; Rogers, 1988; Darienzo and Peterson, 1990; Yamaguchi et al., 1989; Savage and Lisowski, 1991). The installation in 1975 of the first instruments of the Western Canada Telemetered 7  CHAPTER 1. INTRODUCTION  Network improved significantly the detection capabilities of seismicity in the region allowing detection of lower magnitudes and better depth control on earthquake hypocenters. The depth distribution of seismicity in central-south Vancouver Island and /in the southern Vancouver Island-north Puget Sound region shows a distribution of earthquakes in two zones: 1) a zone of crustal seismicity about 20 km thick with a peak of activity at 15-20 km for central-south Vancouver Island and about 30 km thick with a peak at 25 km for the southern Vancouver Island-north Puget Sound region (Rogers, 1983); and 2) a zone of deep seismicity, separated from the shallow zone by a lull of activity of 5 to 10 km, going from about 25 to 55 km and 40 to 65 km, respectively. Depth distribution for central-south Vancouver Island and offshore region is shown in Figure 1.4. The epicenters of the deeper suite lie principally in three regions (Figure 1.5.b): (a) the Nootka fault zone along which left lateral slip occurs due to the different rate of motion between the Juan de Fuca and Explorer plates (Hyndman et al., 1979), (b) southern Vancouver Island-Puget Sound region where it has been suggested that earthquakes are occurring due to arching in the Juan de Fuca plate or triggered by phase changes in the oceanic plate as it subducts into a higher temperature and pressure region (Rogers, 1983; Taber and Smith, 1985; Weaver and Baker, 1988), and (c) along the west coast of Vancouver Island and immediate offshore region beneath the continental shelf where seismicity may be due to rupture of the Juan de Fuca plate as it bends beneath the North American plate. The epicenters of the shallow suite of seismicity are also concentrated in the Nootka fault area and in the southern Vancouver Island-Puget Sound region 0?igure 1.5.a). The latter are due probably to stresses being coupled across the subduction interface as a result of the complex geometry of the plate (Rogers, 1983; Weaver and Baker, 1988). Apart from these two regions there is a broader zone of relatively sparse seismicity extending from the 8  0  0 1975-1988  1982-1988  30  30 c D >  0)  n  o © W  60  o  60  0  _i  i  ' i  50 No. OF EVENTS (km)  0  50 No OF EVENTS (km)  Figure 1.4 Depth distribution in central-soudi Vancouver Island (zone of study shown in Figure 1.1) for seismicity during (a) 1975-1988 and (b) 1982-1988. Note that the peak in activity at about 18 km depth more pronounced in (a) due to afixeddepth assigned to poorly recorded events during 1975-1981.  CHAPTER 1. INTRODUCTION  Figure 1.5 Epicentral distribution of seismicity in southwestern British Columbia for the period 1982-1988, (a) depth < 30 km and (b) depth > 30 km. (Rogers, pers. comm.). Box indicates area of this study. 10  CHAPTER I. INTRODUCTION  continental slope and shelf to central Vancouver Island, the Georgia Strait and the mainland region which defines the North American plate. The largest magnitude earthquakes in this set are located on central Vancouver Island.  1.3 Thesis Outline The general features of seismicity on central Vancouver Island and offshore region have been identified and related to the tectonic model but detailed studies have not been carried out.  Seismograph station coverage from 1975 to 1990 has been adequate to collect a  sufficient number of earthquakes and good quality data to allow more complete and detailed analysis. In this study we undertake the detailed study of seismicity along the west coast of Vancouver Island and immediate offshore region in the 25-55 km depth range to gain a better understanding of earthquake generation processes there. The boundary of the study area is shown in figures 1.1 and 1.5.b. The aims of this study are to use the recorded arrival times, previously calculated magnitudes and thefirstmotion observations to obtain a better understanding of the tectonic processes which generate this set of earthquakes.  More specifically, the frequency of  occurrence andft-valueare determined (Chapter 2), improved hypocenters are calculated and interpreted in terms of both structure and the subduction process (Chapter 3), and the focal mechanisms are examined to gain understanding of the stress field (Chapter 4).  1.4 The data set The seismicity data analyzed in this study are a selection from the Canadian Earthquake Catalogue, recorded between 1975-1989 by the Western Canada Telemetered Network and the Lower Mainland Seismic Array in British Columbia. The selected earthquakes, about 180, 11  CHAPTER 1. INTRODUCTION  are those originally located in the region delineated by the coordinates (124.2°W, 48.8°N), (125.6°W, 50.2°N), (127.1°W, 49.3°N), (125.5°W, 48.1°N) and with depth between 25 and 55 km (Figure 1.4). Figure 1.4 shows the depth distribution in the area for two overlapping periods: 1975-1988 and 1982-1988. Note the strong peak in the 15-20 km depth interval in the 1975-1988 plot and its absence in the 1982-1988 plot. This occurs because some of the stations on the west coast were not installed until 1982-1983. Therefore, because of the inadequate data set, during the processing it is often necessary to assign a depth in order to find a solution. The assigned depth was usually 18 km. In this study data from 1983 and later were used, except for the 6-value calculations (Chapter 2) where data since 1976 to 1988 were used.  12  CHAPTER 2 SEISMICITY RATE AND B-VALUE  The relation between the magnitude of earthquakes and the frequency of occurrence generally satisfy the equation:  log N = a - bM  where N is the number of shocks of magnitude Af or greater for a particulartimeperiod (Gutenberg and Richter, 1949), a is the intercept and b the slope of the straight line defined by the relation between log N and M. The b-value is a characteristic of the tectonic regime and it has been associated with the state of stress in the source region (Wyss, 1973). The 6-value was calculated using the method of Weichert (1980) which uses the maximum likelihood method to estimate a and b for the case of earthquake data grouped in magnitude, with unequal observation periods for each group. From the Canadian Earthquake Catalogue (CEC) we selected the earthquakes in the study area in the depth range 25-55 km for the period 1976-1988 (about 180 events). During this period the detection capabilities of the seismic network of western Canada changed significantly.  When it commenced operation in 1975 the Western Canada Telemetered  Network consisted of 4 short-period vertical stations.  At the end of 1987 the network  consisted of 18 stations, with the greatest improvement made in 1981-1982. These stations 13  CHAPTER 2. SEISMICITY RATE  are in addition to the six regional seismograph stations and one standard station (Munro et al., 1990). Present location of stations is shown in Figure 2.1. To determine a data set representative of the magnitude-frequency distribution in the area, the data were grouped in magnitude intervals of 0.1 and a graph of frequency of earthquakes per year in a given magnitude interval versus magnitude was plotted. Mi magnitudes reported in the CEC were used. To take into account the varying detection capability of the network the data were divided into two periods: 1976-1982 (Figure 2.2.a) and 1983-1988 (Figure 2.2J?). Examination of Figure 2.2 leads to the conclusion that the threshold magnitude locatable for the network is about 1.7 for the 1976-1982 period and about 1.2 for 1983-1988. This conclusion is based on the linearity exhibited for the distribution for magnitudes larger than the threshold value. The maximum earthquake magnitude was chosen as 7.1, the magnitude of the April 13, 1949 subcrustal earthquake south of Puget Sound (Baker and Langston, 1987). Variations in maximum magnitude of one unit did not affect significantly the ^-values. Using these parameters, a 6-value of 0.79 ± 0.07 was obtained (See Figure 2.3.a). This value is very different from 0.38, the value calculated by Rogers (1983) for earthquakes with magnitudes larger or equal to 4.0 in central Vancouver Island prior to 1979. This region has experienced three large earthquakes in the present century (M=7 in 1918, M=7.3 in 1946 and M=6 in 1957) and very few small magnitude earthquakes occur there (Rogers, 1983). The very different 6-values suggest a different tectonic environment associated with deep microseismicity in the area. On the other hand, the &-value of 0.79 compares with 0.73, the 6-value obtained by Crosson (1983) for a set of deep earthquakes (depth > 35 km) in the Puget Sound region (see Figure 2.3.b). They have been interpreted as intraplate earthquakes in the Juan de Fuca plate (Taber and Smith, 1985). Although the Puget Sound events are deeper and further inland compared to the set of earthquakes of 14  CHAPTER  2. SEISMICITY RATE  Figure 2.1 Western Canada telemetered network and other stations (from Munro et al., 1990). Box indicates the area of this study.  15  CHAPTER 2. SEISMICITY RATE  n ' o  •  r  1  I  —  I  —  •  1976-1982 \ CD  3 CD  > -i-> a  a O  0.2 O D D D O O O D D D Q D D O O D a  0  >.  \  SH CD  1  _L  2 3 Magnitude  1983-1988 10  B d CD  > -t->  cd  a 0.2 0  2 3 Magnitude  Figure 2.2 Magnitude frequency distribution. Cumulative number of events per year (on a logarithmic scale) above specified magnitude, (a) Earthquakes occurred during 1976-1982. (b) Earthquakes occurred during 1983-1988. For events with magnitudes larger than about 1.7 in (a) and 1.2 in (b) the distribution approaches a linear relationship.  16  CHAPTER 2. SEISMICITY RATE  t  •  r  Magnitude Figure 23 (a) Maximum likelihood determination of 6-value for deep earthquakes (25-55 km) beneath central-south Vancouver Island from the Nootka fault to the Juan de Fuca Strait. 6-value is 0.79+ 0.07. Maximum magnitude of 7.1 and minimum of 1.2 were used, (b) Recurrence relationship for a deep suite of earthquakes (depth > 35 km) beneath Puget Sound (1970-1978) taken from Crosson (1983). b-value is 0.73. The 6-values are similar for both distributions. 17  CHAPTER 2. SEISMICITY RATE  this study, the ^-values support the idea of a common tectonic environment and suggest an overall £>-value of about 0.7-0.8 for intraplate seismicity in the Juan de Fuca plate beneath central Vancouver Island and Puget Sound. Comparison with other subduction zones indicates similar values for oceanic seismicity at the same depth range; e.g., for Japan, Anderson et al. (1980) found a b-value of 0.7 in the 40 to 70 km depth range, and in the Mid-America trench offshore Guatemala, Ambos et al. (1985) found it to be 0.71 in the 10-40 km range. Assuming that Figure 2.3.a is representative of the deep seismicity in central-south Vancouver Island and taking a rectangular area of 130 km by 200 km, frequencies of occurrence per year were scaled to a 1000 km area and are shown in Figure 2.4. Puget Sound 2  region frequencies in Figure 2.3.b were also scaled to the same period and area, assuming 42 000 km as the area covered by the epicenters (adapted from Crosson, 1983) and a 9 year 2  recording period. The recurrence rate at magnitude 4 is 0.002 events/year/1000 km for our 2  data set and 0.013 events/year/1000 km for deep Puget Sound seismicity. 2  The relatively low level of activity for central-south Vancouver Island indicated by the recurrence rate may be the result of a less deformed slab north of Puget Sound compared with the more pronounced slab arch whose axis is beneath Puget Sound (Crosson and Owens, 1987; Weaver and Baker, 1988). Larger depths for the Puget Sound set also favour increasing activity in this area due to stresses in the subducting plate induced by phase changes (Pennington, 1984), while for central-south Vancouver Island comparatively shallower depths will restrict this type of activity although it still can occur.  18  CHAPTER 2. SEISMICITY RATE  Magnitude  Figure 2.4 Magnitude frequency distribution in central-south Vancouver Island, scaled to one year and 1000 km . Recurrence rate at magnitude 4 is 0.002 events/year/ 1000 km . 2  2  19  CHAPTER 3 HYPOCENTRAL LOCATIONS  3.1 Introduction Accurate location of microseismicity is important to define both crustal and subcrustal seismic zones, to infer the plate geometry at the subduction margin and to impose constraints on the seismic processes and mechanical properties. Some of the focal depths in our data set are only weakly constrained due to the distribution of stations since most of the stations are located to the east of the epicenters and there is almost no coverage to the west (see Figure 2.1 for location of stations). However locations can be improved by using more precise velocity models and appropriate selection of arrivals. The precision and accuracy of hypocentral locations obtained with a local seismic network are controlled by many factors. The most important ones are: 1) distribution of seismic stations (which controls the stability of the inversion procedure); 2) accuracy of the velocity model used in travel time calculations (errors in the model can introduce systematic errors which are difficult to evaluate); 3) accuracy in the time picking of the seismic records (which determines the degree of fit). In order to analyze the reliability of the hypocentral locations, tests using variations in the velocity models and some of the factors controlling the location procedure were performed. Using parameters based on these tests, a set of 138 earthquakes which occurred in the period 1983-1988 with depths in the range 25-55 km reported in the Canadian Earthquake Catalogue 20  CHAPTER 3. HYPOCENTRAL LOCATIONS  (CEC) were relocated. Eleven were located shallower than 25 km, and were not included in the analysis. The final 127 are listed in Appendix A. The tests were performed using the ten largest earthquakes in the data set all with magnitudes > 2.5. These events are distributed throughout the zone of study and, in general, are recorded at a large number of stations. They are event numbers 2, 13, 16, 39, 44, 49, 53, 63, 107, 123 of Appendix A. The numbering system of this appendix is used to designate earthquakes throughout the remainder of this thesis. The program HYPOELLIPSE (Lahr, 1985) was used for locations.  The program  determines the hypocenter and for each event the ellipsoid which encloses the 68% confidence volume. The hypocenter is found using Geiger's method (Geiger, 1912) to minimize the root mean squares (RMS) of the travel time residuals.  3.2 Velocity structure To compute theoretical traveltimesin the location procedure, a model of the seismic velocity structure must be specified.  Thus, earthquakes are located with respect to the assumed  velocity model and not to the real earth. Therefore the quality of the earthquake location will be controlled, among many factors, by the quality of the velocity model. To examine the effect of the velocity model on hypocentral locations the set of 10 earthquakes were relocated using three different velocity models (Table 3.1): 1) model 1, adapted from the refraction onshore-offshore structural model of Spence et al. (1985) in the Vancouver Island region ; 2) model 2, Georgia Strait crustal model (Rogers, 1983), based on a combination of P models for the Georgia Strait-Vancouver Island-Puget Sound region; 3) model 3, adapted from the refraction-reflection model of Drew and Clowes (1990) across Vancouver Island. 21  CHAPTER 3. HYPOCENTRAL LOCATIONS  TABLE 3.1 Regional Earth models used in hypocentral locations (V = P-wave velocity, Z = depth to top of the layer) p  Model 1. Spence et al. (1985)  Vp (km/s)  Z (km)  5.85  0.0  6.60  3.0  6.95  15.0  6.55  19.0  7.70  34.0  8.1 0  35.0  Model 2. Vancouver Island-Puget Sound model (Rogers, 1983)  V  P (km/s)  Z (km)  5.00  0.0  6.00  1.0  6.70  6.0  7.10  30.0  7.75  45.0  22  CHAPTER 3. HYPOCENTRAL LOCATIONS  TABLE 3.1 (cont.)  Model 3. Drew and Clowes (1990)  WEST  V  EAST  P (km/s)  Z (kni)  5.85  P (km/s)  Z (km)  0.0  5.85  0.0  6.60  2.5  6.60  3.5  6.35  13.0  6.95  15.0  7.15  15.0  6.65  18.0  6.35  23.0  7.55  36.0  7.15  25.0  8.00  38.0  6.50  30.0  7.15  32.5  7.70  36.0  V  23  CHAPTER 3. HYPOCENTRAL LOCATIONS  HYPOELLIPSE is restricted to horizontal layering but lateral variations in structure can be accommodated by using the multiple crustal structure option. When using this option each station is assigned to one of the crustal models and that model is always used to calculate travel times to that station. In our tests all the models were approximated by horizontal layers with constant velocity in each layer and with a constant Poisson's ratio throughout the model. The model of Drew and Clowes (1990) was approximated as two horizontally layered models (east and west) to handle the quite different crustal structure beneath the stations on west Vancouver Island and those on east Vancouver Island and the mainland. In the first test the events were located with the three models using a V /V ratio of 1.73, p  s  the value currently used in location procedures for Vancouver Island, and the same set of arrival times as in the Observed Data Record (ODR) of the CEC. To evaluate the performance of each model for this data set, the average RMS of the time residuals for the set (RMSav) and the average of the maximum standard error (the maximum between horizontal and depth error), MSEav, were calculated. It was found that the variations in epicentral locations between the models were at most 0.02 degrees, i.e. less than two kilometers, and in depth were from 2 to 8 km (Figure 3.1). The largest variations in depth occurred for model 1. Variations in RMSs and MSEs between models for a given earthquake rarely exceeded 0.2 sec and 1 km respectively. The largest RMSav and MSEav, 0.35 sec and 2.3 km respectively, correspond to model 1; for model 2 RMSav=0.30 sec and MSEav=2.1 km; and for model 3 RMSav=0.32 sec and MSEav=2.0 km. A second test was performed to analyze the effect of the S-wave velocity structure. To determine the S-wave velocity, the P-wave arrival time differences were regressed against the S-wave arrival time differences for pairs of stations for the 10 largest earthquakes. The slope defined by the points gives the velocity ratio and, thus, Poisson's ratio. The slope was 24  10  20 13  63  30  53 16 44  107 39  49 r  VI/  •  40  50  60  A model 1 O model 2 • model 3  123  20  40 60 DISTANCE (km)  80  100  Figure 3.1 Hypocentral locations of the 10 largest earthquakes in the data set (magnitudes 2.5-3.6), used to test the effect of varying crustal structure on the focal depth. Locations are projected in cross section along AA' (see Figure 3.6). The error bars represent the horizontal and vertical errors (ERH, ERZ) defined in Section 3.3. The 3 crustal models are listed in Table 1.  CHAPTER 3. HYPOCENTRAL LOCATIONS  O CD W  CO  5 P  10  (sec)  Figure 3.2 V /V ratio. S-wave arrival time differences for pairs of stations versus P-wave arrival time differences for pairs of stations selected from the 10 largest earthquakes in the data set. This yields a P velocity to S velocity ratio of 1.76 ±0.02 (Poisson's ratio 0.26± 0.01). p  s  26  CHAPTER 3. HYPOCENTRAL LOCATIONS  calculated using the York algorithm (York, 1966), modified by Williamson (1968), which allows errors in both x and y coordinates in the input data. A set of about 40 points was selected and is plotted in Figure 3.2. P and S time uncertainties were estimated as 0.2 and 0.4 sec, respectively. The resulting value for V IV was 1.76 + 0.02 (Poisson's ratio of P  S  0.26 + 0.01). The 10 earthquakes were relocated using V /V = 1.76 for the three velocity models. p  s  Again model 1 was found to have larger residuals and solutions which were significantly different than for models 2 and 3 and thus was deemed inappropriate; only models 2 and 3 are considered further. When compared with the values obtained for locations with V IV = 1.73 p  s  the depths obtained were up to 1.5 km shallower for model 2. For model 3 depths changed randomly by at most 2.0 km. No systematic variation in the residuals or in the horizontal or vertical errors was observed for any of the models. Comparing the results obtained with V IV = 1-73 and V IV = 1.76 for a given event, the differences in RMS residuals were of P  S  P  S  the order of 0.02-0.15 sec and the differences in standard errors 0.2-1.5 km. Comparing models, model 3 exhibited better results: RMSav=0.29 sec and MSEav=1.76 km while for model 2 RMSav=0.32 sec and MSEav=2.22 km. As a result of this test, we concluded that small variations in the V IV ratio introduce no P  S  significant differences in the locations and error estimates of earthquakes in the area. Also, we observed that locations obtained with model 3 were the most precise, in the sense that in general they exhibited the smallest RMS residuals and standard errors. Thus, for subsequent calculations we retained the value of 1.76 as most representative of the average V /V ratio p  s  for the structure crossed by the wave paths. This is an intermediate value between 1.73 and 1.77-1.78, the two values obtained by Rogers (1983) when experimenting with locations of deeper earthquakes and quarry blasts on southern Vancouver Island. 27  CHAPTER 3. HYPOCENTRAL LOCATIONS  A third test was performed to check the effect in the location of distant P and S wave arrivals. The program HYPOELLJPSE employs P and S arrivals with variable weights depending on the quality of the pick. The 10 earthquakes were relocated modifying the weights of the arrivals. These were assigned taking into account the fact that P arrivals usually are picked with less uncertainty than S arrivals and that time arrivals from distant stations usually have larger uncertainties. Also, arrival times with large residuals in preliminary calculations were assigned low weight for the final hypocentre determination. After several trials, it was found that a maximum epicentral-station distance of 110 to 130 km provided the best constrained solutions. Again, smaller RMS residuals and errors were obtained with model 3 for most of the earthquakes. For model 2 RMSav was 0.27 sec and MSEa 1.59 km, and for model 3 RMSav was 0.22 sec and MSEav 1.16 km. As a result of the previous tests, the quality of the locations (quantified by RMS residuals and horizontal and vertical errors) of 8 of the 10 earthquakes was improved significantly. This is reflected in the moretightlyclustered hypocentral locations compared with the initial locations in the CEC, as shown in Figure 3.3. For earthquake 39, an event with a poor station distribution around the epicenter, the final location showed significandy larger residual and MSE than the starting solution, indicating a very unstable solution, and for earthquake 49, an event with good station distribution, the MSE remained the same and the RMS increased slightly (0.02 sec). The previous results suggest that improvement can be achieved in the locations of intermediate quality recorded earthquakes by using a more precise velocity structure and very accurate arrival times.  3.3 Earthquake locations The complete data set was processed using model 3 and V /V = 1.76. The P and S p  28  s  10  20 63  6  107  30  a H OH  w  13  5 3  ^•  39  O  d  40  3  0  49  O  44  •  O  123  Q  16  O  O 50  60  O •  Can. Earth. Cat. model 3 20  40 60 DISTANCE (km)  80  100  Figure 3.3 Hypocentral locations for the 10 largest earthquakes as listed in the Canadian Earthquake Catalogue (CEC), indicated by open dots, and as calculated with crustal model 3 (described in Table 3.1), indicated by solid dots. Locations are projected in cross section along AA' (see Figure 3.6). Depth errors are not provided in the CEC. When not shown, error bars for locations with model 3 are smaller than the size of the dot.  CHAPTER 3. HYPOCENTRAL LOCATIONS  weights were modified using the same criteria as in the third test. From the 138 earthquakes in the range 25-55 km depth listedin the CECduring 1983-1989, 11 earthquakes were placed shallower than 25 km and most of them were small magnitude and located offshore. They are not included in the summary list to keep the initial depth range criteria and because, as discussed in Section 3.3.1, they do not affect significantly the overall earthquake distribution. The final 127 earthquakes, their locations, error estimates and details of the solutions are summarized in Appendix A. Errors in hypocentral locations are variable across the zone and depend principally on the lack of stations to the west and the small magnitude of some of the earthquakes which were recorded at fewer stations and with lower quality arrivals. With reference to the quality of the solutions in terms of standard errors and RMS residuals, 65% of the solutions have RMS < 0.30 sec, horizontal error (ERH, horizontal 68% confidence limit in the least well constrained direction) < 2.5 km and vertical error (68% confidence limit for depth) < 5.0 km. The remaining solutions have larger values for these parameters (qualities C and D according to HYPOELLIPSE; see Appendix A for details of quality evaluation in this program). The distribution of seismic stations has been mentioned as one of the factors controlling the accuracy of the earthquake locations.  HYPOELLIPSE evaluates the quality of the  distribution with respect to the locations according to the number of time readings used in the solution (NO), to the hypocentral distance to the nearest station (DMIN) and to the larger azimuthal separation between stations as seen from the epicenter (GAP). For our data set, the earthquakes with the best station distributions have NO > 6, DMIN < 50 km and GAP < 180° and correspond to about 50% of the solutions. The remaining solutions, most with epicenters offshore, have larger values for these parameters. Combining these two factors, i.e., quality of the solution and quality of the station 30  CHAPTER 3. HYPOCENTRAL LOCATIONS  distribution, we have selected our highest quality locations as those with RMS < 0.30 sec, ERH < 2.5 km, ERZ < 5.0 km, NO > 6, DMIN < 50 km and GAP < 180°. These correspond to about 50% of the events in the data set and are indicated with solid dots in the subsequent figures in this chapter. Additional locations are indicated by open dots. It is worth noting that standard errors and RMS residuals, calculated for each hypocentral location, are associated mainly with the precision of the solution, i.e., how well the arrival time data fit the calculated solution. However, the. accuracy of the solution, measured by the approximation of the calculated hypocenter to the true hypocenter, is not evaluated. The accuracy is usually worse than the precision and is influenced primarily by imperfect approximation of the crustal structure in three dimensions between the hypocenter and each of the stations (Ward and Gregersen, 1973). The main constraints to estimate the accuracy of our solutions are provided by the geometry of the plate inferred from other studies (see Section 3.5) and the fact that earthquakes can only occur in the portion of the plate that is cold enough to deform by brittle fracture. A conservative estimate of the overall hypocentral errors, taking into account the minimum value imposed by the standard errors, station distribution, and accuracy of the velocity model would be: ±3 km for earthquakes onshore for which accurate crustal structure is known and are recorded at several nearby (less than 40 km distant) stations; ± 5 km for earthquakes offshore and no more distant than 20 km from the coast which, although frequently recorded in a good number of stations, have GAPs larger than 180° and the nearest station from the epicenter can be more than 50 km; and ±10 km for those farther than 20 km from the coast which have very larger GAPs (typically 250°) are far from the stations and the velocity structure is complex and not well known. These are rough error estimates and probably they increase with decreasing magnitude of earthquakes.  31  CHAPTER 3. HYPOCENTRAL LOCATIONS  3.3.1 Epicentral distribution The final epicentres are shown in Figure 3.4. For this period (1983-1989), there are clear variations in seismic activity across the area. For the offshore seismicity near the shelf edge, more frequent activity and larger magnitudes are observed to the northwest than to the southeast For the seismicity onshore very little activity is detected in the northeast region, near Flores Island, and the highest frequency and larger magnitudes are observed to the southeast in the Barkley Sound area. These features do not seem to be artifacts of the station distribution since high activity is detected offshore to the northwest where station distribution is poor and few earthquakes are located onshore to the northeast where it is comparatively excellent. Also the sparse activity near the shelf edge in the southwest appears to be real since the network has the capability to detect earthquakes of magnitudes as low as to 1.2 (see Chapter 2). To check if the low seismicity near the shelf edge to the southwest could be an effect of the depth limit of 25 km chosen for the data set, a search was made for earthquakes in the area in the range 20-25 km depth in the CEC and in the earthquakes relocated shallower than 25 km. Fifteen earthquakes were found, the maximum magnitude 1.7, and 13 of them located less than 25 km from the coast The distribution of seismicity remained the same near the shelf edge. Then, from the epicentral distribution, it is observed that the band of seismicity in the 25-55 km range is not aligned parallel to the shelf edge. This feature will be examined in the next section. 3.3.2 Hypocentral locations In Figure 3.5 the earthquakes divided in four depth intervals (25-30, 30-35, 35-40, 40-45 km) have been plotted. Except for the interval 25-30 km only data of the highest quality have been included. 32  Figure 3.4 Distribution of epicenters in the depth range 25-55 km in the study area in the period 1983-1989. Solid dots indicate the best constrained locations (see text for criteria). Abbreviations are as follows: NI=Nootka Island, MI=Meares Island, pS=Barkley Sound, JFS=Juan de Fuca Strait. Dashed line indicates approximate shelf edge line.  CHAPTER 3. HYPOCENTRAL LOCATIONS  -127  -126  -125  -124  Figure 3.5 Depth interval plots for earthquakes, a earthquakes between 25-30 km depth, b earthquakes between 30-35 km depth. Solid dots indicate the best constrained locations. In a solutions of all qualities are shown, in b only best constrained solutions (see text for criteria) are shown. NFZ=Nootka fault zone. Same abbreviations as in Figure 3.4 are used. 34  CHAPTER 3. HYPOCENTRAL LOCATIONS  -127  -126  -125  -124  Figure 3.5 (cont.) Depth interval plots for earthquakes, c earthquakes between 35-40 km depth, d earthquakes between 40-45 km depth. Only best constrained solutions are shown. Same abbreviations as in Figure 3.4 are used. 35  CHAPTER 3. HYPOCENTRAL LOCATIONS  Most of the hypocenters in the range 25-30 km are located offshore near Barkley Sound and Meares Island (Figure 3.5.a). Four small earthquakes are located near Nootka Island, the northernmost probably occurring within the North American plate and the ones offshore on the projection of the Nootka fault. This band is not defined to the north of Flores Island. Hypocentres in the depth range 30-35 km (Figure 3.5.b) are east of the offshore band in the 25-30 km range and the distribution has similar orientation. Activity is concentrated also around Barkley Sound and Meares Island. Two small earthquakes are located near Flores Island. The 35-40 km distribution (Figure 3.5.c) follows the same pattern as the two previous intervals: strikes north-northwest, is shifted to the east of the previous and is more concentrated around Barkley Sound. Between 40-45 km O^igure 3.5.c) only four well located, small magnitude earthquakes were detected. The two to the west overlap with the 35-40 km distribution. Two clear features are evident from the hypocenter distributions: a band of seismicity striking north-northwest, deepening to the northeast and a high concentration of activity near Barkley Sound area. Two cross-sections separating the earthquakes to the northwest and to the southeast have been plotted along lines parallel to the subduction direction (\ines AA' and BB' in Figure 3.6). They are shown in Figure 3.7, solid dots indicate the best constrained hypocentral locations (see Section 3.3 for criteria). Section along AA' (Figure 3.7.a) defines a thin slab of seismicity, about 9 km thick and dipping 12° ± 4° in the direction of AA'. The distribution is best defined to the NE of the 60 km mark along AA'. A significant number of earthquakes with well constrained depths and some additional, not so well constrained buttightlyclustered, earthquakes delineate the distribution very clearly. 36  Figure 3.6 Areas plotted in cross-sections in Figures 3.7 and 3.8. All earthquakes within each area have been projected onto vertical planes oriented along AA', BB', CC and DD'. Dashed line indicates approximate shelf edge line. Dot-dashed lines show cross section locations.  CHAPTER 3. HYPOCENTRAL LOCATIONS  (NE) A'  A (SW) -i  r~  o  e  0_. A  .o° OH  w Q  60  |  0  ,  ,  , __,  50  1  L _  100 WIDTH (km)  (NE) B'  B (SW) • •  a  o°  E  O 86©  w  •  °°° -• °o °  ^  • •• ^  .  -  o  b 60  0  1  ,  ,  50  .  .  1  100  .  1  WIDTH (km)  Figure 3.7 Cross-sections of seismicity in the range 25-55 km depth projected onto planes parallel to the subduction direction (no vertical exaggeration), a corresponds to earthquakes in the southeastern area projected along AA'and b corresponds to earthquakes in the northwestern area projected along BB' (Figure 3.6). Solid dots indicate best constrained solutions. Squares are earthquakes shallower than 25 km, not examined here. Dotted line correspond to 25 km depth. Dashed lines show the inferred position of the brittle structure in the plate.  38  CHAPTER 3. HYPOCENTRAL LOCATIONS  The earthquake distribution along BB' shows a similar thin band of earthquakes as in the section along AA', to the NE of the 50 km mark (see Figure 3.7.b). The band is about 8 km thick, dipping 14° ± 4° in the direction of BB' and it is about 8 km deeper than the band in cross section along AA'. This band cannot be unambiguously delineated due to the limited seismic activity in the area, but the high quality of the hypocentral locations contributes to constrain its shape. The largest earthquake in this cross section is apparently at the interface and this is also observed in cross-section CC, Figure 3.8. Offshore, to the SW of the 60 km model distance in AA' and of the 50 km model distance in BB', the distributions are not well defined. Small magnitudes and a poor distribution of stations for the earthquakes offshore result in larger uncertainties in hypocentres and thus, in the disperse distribution of seismicity in this area. Consequently the bend in the slab indicated by the seismicity along AA' and BB' is unlikely to be real. Two additional cross sections were plotted along lines parallel to the shelf edge (lines CC and DD' in Figure 3.6) projecting the earthquakes along corridors EFGH and HGJJ, they are shown in Figure 3.8. The earthquake distribution in both sections exhibit a thin band of seismicity, dipping 4° ± 2° along the CC and DD' directions, which is particularly well defined SE (notice vertical exaggeration 2:1 infigure3.9). To the NW the band cannot be traced due to the low level of activity but the few well located earthquakes are compatible with this geometry. From the apparent dips determined in the cross sections, 13° ± 4° along the subduction direction and 4° + 2° parallel to the shelf edge, we determine the 'true' dip, 14° ± 4°, and strike, N29°E + 5°, of the seismicity slab.  39  CHAPTER 3. HYPOCENTRAL LOCATIONS  C (NW)  (SE) C  10  TJ—  •  •  20 -  x  30  E-"  au  w Q  o 40  50  °m  . - - -r- -I  1  1  • ..1 •  50  100 LENGTH (km)  150  D (NW)  200  (SE) D'  10 -I  1  1  1  20 p .  X  30  a. w p  •  — — V  ~5~«"  40  50  50  100 LENGTH (km)  150  Figure 3.8 Cross-sections of seismicity in the range 25-55 km depth projected onto planes parallel to the shelf edge line, a corresponds to earthquakes in the corridor EFGH projected along CC and b corresponds to earthquakes in the corridor HGLT projected along DD'. Solid dots indicate best constrained solutions. Squares are earthquakes shallower than 25 km, not examined here. Dotted line corresponds to 25 km depth. Inferred position of the brittle structure in the plate is indicated by a dashed line where it is more certain and by a short dashed line where it is uncertain. 40  200  CHAPTER 3. HYPOCENTRAL LOCATIONS  3.3.3 Relocation of some earthquakes offshore  The calculated depths for some of the earthquakes offshore, those 30-^40 km from A in Figure 3.7.a, and 10-40 km from B in Figure 3.7.b, are in excess of what is expected from the position and thermal state of the plate. Based on the age of the plate, Rogers (1983) estimated the thickness of the plate when it starts to subduct as 30 km and the brittle thickness as 10 km from the order of seismic rupture depth for earthquakes in the oceanic lithosphere west of Vancouver Island (Hyndman and Rogers, 1981). Assuming a temperature of 450°C at the top of the plate at 20 km depth and a temperature gradient within the oceanic lithosphere of 35°C/km from Lewis et al. (1988), earthquakes within the plate in this area are expected to occur at depths less than 30 km to be consistent with the 300°C-700°C temperature range at which the plate is rigid enough to deform by brittle fracture (Rogers, 1983). The calculated depths of earthquakes in this area within 33 to 43 km raise doubts about the reliability of these solutions. To gain more insight into the solutions three large earthquakes in the area (39, 91, 107) were relocated by fixing the depths to determine the RMS residual error of the hypocenter solutions as a function of depth. Velocity structure of model 3 (Drew and Clowes, 1990) was used for the relocations. The results summarized in Figure 3.9 show well defined minima at a depth of 33 to 35 km for all three earthquakes. The solid symbols show the hypocenters determined using a free depth. Thus, this velocity model and station distribution provide an apparently well defined depth. However the GAPs are 228° to 256° and although the closest station is located about 25 km from the epicenters, the next closest is about 80 km distant. Thus, with this station distribution and uncertainties in the velocity model, these depths are not well constrained.  41  CHAPTER 3. HYPOCENTRAL LOCATIONS  RMS (sec)  Figure 3.9 RMS residuals of the hypocentral solutions as a function of the assumed depth for offshore earthquakes 39, 91and 107. Solid symbols are the values obtained for the same earthquakes located with crustal model 3 and free depths.  42  CHAPTER 3. HYPOCENTRAL LOCATIONS  3.4 Plate geometry The depth distribution and cross section analyses clearly show a gently dipping zone of seismicity beneath south-central Vancouver Island extending about 60 km down-dip at a distance range of 35 to 95 km from the shelf edge and at a depth of 25 to 45 km. This zone, the shallow Benioff zone of seismicity embedded within the Juan de Fuca plate, constrains the shape and position of the subducting plate between the Nootka fault on the northwest and the Juan de Fuca Strait on the southeast. Figure 3.10 shows a schematic of the contour depths to the top of the seismogenic zone. It is based on the depth distributions in Figure 3.5, on the cross sections of figures 3.7 and 3.8 and on the dip calculated from them. Except for the 25 km contour only well constrained hypocenters (as defined in the introduction of Section 3.3) have been used. The position of the top of the seismogenic zone is consistent, within the uncertainties in hypocentral locations, with the position of the downgoing oceanic crust determined by Hyndman et al. (1990) for central Vancouver Island from reflection data (Figure 3.11). The only significant difference is in the 25 km contour which is about 12 km to the southwest relative to the corresponding contour from reflection data and can be explained by the poorly constrained solutions used to define this contour. A difference of about 20° between the azimuth of the approximated shelf edge line (considered to be parallel to the subduction trench) and the azimuth of the contour depths, observed infigure3.10, indicates upward arching of the slab from northwest to southeast beneath central-south Vancouver Island as suggested previously by Rogers (1983) and Cassidy (1991). The relatively small thickness of the seismicity slab, about 9 km, is consistent with the young age of the Juan de Fuca plate which restricts the low temperature zone where earthquakes can occur to a thin core of the plate as pointed out by Rogers (1983) and 43  CHAPTER 3. HYPOCENTRAL LOCATIONS  Figure 3.10 Contour depths in kilometers to the top of the brittle zone as inferred from seismicity in this study. Dashed line indicates uncertain position.  44  CHAPTER 3. HYPOCENTRAL  LOCATIONS  Figure 3.11 Contour depths in kilometers to the top of the downgoing oceanic crust from reflection data in the southern Vancouver Island area (from Hyndman et al., 1990). The geometry of the brittie slab is consistent with the position of the subducting plate except for the 25 km contour due probably to uncertainties in hypocentral locations used to define it.  45  CHAPTER 3. HYPOCENTRAL LOCATIONS  Hyndman et al. (1990). The decreasing magnitudes and frequency of activity toward the northeast, apparent in cross section a in Figure 3.7, most probably reflects warming of the slab as it deepens (e.g., Keen and Hyndman, 1979; Rogers,1983).  3.5 Discussion The plate geometry indicated by seismicity agrees with that expected beneath south-central Vancouver Island on the basis of the volumetric problem caused by the change in strike"of the subduction zone from almost NS off the Washington coast to approximately N30°W off the Vancouver Island coast (Keen and Hyndman, 1979; Rogers, 1983). The orientation of the dip of the plate calculated here, 14°±4° at N29°E±5°, agrees with and is better constrained than that of 15°±5° at N30°E±20° as calculated by Cassidy (1991) for the plate beneath central Vancouver Island based on receiver function studies. The dip direction compares with N56°E for the Juan de Fuca plate motion. This orientation is consistent with the arched geometry model of the plate suggested by Rogers (1983) as the way of deformation of the plate to accommodate the space problem due to the bent margin. In this model the plate is folded in the vicinity of Puget Sound and the limbs of the fold dip southeasterly to the south of Puget Sound and northeasterly under south-central Vancouver Island. This model has received support from various earthquake hypocenters and receiver function studies in Washington and central-south Vancouver Island (Crosson and Owens, 1987; Weaver and Baker, 1988; Owens et al., 1988; Lapp et al., 1990; Cassidy, 1991). Based on our results, on studies of hypocentral locations of deep earthquakes beneath Washington and northern Oregon by Taber and Smith (1985) and Weaver and Baker (1988) and by Rogers et al. (1990) along a corridor crossing Vancouver Island, we have sketched in Figure 3.12 the geometry of the Juan de Fuca plate inferred from seismicity beneath 46  CHAPTER 3. HYPOCENTRAL LOCATIONS  \Z5°  Figure 3.12 Summary of plate geometry beneath western Washington and Vancouver Island inferred from seismicity, based on this and Rogers et al. (1990) studies for Vancouver Island and on Taber and Smith (1985) and Weaver and Baker (1988) for western Washington. Dip direction estimates (solid arrows) from receiver function analysis given by Cassidy (1991), Owens et al. (1988) and Lapp et al. (1990) are also shown. Contour depths in kilometers.  47  A (SW) —,  ,  .  ,  1  ,  a  i  • . "i  .  g  .  .  >  '  I  '  '  '  (NE) A' 1  •  o o  _ •  u  o  0°" o o  50  100 WIDTH (km)  150  200  Figure 3.13 Cross-section along AA" (Figure 3.6) including seismicity in cross-section AA' (Figure 3.7.a) and seismicity in the depth range 55-70 km (solid squares) given by Rogers et al. (1990). Shallow seismicity beneath Georgia Strait is not shown.  CHAPTER 3. HYPOCENTRAL LOCATIONS  (NE) A'  A (SW) -n  1- 1  r-  ._r-  :  \  6 x  30  OM  E-  w Q  60  50  WIDTH (km)  100  B (SW)  (NE) B'  1 -  —I  6 x  30  o 0  — X. o  o° ^  Or  E-<  a, w a  o  D-  oc  OM  60  50  _J  L_  WIDTH (km)  100  Figure 3.14 (a) Overlay of a section of Drew and Clowes (1990) structural model and the same earthquake hypocenters of cross-section AA' (Figure 3.7.a). (b) Overlay of seismicity in cross-section BB' OFigure 3.7.b) and the subducting plate position beneath Flores Island determined by Dehler (1991). Solid dots indicate best constrained solutions (see text for criteria). Abbreviations are OC=oceanic crust, OM=oceanic mantle. 49  CHAPTER 3. HYPOCENTRAL LOCATIONS  Washington and central-south Vancouver Island.  Dip direction estimates from receiver  function studies given by Cassidy (1991) are also shown. Combining the hypocentral locations at depths of 60 to 70 km of Rogers et al. (1990) for a group of earthquakes beneath the Strait of Georgia with hypocentral locations of earthquakes in cross-section AA' in Figure 3.7.a there is an indication that the plate dips at a steeper angle, near 18°, at depths between 40-60 km beneath central Vancouver Island (see Figure 3.13). A comparison of hypocentral locations in cross-section along AA' with the structural model of Drew and Clowes (1990) and of cross-section along BB' with the position of the subducting plate determined from a magnetic and gravity model along a profile parallel to the plate convergence direction crossing about 5 km to the north of Flores island, line 2A of Dehler (1991), shows that most of the deep earthquakes are occtirring in the oceanic crust (Figure 3.14). Some earthquakes offshore, apparently occurring in the upper mande, have depths poorly constrained and are probably occurring shallower. More precise locations for seismicity in this area cannot be improved substantially without the deployment of ocean bottom seismograph arrays.  50  CHAPTER 4  FOCAL MECHANISM SOLUTIONS  4.1 Introduction  Tne small magnitudes, relatively deep focus and the station distribution, all make difficult to determine a well constrained focal mechanism solution from a single earthquake. An average magnitude of 1.5, an average gap (largest azimuthal separation between stations as seen from the epicentre) of 180° and a minimum gap of 95° for our set of earthquakes, and the fact that of the almost 130 earthquakes only 21 have 6 or more reliable first motion observations illustrate well the limitations. To overcome these difficulties, a composite focal mechanism approach was used. In this procedure, first motion observations for more than one earthquake are combined under the assumption that the earthquakes share a similar mechanism. This assumption is justified in this study by the fact that earthquakes used in each solution are closely located. In the composite method different station distributions with respect to the focus may constrain a solution. This approach may provide a picture of the regional stress distribution. The azimuths and take-off angles required to plot thefirstmotion pattern were calculated with HYPOELLIPSE using the same velocity model as for hypocentral locations (Section 3.3). The solutions were obtained from plots of the first motions on a lower hemisphere equal area projection of the focal sphere by visual fitting. 51  CHAPTER 4. FOCAL MECHANISMS  In general those earthquakes with more than six observations coincided with the largest events in the set and/or were located in a region of good station coverage.  Initially,  first motion distributions for 19 earthquakes with magnitudes greater or equal to 2.0 were examined and 4 tentative mechanisms were delineated.  Various approaches were made  to obtain additional information to better constrain these mechanisms: observations from original digital records were checked, first arrivals from smaller earthquakes were included and plotted jointly with those of the larger ones, and observations from distant stations were used. The results of these tests showed that using small events with a few observations ended frequently in patterns inconsistent with those determined by the larger events. This also occurred when the hypocenters of the small earthquakes were close to those of the larger earthquakes. This observation suggests local heterogeneities in the stress field. The same effect occurred when the epicentre-station distance was more than about 150 km. Further, it was found that events in the period 1983-1984 showed inconsistent observations, suggesting polarity problems during the network setup phase. As a result of these tests, only data from earthquakes in the period 1985-1988, each with at least 6 clear first motions, were used. Twenty earthquakes with local magnitudes from 1.2 to 3.5 fitted these conditions. 4.2 Composite focal mechanism solutions The composite focal solutions and the set of earthquakes used to obtain each solution are shown in Figure 4.1 and listed in Table 4.1. The best solutions have at least one nodal plane with strike and dip constrained to within 20°. These are shown in Figure 4.1 in solid lines. Detailed first motion observations on which the solutions are based are shown in Figure 4.2. Table 4.2 provides the locations and previously determined focal solutions for five large earthquakes in the region, which are relevant for the discussion . These epicenters and mechanisms are plotted in Figure 4.3. 52  CHAPTER 4. FOCAL MECHANISMS A  0 km  50 km  30 km -  60 km \ !  50  A \ \ ~\  '.If/  :  i  -  V  JO  I  O AiU^\'- ^ v  BR  49  km  48  -127  50 —i  -126  -125  -124  Figure 4.1 Equal area, lower focal hemisphere composite focal mechanism solutions for five groups of earthquakes. Groups correspond to the list in Table 4.1 and numbers to individual earthquakes in Figure 4.2. Solutions show normal and strike-slip faulting. Dashed solutions are weakly constrained. Solid pentagons indicate earthquakes used in the solutions. Shaded areas indicate compression. Dotted line is the northeast projection of the Nootka fault. Cross section of hypocenters along AB is shown at top. 53  TABLE 4.1 Composite focal mechanism solutions  MECHANISM GROUP  a  b  c  d  e  EVENT  39  DEPTH  MAGNIT.  Azimuth  Strike  Dip  2.9  N5°W±15°  10°NE±20°  N84°E±5  78°N+10°  (km)  [ML)  34.6  91  33.1  2.3  107  35.2  2.9  100  35.8  1.2  123  36.8  2.5  24  33.1  2.4  92  34.9  2.0  77  39.1  1.9  78  37.3  1.6  23  32.6  2.0  44  36.4  3.0  49  37.3  3.2  53  32.0  3.5  98  33.8  1.5  117  32.0  2.3  122  29.7  2.4  TAXIS  P AXIS  NODAL PLANES  Dip  Azimuth  Dip  310°±15°  2°±20°  89°±10°  24°±35°  220°±15°  16°±20  N48°W±10° 70°SW±35° 78°NW±30 N42°E±5°  355°±10°  10°±35  N28°W±10° 52°NE±20° 48°SW±10« S74°E±5  302°±10°  68°±20°  32°±10°  2° ±20°  N12°W±20° 80°W±30° N48°E±10° 20°SE±10°  62°±20°  50°±30°  268°±20°  32°±30°  N34°W±10° 80°NE±20° N52°E±5° 70°NW±8°  186°±10°  20°±20°  96°±10°  8°±20°  0  C  0  0  ...  0  Table 4.1 Grouping and earthquakes used in the composite focal solutions and details of the solutions. Event number corresponds to general list of earthquakes in Appendix A. Depths of earthquakes in group a are weakly constrained.  CHAPTER 4. FOCAL MECHANISMS  GROUP A: Observations from three earthquakes with magnitudes 2.3 to 2.9 located offshore Vancouver Island (earthquakes 39, 90 and 107) were used to derive the composite focal mechanism for group a. The number of observations is enough to clearly define afirstmotion pattern characteristic of strike-slip faulting (Figure 4.1). The tension and compression axes have shallow dip (Figure 4.2). Earthquakes of group a are located in a zone of frequent seismic activity. Their focal depths are poorly constrained (Section 3.3.3) and due to uncertainties in these depths it is not possible to establish whether they are in the subducting plate. However a change in the focal depth of these earthquakes does not change the nature of the mechanism solution. A possible relation between earthquakes of group a and the inferred subducted Nootka fault is suggested by the similarity in the type of faulting, strike-slip, and orientation of the P and T axes between these events and two large earthquakes (December 16, 1957 and July 5, 1972), which were located near the northeast projection of the Nootka fault (see Figure 4.3 and Table 4.2). Mechanism solutions for these large earthquakes indicate dominant strike-slip faulting. The sense of motion is sinistral on the northeast striking plane or dextral on the northwest plane (Rogers, 1983; Cassidy et al., 1988). Rogers (1983) supported the hypothesis that these large earthquakes were located in the continental plate and were a result of the stress regime generated by the interaction of the Explorer plate with the America plate. Cassidy et al. (1988), based on improved locations, focal mechanisms and stress drops, reinterpreted them as interplate earthquakes, the result of left-lateral motion along the Nootka fault as it subducts beneath Vancouver Island. Though the type of focal mechanism for group a suggests an association with the subducted Nootka fault, orientation of nodal planes and epicentral locations do not. Strike of the nodal planes are N65°E and N26°W for the 1957 earthquake (Cassidy et al., 1988), 55  CHAPTER 4. FOCAL MECHANISMS  N64°E and N26°W for the 1972 earthquake (Rogers, 1983), and N84°E±5° and N5°W±15° for  group a.  The difference of about 20° in the strike of the planes oriented NE between the  large earthquakes and  group a solution may be significant. Additionally, epicentral locations  show that they are 20-50 km to the southeast of the northeast projection of the fault, too far to be associated directiy with slip along it. It is still possible that these events are indirectly related to the Nootka fault. They may result from the redistribution of stresses in the region adjacent to the fault and could be associated with branching splay faults or to reactivated zones of weakness on the oceanic plate, derived from the complex history of interactions between the Juan de Fuca and Explorer plates (Hyndman et al., 1979). GROUP B: The focal mechanism for  group b was  obtained using two earthquakes of  magnitudes 1.4 and 2.5 (earthquakes 100 and 123) in a zone where few events were detected in this study. Although the number of observations is small, the distribution of compressions and dilatations in the projection of first arrivals is indicative of a dominant strike-slip mechanism. The nodal planes are near vertical. The pressure axis is oriented almost north-south and the tension axis east-west Both axes have shallow dip (Figure 4.2). Strike-slip mechanisms for  group b may indicate internal shearing on the slab, although  the mechanism responsible for it is unclear. Various mechanisms are proposed: (1) redistribution of stresses in the region adjacent to the Nootka fault (similar to  group a); (2) variations  along the margin in the strength of coupling of the subducted plate with the North America plate; and (3) stresses generated from deformation of the plate due to the change in direction of the margin. The latter will be discussed when the solution of  group e is analyzed.  On the large scale the second mechanism is the least plausible. The strength of coupling at a subduction interface has been correlated positively with the convergence rate and negatively with slab age (Ruff and Kanamori, 1980; Kanamori, 1986). Under south to central Vancouver 56  CHAPTER 4. FOCAL MECHANISMS  Island the age of the subducting plate is very similar, about 10 my (Hyndman et al., 1979), and the difference in convergence velocity between south and central Vancouver Island of about 0.2 cm/yr (Riddihough, 1984) does not seem to be large enough to cause shearing on the slab as it subducts. GROUP C: Solution for group c is acceptably constrained. Two earthquakes of magnitude 2.0 and 2.4 (earthquakes 24 and 92) define the focal mechanism. The distribution of first motions with dilatations at the center and compressions at the periphery (Figure 4.2) is characteristic of a normal faulting mechanism. The tension axis is nearly horizontal and is oriented N32°E. The strike of the T axis is, within uncertainties, in the direction of dip of the subducting slab determined in Chapter 3 (see Figure 4.4). This suggests that the down-dip direction of the slab is the factor controlling the orientation of the T axis. The presence of earthquakes with normal mechanisms in this zone may indicate bending stresses due to the change in the dip of the plate as it deepens beneath east-central Vancouver Island (see Section 3.5). The nodal plane oriented N28°W dipping N52°E, which is chosen as the fault plane is consistent with this interpretation. GROUP D: Two events of magnitudes 1.9 and 1.6 located near the deeper edge of the seismic zone (earthquakes 77 and 78).studied here constitute group d. Although the observations do not provide a well constrained mechanism (see Figure 4.2) a solution consistent with normal faulting is indicated. The pressure axis is oriented east of north and the tension axis east-west. The orientation of the P and T axis is significandy different from those of the other solutions (see Figure 4.4) and, in particular,fromthe solution of group c, which exhibits normal faulting also. The principal axes show no correlation with tectonic stresses in the region. It may be that events of group d reflect more complex stresses developed within the 57  CHAPTER 4. FOCAL MECHANISMS  9  r o u  P  group d  c  Figure 4.2 Detailed first motion observations for the five group of microearthquakes of Figure 4.1 and Table 4.1 and the events discussed in the text with insufficient number of observations to provide a constrained solution. Projections are equal area and lower hemisphere. Solid dots indicate compressions; open dots, dilatations; solid diamonds, pressure axis; open diamonds, tension axis. Dashed solutions are weakly constrained. 58  CHAPTER 4. FOCAL MECHANISMS  group e  earthquakes 81 and 103  earthquake 43  earthquake 50  Figure 4.2 (cont.)  59  CHAPTER 4. FOCAL MECHANISMS  Figure 4.3 Focal plots for significant earthquakes in the Juan de Fuca plate beneath British Columbia and Washington. Earthquake locations correspond to list in Table 4.2. Double arrow represents subduction direction. (Modified from Baker and Langston, 1987). 60  TABLE 4.2 Significant earthquakes in the Juan de Fuca plate under British Columbia and Washington  MECHANISM DATE  April 13, 1949  1  Dec 16, 1957  2  April 29, 1965  July 5,1972  3  4  May 16,1976  LOCATION (°)  DEPTH (km)  MAGNITUDE  47.2N  54  M=7.1  122.62W  intraplate  49.6N  30  127.0W  interplate  47.4N  59  122.3W  intraplate  49.5N 127.3W  4  48.8N 123.2N  1  Baker and Langston, 1987.  2  Cassidy et al., 1988.  3  Isacks and Molnar, 1971. -  4  Rogers, 1983  Type  P axis  Taxis  Azimuth  Dip  Azimuth  Dip  strike-slip  245°  -20°  136°  -16°  M =5.9  strike-slip  203°  18°  106°  -21°  m =7.1  normal  274°  -60°  63°  26°  25 interplate  mb=5.7  strike-slip  200°  10°  108"  13°  62 intraplate  M =5.4  normal  215°  -44°  78°  -16°  s  s  b  L  CHAPTER 4. FOCAL MECHANISMS  plate by folding or bending or by phase change processes. A similar phenomenon of a group of earthquakes showing principal axes reversed from most of the rest of events has been observed for deep seismicity (depths from 35 to 55 km) in the Juan de Fuca plate beneath the Olympic Peninsula in Washington (Taber and Smith, 1985). There, while the average azimuth of the T axis is in the direction of plate convergence, four earthquakes have P axes in almost the same direction. GROUP E: The composite focal mechanism for group e was determined using data from seven earthquakes with magnitudes 2.0 to 3.5 (earthquakes 23, 44, 49, 53, 98, 117 and 122). With these relatively large magnitudes and a good station distribution, a very well constrained solution was obtained. Five smaller earthquakes not used in the solution, also showed arrivals which are compatible. The mechanism is primarily strike-slip on an almost vertical plane. This focal mechanism suggests lateral shearing within the slab. The very shallow pressure axis (20°) is not oriented in the plate convergence direction. The strike of the T axis, N96°E, is significantly different from the dip direction of the plate inferred from hypocentral locations in Section 3.3.2 (N29°±5°E). Also it is about 30° from the dip direction of the T axis of the May 16, 1976 Pender Island earthquake. This was an intraplate event of magnitude Af^=5.4, located at 62 km depth and exhibited normal faulting (Rogers, 1983). For its magnitude, it is likely representative of the regional state of stress. Strike-slip faulting mechanisms within the plate have been also observed beneath Washington for small earthquakes in a depth range 43-52 km by Taber and Smith (1985), but only one shows a similar P and T axis orientation. The large April 13, 1949 earthquake, (Mj=7.1), located at 54 km depth (see Figure 4.3 and Table 4.2) also showed strike-slip faulting (Baker and Langston, 1987). They provide additional evidence of compression acting within the shallow dipping portion of the slab. 62  CHAPTER 4. FOCAL MECHANISMS  Figure 4.4 Position on the lower focal hemisphere of the axes of compression (solid circles) and tension (open circles) for composite focal mechanisms of Figure 4.1 and Table 4.1. Dashed line shows projection of the subducted plate interface inferred in Chapter 3. Arrows indicate directions of interaction of the Juan de Fuca and America plates (JF/A) and Explorer and America plates (EX/A). Letters a to e refer to the five groups of earthquakes.  63  CHAPTER 4. FOCAL MECHANISMS  OTHER EARTHQUAKES: Near to groups a to e there are individual earthquakes which have at least 6 clear first motion observations, for example earthquakes 43, 50, 81 and 103, shown also in figures 4.1 and 4.2. Comparison of their first motion observations with the mechanisms of the nearby constrained solutions (earthquakes 81 and 103 with group e, 43 with group c, and 50 with group a) shows inconsistency. This observation suggests local heterogeneities in the stress field.  4.3 Discussion The composite fault plate solutions, although some are weakly constrained, indicate normal faulting and strike-slip mechanisms. There is no indication in these data of shallow angle thrusting mechanisms along the plate interface. One solution showing normal faulting is consistent with down-dip tension. This solution is suggestive of bending stresses as the dip increases.However the strike-slip solutions are an unexpected result of this study and may indicate internal shearing of the slab. Rogers et al. (1990) have examined earthquakes within the descending plate along the LITHOPROBE southern Vancouver Island corridor. They find two groups of events, one consistent with down-dip tension and the other with down-dip compression. In contrast, group e from this area shows strike-slip motion. This discrepancy may be explained in part by differences in the data set Their data set includes small earthquakes and uses events as far from the shelf edge as - 180 km where the depths are near 70 km. For our solutions all the T axes but one dip slightly, indicating within uncertainties inplate tension. Their strikes vary over a range of azimuths and only one is consistent with the down-dip direction of the plate. The compression axes show a wide range of azimuths and dips. The diversity in the orientation of the tension and compression axes and a lack of 64  CHAPTER 4. FOCAL MECHANISMS  a clear association with the tectonic stress regime reflected by the large earthquakes in the region, suggest that seismicity in the range 25-55 km depth reflects more complex stresses involved in the subduction process. Proposed explanations for such complexity are many: the complicated geometry of the plate as it bends (archs in the dip direction) and folds (archs in a direction parallel to the margin), phase changes from gabbro to eclogite which are likely to occur at this depth in the plate, and interactions at the interfaces and boundaries of the slab. No single mechanism can explain all the focal solutions observed and it is likely that they are superimposed. A similar broad distribution of principal axes also has been observed for seismicity in the JdF plate beneath Washington (Taber and Smith, 1985; Ma et al., in prep.) and it has been associated with variations in the slab stress state due to the arched plate geometry (Ma et al., in prep.).  •  ,  Increase in the dip of the Juan de Fuca plate at about 40 to 70 km depth is suggested from hypocentral locations (Section 3.5) and in several studies (Rogers, 1983; Weaver and Baker, 1988, Cassidy, 1991) and bending of the plate has been inferred. As a result of bending, tensional stresses in the upper part of the plate and compressional at the base are likely to develop. Results from hypocentral locations derived in Chapter 3 are supportive of an arched geometry (termed here folding) model for the subducting plate beneath western Washington and British Columbia (see section 3.4). This geometry probably complicates the pattern of stresses in the plate and should be considered as one of the factors controlling the orientation of the principal axes. Strike-slip folding mechanisms frequently are found in association with folded structures (Billings, 1972) and are likely to form when faulting is caused by horizontal pressure (Jaroszewski, 1984), although they do not represent the typical stress system. The stress r  65  CHAPTER 4. FOCAL MECHANISMS  pattern of folded slabs is a function not only of the flexure of the slab, but also of the underlying forces and the circumstances accompanying folding. In the case of folding by horizontal compression proceeding under considerable weight of overburden the formation of vertical or high angle strike-slip faults in an oblique position with respect to the axis of the fold may be expected (Jaroszewski, 1984). Thus the idea that the strike-slip mechanisms observed are a result of stresses generated in the folding process appears feasible. Phase changes from grabbro to eclogite in the oceanic crust and from plagioclase-to spinel and garnet in the upper mande take place at depths from 30 to 70 km as the oceanic lithosphere descends into the mantle (Pennington, 1984) and have have been suggested as the origin of the concentration of seismicity under Texada Island at about 60-70 km depth (Rogers, 1983; Rogers et al., 1990). Thus, some of the earthquakes used to calculate our focal mechanism solutions, which are in a range 30-40 km depth, may originate in this way. In other subduction zones (New Zealand, Robinson (1978) and Reyners (1980); Alaska, Reyners and Coles (1982)) observations of earthquakes at a similar depth range show that a common feature of focal mechanisms is their large diversity and the lack of a consistent pattern except for down-dip extension among a few events (Rogers,1983). A similar feature is observed in the focal mechanisms of this study. In the section beneath central-south Vancouver Island, the Juan de Fuca/America interaction is normal subduction (Riddihough, 1984); i.e., the direction of the interaction vector is orthogonal to the boundary. However, a deep Benioff zone of earthquakes and large thrust earthquakes have not been observed in this area nor, in general, in the Cascadia subduction zone (Rogers, 1983; Crosson 1983; Heaton and Kanamori, 1984; Ludwin et al., preprint). Our results from hypocentral locations and composite mechanism solutions confirm the previous observations at the small magnitude level. Seismicity in the range 25-55 km beneath 66  CHAPTER 4. FOCAL MECHANISMS  the continental margin of Vancouver Island is confined to the subducting slab and not to the interface (see Chapter 3) and no thrust mechanisms have been observed. The question of whether this is the result of aseismic slip along the interface or of the plates being locked and elastic strain accumulating in the upper plate to be eventually released in a great earthquake, has been widely discussed in recent years (e.g. Heaton and Kanamori, 1984; Rogers, 1983; Rogers, 1988; Weaver and Smith , 1983; Taber and Smith, 1985). However, recent results, e.g. of deformation studies in northern Washington by Savage and Lisowski (1991) support the hypothesis of a segment of the plate interface underlying the continental slope and outer continental shelf being locked but the remainder of the interface slipping continuously. If their model were applicable to the plate interface beneath central-south Vancouver Island, the set of earthquakes of this study would be located in the transition segment over which the slip deficit decreases continuously down-dip. Also results from stratigraphic studies of coastal lowlands (Atwater, 1987; Darienzo and Peterson, 1990) show evidence that great subduction earthquakes do occur in this region.  67  CHAPTER 5 SUMMARY AND CONCLUSIONS  The distribution of deeper earthquakes (> 25 km) along the west coast of Vancouver Island and immediate offshore region define a band of activity between the Nootka Fault to the north and the Juan de Fuca Strait to the south, almost parallel to the coast line and about 80 km wide. Magnitudes are in the 0.3-3.6 range. Seismicity is concentrated on the Barkley Sound and immediate onshore area and beneath the continental shelf west of Flores Island. In those areas the larger magnitudes (10 earthquakes between 2.5 and 3.6) are also observed. Magnitudes are smaller to the east and deep activity almost ceases at about 40 km from the coast at a depth of about 45 km, to reappear again beneath Georgia Strait at the 60-70 km depth range. Hypocentres, recalculated with a velocity structure determined from refraction-reflection data (Drew and Clowes, 1990), and using a maximum epicentral-station distance to 110 to 130 km, azimuthal separation of less than 180° and the more reliable arrivaltimes,provided a good number of earthquakes (about 80) with depth uncertainties of less than + 5 km. These earthquakes delineate clearly a gently dipping zone of seismicity beneath centralsouth Vancouver Island at 25 to 45 km depth, the shallow Benioff zone, associated with the brittle portion of the subducted Juan de Fuca plate. The brittle structure is about 9 km thick and is dipping 14°+4° at N29°E±5°. The geometry of this structure is in good agreement with the position of the downgoing oceanic crust determined by Hyndman et al. (1990) based on reflection data and other geophysical studies. These results better constrain the dip of the 68  CHAPTER 5. SUMMARY AND CONCLUSIONS  I plate under central Vancouver Island compared with the orientation of 15°±5° at N30°E±20° determined by Cassidy (1991) from receiver function studies. A comparison with the velocity structure used for locations (Drew and Clowes, 1990) and a gravity and magnetic model for the region (Dehler, 1991) shows that most of the activity is occurring in the oceanic crust. The geometry outlined by seismicity is supportive of the model of plate folding proposed by Rogers(1983) to accommodate the space problem due to the bent margin and which has received recent support from other studies (Crosson and Owens, 1987; Weaver and Baker, 1988; Owens et al., 1988; Lapp et al., 1990; Cassidy, 1991). In this model the axis of the fold would be located beneath the Puget Sound region and the limbs would correspond to the portions of the plate under central-south Vancouver Island and southern Washington. Composite focal mechanism solutions were determined for five groups of earthquakes. Three solutions show strike-slip faulting mechanisms and two normal ones. One solution showing normal faulting is consistent with down-dip tension and may suggest bending stresses due to the increase in the dip of the plate in this area. The presence of strike-slip faulting indicates that shearing is occurring within the slab. Principal stress axes inferred from the solutions show that all the T axes but one dip slightly (less than 21°), indicating in-plate tension. Their strikes vary over a range of azimuths and only one is consistent with the down-dip direction of the plate. The compression axes show a wide range of azimuths and dips. Wide scatter of P and T axes directions for intraslab seismicity suggestive of an inhomogeneous stress field also has been observed in western Washington (Ma et al., preprint).  The diversity in the orientation of the principal axes and a lack of a clear  association with the tectonic stress regime suggest that seismicity in this depth range reflects more complex stresses involved in the subduction process. These include bending stresses generated by increasing dip of the plate as it subducts or by fold formation to accommodate 69  CHAPTER 5. SUMMARY AND CONCLUSIONS  the excess of volume due to the change in the direction of the margin, stresses generated in volume changes due to the phase change from gabbro to eclogite which is likely to occur at this depth range (Rogers, 1983; Pennington, 1984), or local heterogeneities in the stress field. No single mechanism can explain all the focal solutions and it is likely that they are superimposed. Frequency-magnitude data were used to calculate the 6-value for this set of earthquakes. The calculated value of 0.79±0.07 is significandy different from a previous estimate (0.38 from Rogers, 1983) for central Vancouver Island which includes crustal and subcrustal earthquakes, but is in good agreement with values obtained for seismicity in a similar depth range in other subduction zones (e.g. 0.73 in the Puget Sound region (Crosson, 1981), 0.7 for Japan (Anderson et al., 1980) and 0.71 for the Mid America trench offshore Guatemala (Ambos et al., 1985)). In conclusion, we have found that: (1) Reprocessed earthquake locations have provided a refined picture of a subcrustal seismicity zone, the shallow Benioff zone, clearly separated from the shallower crustal earthquakes and not associated with the interface between the Juan de Fuca and North America plates. The structure defined by the brittle zone has both confirmed and extended the plate geometry suggested by other geophysical studies in the area. Hypocentral locations delineate a slab of seismicity beneath central-south Vancouver Island at a distance range of 35 to 95 km from the shelf edge and at a depth of 25 to 45 km, about 9 km thick and dipping 14°+4° at N29°E±5°. (2) The 6-value of 0.79 is also consistent with the conclusion that seismicity is occurring within the slab. (3) Composite focal mechanism solutions suggest strike-slip and normal faulting mech70  CHAPTER 5. SUMMARY AND CONCLUSIONS  anisms. Thus, the previous observations which indicated an absence of thrust focal mechanisms in the deeper seismicity are confirmed at the small magnitude level. P and T axes inferred from focal mechanisms show a wide range of orientations with the only predominant feature being that the T axes are shallow. This diversity suggests a more complex state of stress in the subducting slab than expected from the application of a simple subduction model which predicts down-dip tensional stresses at the upper zone of the slab, compressional stresses at its bottom and compressional stresses at the interface between the plates. Proposed explanations for such a complexity are the arched plate geometry, stresses generated by phase changes in the plate and interactions at the boundaries. The observations are likely to be a superposition of these effects. The state of stress in the subducted slab and its relation to stresses in the crust are still not clear. Further studies to determine the contribution of the already mentioned mechanisms to seismicity in the area should be performed. One approach which could contribute to our understanding is the use of spectral analysis of the digital data to determine the seismic moments and stress drops of the same set of earthquakes. Some improvement in the definition of the geometry of the plate in the offshore region west of Flores and Nootka Island requires the installation of ocean bottom seismograph arrays, which would provide a better station distribution and consequently more reliable hypocentral locations. The structure of the Juan de Fuca plate from seismicity beneath the Olympic Peninsula and Juan de Fuca Strait area is not well defined either. In this region, there is a low level of seismicity and, independently, an inadequate station distribution by the Western Canada Seismic Network and the Washington State Network. Analysis of this set of seismicity, matching data from both seismic networks, would probably provide an improved picture of the shallow Juan de Fuca plate.  71  REFERENCES  Ambos, E. L., Hussong, D. M., and Holman, C. E., 1985. An ocean bottom seismometer study of shallow seismicity near the Mid-America Trench offshore Guatemala. Journal of Geophysical Research, v. 90, p. 11397-11412. Anderson, R. N., Hasegawa, A., Umino, N., and Tagaki, A., 1980. Phase changes and the frequency-magnitude distribution in the upper plane of the deep seismic zone beneath Tohoku, Japan. Journal of Geophysical Research, v. 85, p. 1389-1398. Ando, M., and Balazs, E. I., 1979. Geodetic evidence of aseismic subduction of the Juan de Fuca plate. Journal of Geophysical Research, v. 84, p. 3023-3028. Atwater, B., 1987. Evidence for great Holocene earthquakes along the outer coast of Washington State. Science, v. 236, p. 942-944. Baker, G. E. and Langston, C. A., 1987. Source parameters of the 1949 magnitude 7.1 south Puget Sound, Washington earthquake as determined from long-period body waves and strong ground motion. Bulletin of the Seismological Society of America, v. 77, p. 1530-1557. Billings, M. P., 1972. Structural Geology. 3rd Ed., Prentice Hall Inc, 606 p. Cassidy, J. F., 1991. Teleseismic receiver function analysis of the crust and upper mande of southwestern British Columbia Ph. D. thesis, University of British Columbia, Vancouver, 174 p. Cassidy, J. F., Ellis R. M. and Rogers G. C , 1988. The 1918 and 1957 Vancouver Island earthquakes. Bulletin of the Seismological Society of America, v. 78, p. 617-635. Crosson, R. S., 1983. Review of seismicity in the Puget Sound region from 1970 through 1978. In Proceedings of Workshop XIV, Earthquake Hazards of the Puget Sound Region, Washington, USGS Open File Rept. 83-19, J.C. Yount and R.S. Crosson, Editors, United States Geological Survey, p. 6-18. Crosson, R. S. and Owens T. J., 1987. Slab geometry of the Cascadia subduction zone beneath Washington from earthquake hypocenters and teleseismic converted waves. Geophysical Research Letters, v. 14, p. 824-827. 72  REFERENCES  Darienzo, M. E. and Peterson, C. D., 1990. Episodic tectonic subsidence of Late Holocene salt marshes, northern Oregon central Cascadia margin. Tectonics, v. 9, p. 1-22. Dehler, S. A., 1991. Integrated geophysical modelling of the northern Cascadia Subduction Zone. Ph.D. thesis, University of British Columbia, Vancouver, 151 p. Drew, J. J and Clowes, R. M., 1990. A re-interpretation of the seismic structure across the active subduction zone of western Canada. In Studies of laterally heterogeneous structures using seismic refraction and reflection data, A. G. Green, editor, Geological Survey of Canada, Paper 89-13. Geiger, L., 1912. Probability method for the determination of earthquake epicenters from the arrival time only. Bulletin of St. Louis University, v. 8, p. 60-71. Gutenberg, B. and Richter, C. F., 1949. Seismicity of the Earth and associated phenomena. Princeton University Press, Princeton, N. J., 310 p. Heaton, T. H. and Kanamori H., 1984. Seismic potential associated with subduction in the northwestern United States. Bulletin of the Seismological Society of America, v. 74, p. 933-941. Hyndman, R. D., Riddihough, R.P. and Herzer, R., 1979. The Nootka fault zone - a new plate boundary off western Canada. Geophysical Journal of the Royal Astronomical Society, v. 58, p. 667-683. Hyndman, R. D. and Rogers, G. C , 1981. Seismicity surveys with ocean bottom seismographs off western Canada. Journal of Geophysical Research, v. 86, p. 3867-3880. Hyndman, R. D., Yorath, C. J., Clowes, R. M. and Davis, E. E., 1990. The northern Cascadia subduction zone at Vancouver Island: seismic structure and tectonic history. Canadian Journal of Earth Sciences, v. 27, p. 313-329. Isacks, B. and Molnar, P., 1971. Distribution of stresses in the descending lithosphere from a global survey of focal mechanism solutions of mantle earthquakes. Reviews of Geophysics, v. 9, p. 103-174. Jaroszewski, W., 1984. Fault and fold tectonics. British Micropaleontological Society Series. Ellis Horwood Ltd., Chichester, 565 p. Kanamori, H., 1986. Rupture process of subduction zone earthquakes. Annual Review of Earth and Planetary Sciences, v. 14, p. 293-322.  73  REFERENCES  Keen, C. E. and Hyndman, R. D., 1979. Geophysical review of the continental margins of eastern and western Canada. Canadian Journal of Earth Sciences, v. 16, p. 712-747. Lahr, J. C , 1985. HYPOELLJPSE/VAX: a computer program for determining hypocentral parameters, magnitude andfirstmotion patterns. United States Geological Survey, Open File Report 84-519. Lapp, D. B., Crosson, R. S. and Owens, T. J., 1990. P-waveform analysis for local subduction geometry south of Puget Sound, Washington. Pure and Applied Geophysics, v. 133, p. 349-365. Lewis, T. J., Bentkowski, W. H., Davis, E. E., Hyndman, R. D., Souther, J. G., and Wright, J. A., 1988. Subduction of the Juan de Fuca Plate: Thermal consequences. Journal of Geophysical Research, v. 93, p. 15207-15225. Ludwin R. S., Weaver C. S. and Crosson R. S. (preprint). Seismicity of Washington and Oregon. In Neotectonics of North America, DNAG Associated Volume GSMV-1, eds. Slemmons D.B., Engdahl E.R., Blackwell D., Schwartz D. and Zoback M. Ma, L., Crosson, R. S. and Ludwin, R. S., (preprint). Focal mechanisms of western Washington and their relationship to regional tectonic stress. Milne, W. G., Rogers, G. C , Riddihough, R. P., McMechan, G. A, and Hyndman, R. D. 1978. Seismicity of western Canada. Canadian Journal of Earth Sciences, v. 15, p. 1170-1193. Munro, P. S., Halliday, R. J., Shannon, W. E. and Schieman, D. R. J., 1990. Canadian seismograph operations - 1987. Geological Survey of Canada, Paper 88-25. Owens, T. J., Crosson, R. S. and Hendrickson, M. A., 1988. Constraints on the subduction geometry beneath western Washington from broadband teleseismic waveform modeling. Bulletin of the Seismological Society of America, v. 78, p. 1319-1334. Pennington, W. D., 1984. The effect of oceanic crustal structure on phase changes and subduction. Tectonophysics, v. 102, p. 377-398. Reyners, M . , 1980. A microearthquake study of the plate boundary, North Island, New Zealand. Geophysical Journal of the Royal Astronomical Society, v. 63, p. 1-22. Reyners, M. and Coles, K. S., 1982. Fine structure of the dipping seismic zone and subduction mechanics in the Shumagin Islands, Alaska. Journal of Geophysical Research, v. 87, p. 356-366.  74  REFERENCES  Riddihough, R. P., 1977. A model for recent plate interactions off Canada's west coast. Canadian Journal of Earth Sciences, v. 14, p. 384-396. Riddihough, R. P., 1984. Recent movements of the Juan de Fuca plate system. -Journal of Geophysical Research, v. 89, p. 6980-6994. Riddihough, R. P. and Hyndman, R. D., 1990. The modem plate tectonic regime of the western Canada continental margin. In The Cordillera Orogen: Canada. Gabrielse, H. and Yorath, C. J. editors. Decade of North American Geology (DNAG), Volume G-2 (in press). Robinson, R., 1978. Seismicity within a zone of plate convergence - the Wellington region, New Zealand. Geophysical Journal of the Royal Astronomical Society, v. 55, p. 693-702. Rogers, G. C , 1983. Seismotectonics of British Columbia. Ph.D. thesis, University of British Columbia, Vancouver, 227 p. Rogers, G. C , 1988. An assessment of the megathrust earthquake potential of the Cascadia subduction zone. Canadian Journal of Earth Sciences, v. 25, p. 841-852. Rogers, G. C. and Hasegawa, H. S., 1978. A second look at the British Columbia earthquake of 23 June, 1946. Bulletin of the Seismological Society of America, v. 68, p. 653-675. Rogers, G. C , Spindler, C. and Hyndman, R. D., 1990. Seismicity along the Vancouver Island Lithoprobe corridor. In Proc. of the Project Lithoprobe: Southern Canadian Cordillera Transect Workshop, Calgary, 3-4 March 1990, p. 166-169. Ruff, L. J. and Kanamori, H., 1980. Seismicity and the subduction process. Physics of the Earth and Planetary Interiors, v. 23, p. 240-52. Savage, J. C. and Lisowski, M., 1991. Strain measurements and the potential for a great subduction earthquake off the coast of Washington. Science, v. 252, p. 101-103. Spence, G. D., Clowes, R. M., and Ellis, R. M., 1985. Seismic structure across the active subduction zone of western Canada. Journal of Geophysical Research, v. 90, p. 6754-6772. Taber, J. J. and Smith, S. W., 1985. Seismicity and focal mechanisms associated with the subduction of the Juan de Fuca plate beneath Washington. Bulletin of the Seismological Society of America, v. 75, p. 237-249. Ward, P. and Gregersen, S., 1973. Comparison of earthquake locations determined with data from a network of stations and small tripartite arrays on Kilauea Volcano, Hawaii. Bulletin  75  REFERENCES  of the Seismological Society of America, v. 63, p. 679-711. Weaver, C. S. and Smith, S. W., 1983. Regional tectonic and earthquake hazard implications of a crustal fault zone in southwestern Washington. Journal of Geophysical Research, v. 88, p. 10371-10383. Weaver, C. S. and Baker, G. E., 1988. Geometry of the Juan de Fuca plate beneath Washington and northern Oregon from seismicity. Bulletin of the Seismological Society of America, v. 78, p. 264—275. Weichert, D. H., 1980. Estimation of the earthquake recurrence parameters for unequal observation periods for different magnitudes. Bulletin of the Seismological Society-of America, v. 70, p. 1337-1346. Williamson, J. H., 1968. Least-squares fitting of a straight line. Canadian Journal of Physics, v. 46, p. 1845-1847. Wyss, M., 1973. Towards a physical understanding of the earthquake frequency distribution. Geophysical Journal of the Royal Astronomical Society, v. 31, 341-359. Yamaguchi, D. K., Woodhouse, C. A., and Reid, M . S., 1989. Tree ring evidence for syncronous rapid submergence of the western Washington coast 300 years BP., EOS, v. 70, p. 1330 (abstract). Yorath, C. J. and Hyndman, R. D., 1983. Subsidence and thermal history of Queen Charlotte Basin. Canadian Journal of Earth Sciences, v. 20, p. 135-139. York, D., 1966. Least-squaresfittingof a straight line. Canadian Journal of Physics, v. 44, p. 1079-1096.  76  APPENDIX  A  Summary of HYPOELLIPSE solutions for earthquakes in the depth range 25-55 km in the study area. EVENT 1 2 3 4 5 6 7 8 9 10 11 12 13 14" 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ' 35 36 37 . 38 39 40 41 42 43 44 45 46 47 48 49. 50  DATE 830124 830205 830205 830209 840102 840421 840427 840611 840715 840806 840819 840825 840828 840909 841004 841018 850109 850118 850209 850223 850317 850318 850319 850413 850424 850428 850430 850521 850726 850807 850825 850904 851009 851017 851101 851114 851209 851213 851225 860130 860201 860222 860223 860302 860316 860326 860404 860404 860608 860716  LAT 10 5 6 10 19 19 3 18 12 13 15 7 22 18 14 1 7 20 13 11 10 1 17 10 11 13 15 21 19 12 17 21 20 10 2 22 16 22 23 22 13 0 9 4 13 2 3 11 5 6  10 27 38 52 29 54 10 27 7 38 34 54 15 38 6 1 30 15 42 34 44 44 22 53 44 39 38 30 0 19 58 5 41 39 16 6 10 27 40 23 32 43 9 58 49 41 58 22 17 44  49. .11 49. .05 48. .76 49. .11 49. .12 49. .17 48. .74 49. .37 48. .88 49. .02 48. .85 49. .41 49. .03 49. .18 48. .88 48. .94 48. .89 49. .20 49. .77 49. .42 49. .21 49. .13 48. .75 49. .13 49. .02 49. .69 49. .18 48. .87 49. .49 48. .74 48. .96 49. .23 49. .12 48. .93 48. .93 48. .89 48. .96 48. .24 49. .18 49, .06 49, .29 48, .74 49, .17 48 .74 49, .47 48, .99 49, .15 49, .10 48. .99 49, .06  LONG -124. .99 -125. .07 -125. .32 -126. .14 -125. .28 -125. .62 -125. .13 -125. .48 -125. .90 -125. .12 -125. .09 -126. .87 -125. .52 -126. .30 -125. .10 -125. .00 -125. .15 -125. .45 -125. .90 -126. .26 -126. .30 -125. .61 -124. ,91 -125. .52 -125. .13 -126. .32 -125. .41 -125. .71 -125. .61 -125. .75 -125, .35 -126. .86 -125, .36 -124, .53 -125, .08 -125, .44 -125, .65 -125, .72 -126, .60 -125, .24 -125 .70 -124, .74 -125, .86 -124 .92 -126, .38 -125, .92 -125 .63 -125, .61 -125, .35 -126 .21  DEPTH MAG  NO DI GAP  39, .04 36. .90 31. .45 28. .13 31. .97 37. .36 31. .72 42. .50 30. .82 36. .47 32. .72 31. .67 32. .99 26. .86 37. .37 37. .08 28. .88 36. .82 44. .37 37. .20 37. .02 34. .65 32. .61 33. .13 37. .07 40. .89 39. .80 28. .60 38, .64 31. .75 35, .35 30. .22 35, .29 43, .94 37 .24 29, .29 31, .52 43 .45 34, .58 35 .51 34 .18 31 .58 28 .60 36 .85 33 .55 25 .24 35 .04 35 .19 37 .34 37 .30  12 6 11 7 5 11 10 7 5 6 9 8 11 9 12 13 13 11 9 8 11 9 14 12 16 10 10 11 12 9 10 9 16 20 12 10 9 7 7 6 7 12 16 19 9 12 8 12 14 12  77  1..90 0..90 3. .60 1..40 0..90 1..90 1..90 1.,50 0..90 0..30 1..10 1..70 2..80 1.,90 1..70 2..90 0.,90 0..90 1..00 1..10 1..40 0..90 2..00 2..40 1..90 1..50 0..80 1..60 0..90 1..30 0..70 1..50 1..30 1..60 0..90 1..20 1..60 1..60 2..90 0..80 1,.20 1..40 2,.20 3,.00 0,.90 2,.00 1,.20 1,.20 3,.20 2,.00  21 30 26 41 37 25 36 46 31 28 32 24 8 28 30 36 26 27 12 43 26 20 32 18 27 22 26 18 7 31 10 29 20 14 29 9 11 81 22 21 60 29 35 33 15 31 23 18 11 43  132 320 214 206 288 177 201 186 347 219 170 225 131 209 162 128 165 138 134 247 203 182 179 159 121 131 152 261 120 220 147 271 116 102 145 220 227 298 228 172 243 164 164 145 167 201 181 187 110 224  RMS  ERH  ERZ SQ DQ  0..20 1..2 1..3 0..02 0..5 0..9 0..15 0..8 1..3 0..17 5..6 2..0 0..45 4..1 2,.2 0..10 0..7 0..8 0..10 0..5 0..7 0..11 1.,2 2..9 0..15 17. .3 10. .0 0..07 5..6 1..1 0..15 0..9 2..4 0..17 C..9 3. .9 0..16 1..2 0..7 0..34 7..3 3..9 0..14 0.,7 0..9 0..15 0..7 0..9 0..15 ,7 0. 1..4 0..16 0.,9 1..1 0..16 1..0 1..4 0..17 2.,1 1., 6 0..22 1..5 1..9 0..08 0..7 0..6 0..14 0.,7 0..7 0..15 0..9 0..8 0..14 0..6 0..6 0 .11 , 0..9 1..0 0..16 1..2 1..7 0..19 1..8 1,.6 0..17 1,.1 1,.0 0..19 2..3 3..0 0..10 0..6 0,.9 0..22 2,.9 1,.8 0..12 0..5 0..4 0..14 0,.5 0,.8 0,.09 0,.4 0..5 0,.09 0,.7 0,.8 0,.13 1,.2 0,.7 0,.23 6,.6 17 .5 0,.12 2,.7 1,.1 0,.02 0,.3 0,.5 0,.08 1,.1 0,.8 0,.13 0,.6 1,.0 0,.24 1,.0 2 .2 0 .15 0 .6 0 .7 0 .45 2,.9 6,.7 0,.18 1,.4 2 .0 0 .13 1 .2 1 .5 0,.11 0,.8 0 .5 0,.12 0 .5 0,.6 0,.22 1,.4 3,.1  B E C c D B B C D D B D C D B B B B B C C B B B B B B C B C B D B B A B C D D A B B B B D C B B B C  B D D D D C D D D D C D B D C B C C B D D D C C B B C D B D C D B B C D D D D C D C C C c D D D B D  APPENDIX A. EARTHQUAKE LOCATIONS EVENT 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127  DATE 860724 860815 860906 860923 861002 861011 861101 861103 861125 861206 861209 861212 861217 861226 870105 870114 870201 870213 870511 870518 870531 870619 870619 870620 870621 870626 870712 870825 870906 870909 870921 870924 871015 871022 871201 871216 871217 871221 880102 .880118 880229 880315 880319 880329 880403 880405 880411 880412 880414 880418 880421 880423 880501 880502 880509 880604 880605 880607 880613 880625 880705 880711 880722 880723 880817 880904 880905 880923 881011 881012 881101 881101 881115 881125 881207 881211 881220  LAT  LONG  9 43 49..13 -125 .41 4 1 48..59 -125 .23 14 46 48..75 -125 .24 18 40 48..56 -124 .84 22 31 48..69 -124 .83 10 41 49..30 -125 .57 14 27 49..07 -126 .57 16 42 48..90 -125 .34 9 53 ' 49..48 -125 .58 7 45 49..32 -126 .17 18 12 49..11 -125 .51 15 39 49..40 -126 .23 13 29 49..44 -126 .48 14 42 48..79 -125 .14 19 56 49..02 -125 .20 16 47 48..66 -125 .16 3 9 49..42 -126 .81 13 19 49..00 -125 .38 4 55 48..90 -125 .44 18 18 48..99 -125 .51 22 4 48..82 -.124 .69 7 9 49..00 -125 .53 10 28 49..16 -125 .52 9 49 48..76 -125 .13 22 25 48..28 -125,.11 9 49 49;,10 -125 .74 1 29 49..04 -125 .14 0 31 49.,15 -125,.29 9 56 49. 46 -125,.61 5 50 48.,94 -125,.71 6 39 48.,78 -124,.72 15 50 49.,13 -126,.49 6 52 49.,36 -126,.44 19 11 48. 91 -125..35 5 11 49.,11 -125,.19 4 58 49.,08 -126,.43 0 48 49.,10 -125,.59 20 5 .48.,93 -125,.07 21 13 49.,07 -126,.26 22 38 49.,53 -126,.56 0 • 7 49.,17 -126,.64 11 4 49..18 -125 .63 3 46 49..30 -126 .19 12 56 49.,12 -125..60 19 3 48..85 -126 .01 11 18 48..94 -125 .06 48..54 -125 .00 1 31 7 58 48..85 -124 .92 0 53 49.,85 -126 .17 4 6 49.,45 -125 .95 15 41 48..91 -125 .89 6 5 49..58 -126 .17 2 48 48..85 -124 .80 17 1 49..06 -126 .65 48..98 -125 .24 9 55 7 17 49..64 -126 .23 20 53 49..37 -126 .89 9 58 48..32 -125 .28 10 38 48..89 -124 .67 13 46 48..74 -125 .32 9 14 48..95 -125 .31 19 7 48,.51 -125 .19 2 26 49..70 -125 .67 10 37 ' 49, .28 -126 .49 23 22 49..12 -125 .58 8 24 49..09 -126 .21 48,.84 -125 .05 21 21 6 2 49..03 -125 .20 10 16 49,.23 -125 .62 19 31 48..86 -125 .40 13 19 48 .89 -125 .23 21 1 48 .83 -125 .25 8 43 49..39 -125 .97 48..89 -124 .70 18 29 20 2 48..42 -124 .91 4 6 49 .01 -125 .14 17 49 49 .05 -126 .08  DEPTH MAG 40..57 30..25 31,.63 43,.51 32..34 36..47 43..11 31,.32 30..12 38..13 35..76 34..88 32..34 31..37 35..83 30..30 28..52 33..26 28..24 30..83 32.i89 33..09 34..73 27..80 32..30 33..90 39..11 37.,25 43.,67 27.,97 32.,94 34.,71 32.,24 27.,96 38.,72 34.,40 33.,53 35..91 36.,14 28.,06 33..11 34..85 34..92 31..97 37..56 33..76 29..29 33...77 29..16 35..82 27..46 34..94 37..94 31..21 34..12 35..19 35,.23 33..11 36,.83 27,.09 33,.90 25 .31 52,.17 33,.16 31 .59 32 .41 32 .02 33..27 37 .01 27 .77 33 .75 29 .65 36 .80 36 .44 30 .29 35 .24 31 .83  1..20 1 .00 3 .50 1 .30 1 .20 1,.40 0 .80 1 .20 1 .50 0..50 1..30 1,.60 2 .50 1 .20 1,.90 1,.40 1,.60 i , .60 l , .10 l . . 40 l . .00 0..70 0,.70 l , .20 l . .70 0..80 1..90 1..60 1..00 0..50 1..80 2..10 1..60 1.,30 1..30 1..50 1..40 1..80 1..60 1..30 2..30 2,.00 1,.50 1,.10 1..50 1,.50 1,.20 1 .50 1,.10 1,.20 1,.00 1 .50 1 .50 1,.90 1 .30 2 .40 2 .90 1 .20 0 .70 1 .40 1 .00 1 .00 1 .00 1 .80 1 .00 1 .30 2 .30' 1 .10 1 .20 1 .70 1 .40 2 .40 2 .50 0 .90 1 .50 1 .10 1 .90  78  NO DI GAP 13 20 10 46 18 30 6 29 6 31 9 19 7 34 10 13 9 48 9 27 11 17 11 23 12 8 7 32 10 23 12 42 10 20 10 9 13 8 S 4 13 20 7 5 10 23 13 35 14 38 9 24 14 28 16 25 10 7 5 16 17 24 9 27 5 8 9 12 10 28 5 34 11 18 13 29 9 40 7 17 11 25 13 26 11 27 8 20 7 40 11 28 6 42 13 24 4 11 11 31 7 29 11 24 9 19 8 36 10 19 10 37 8 26 7 68 8 13 11 28 ' 10 14 7 55 6 28 10 11 6 19 9 40 12 32 8 23 12 28 13 14 10 20 11 23 12 34 8 12 7 39 9 26 11 44  130 244 198 225 188 156 238 180 202 155 156 132 105 191 120 224 219 116 215 174 138 210 157 198 208 250 115 106 148 291 151 247 208 177 95 291 181 146 229 140 232 130 164 180 279 144 299 197 360 200 274 121 145 276 134 122 256 282 124 226 153 254 126 210 206 219 169 119 170 214 170 196 123 127 260 125 215  RMS  ERH  ERZ SQ DQ  0..22 1..2 1..4 0..08 0..8 1..3 0,.19 0,.8 0..7 0,.27 5,.8 9,.4 0,.02 0,.4 0,.6 0..11 1..2 1..5 0,.14 1..7 2,.4 0,.10 0..6 1,.0 0,.09 0..6 1.. 6 0..13 1..0 1..7 0..12 0..7 0..8 0..16 0,.8 0..9 0..23 1,.4 2,.2 0..11, 1,.0 2..1 0..06 0.,3 0..3 0..10 0..6 1..0 0..24 2..4 2..0 0..12 0..9 0..8 0..06 0..4 0..4 ,1 c<,12 1. 1 0..18 0.!7 i . .3 0..09 1..3 i . .3 0..15 1..1 i . .1 0..18 0..9 2..1 0.,21 2..1' 3.,5 0.,16 2...0 1.,2 0..15 0..7 0.,9 0..18 0..9 1..0 0..31 2..8 5..2 0.,03 2.,9 2..6 0..17 0.,6 0.,8 0..08 0..8 0..8 0..03 5..3 2.,6 0.,04 0.,3-. 0.,5 0..09 0..6 0..8 0.,03 2.,2 8.,0 0.,11 0..8 0.,7 0..07 0..3 0..3 0.,15 1.,5 2.,3 0.,11 1.,1 2.,7 0.,18 1..3 1..2 0..23 1..1 1..2 0..09 0..4 0..5 0..09 0..8 1..0 0..08 1..3 1..5 0..14 0..8 1..7 0,.15 7..8 10..8 0..19 1..4 0..8 0..03 99..0 99..0 0..16 1..2 1..0. 0..11 2..1 1..8 0,.16 1..0 1..0 0,.05 0,.4 0..4 0..19 3..4 3..2 0..07 0,.4 0..7 0,.17 0..7 1,.6 0,.09 1,.1 0..7 0 .14 2 .1 r..8 0,.07 0,.5 0 .8 0 .05 0..4 0..6 0 .05 0 .3 0 .5 0 .07 1 .1 2 .9 0 .09 1..9 1 .8 0 .10 0 .6 0 .6 0 .02 0 .3 0 .4 0 .22 2 .1 2 .7 0 .10 0 .5 0 .5 0 .11 1 .0 2 .0 0 .15 1 .1 0 .9 0 .08 0 .5 0 .5 0 .11 0 .7 1 .7 0 .08 0 .6 0 ,7 0 .15 1 .0 1 .6' 0 .04 0..3 0 .4 0 .04 0 .8 • 1 .0 0 .12 1 .1 2 .1 0 .18 1 .3 ' 1.1  B B B D . B D D D A D B C C D B D B D B C B C B B C B B D A B B D C D B B A D 3 C B C B D . B C B D C D C D B B B B C C D D B C B D D D A C B B C D B D A C C D C C C D B B A C B D C D B C D D C D D D B D D D B B • A C D D A B B B B D C D B B A D A C C D C B B D A D C D B C B B B C B D B C B' D B B A B B D B B C D  APPENDIX A. EARTHQUAKE LOCATIONS  Abbreviations: DATE: yymmdd hh mm; year, month, day, hour, minute. NO: Number of P, S, S-P readings used in the solution. DI: distance to the closest station used in the solution. GAP: Largest azimuthal separation in degrees between stations as seen from the epicenter. RMS: root mean squared residuals. ERH: horizontal 68% confidence limit for the least well constrained direction. ERZ: 68% confidence limit for depth. SQ: solution quality. SD: station distribution quality.  QUALITY EVALUATION:  SOLUTION QUALITY SQ  RMS (sec)  ERH (km)  ERZ (Ton)  A  <0.15  < 1.0  <2.0  B  < 0.30  < 2.5  <5.0  C  <0.50  < 5.0  D  Others  79  APPENDIX A. EARTHQUAKE LOCATIONS  STATION DISTRIBUTION QUALITY SD  NO  GAP  DI  A  >6  < 90  < depth or 5 km  B  >6  < 135  < 2*depth or 10 km  C  >6  <180  <50km  D  Others  80  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0052922/manifest

Comment

Related Items