UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

New data and re-evaluation of the 1965 Hope Slide, British Columbia Von Sacken, Rosanna S. 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 V66.pdf [ 20.08MB ]
Metadata
JSON: 831-1.0052603.json
JSON-LD: 831-1.0052603-ld.json
RDF/XML (Pretty): 831-1.0052603-rdf.xml
RDF/JSON: 831-1.0052603-rdf.json
Turtle: 831-1.0052603-turtle.txt
N-Triples: 831-1.0052603-rdf-ntriples.txt
Original Record: 831-1.0052603-source.json
Full Text
831-1.0052603-fulltext.txt
Citation
831-1.0052603.ris

Full Text

NEW DATA AND RE-EVALUATION OF THE 1965 HOPE SLIDE, BRITISH COLUMBIA by ROSANNA S. VON SACKEN B.Sc, The University of Waterloo, 1985 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Geological Sciences) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA April 1991 ® Rosanna S. von Sacken, 1991 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) 11 ABSTRACT The 1965 Hope Slide is one of the largest rock avalanche to have occurred in recent historic time. Although this landslide is very well known, virtually no comprehensive investigation was undertaken. This study represents a first, but essential, step to begin a detailed evaluation of the mass movement; it is also part of a research program investigating the landslide hazards along strategic transportation corridors in southwestern British Columbia (Savigny, 1990, in prep.). The geology at the slide site was confirmed to consist of greenstone and felsite, however, two varieties of each of the rock type were found: the greenstone occurs in a massive and a slightly schistose form, and the felsite occurs in a buff coloured and a greyish-white coloured variety. Discontinuities at the study site include two steeply dipping faults, three dominant sets (Jl, J2 and J3) and a shallower dipping set of joints, the orientations of the latter set closely relate to those of Jl, and a number of gouge filled shear zones along the buff felsite and greenstone contacts. The 1965 failure surface was probably controlled by two mechanisms, in which the steeper upper portion of the slope was largely controlled by pervasive step-like discontinuities (Jl and the shallower joints); the shallower lower part of the slope was controlled by gouge filled buff felsite-greenstone contacts. These two mechanisms also support the two slide events hypothesis put forward by Weichert et al. (1990), who suggested that the seismic signals recorded on the day of the landslide were the results, rather than the cause, of the mass movement(s). Based on the evidence found in this study, it is proposed that the lower slope (below the upper northeast trending fault) failed first along the gouge filled lithologic contacts, due to the debuttressing effects of the lower slope and the existing weakness along the joints, the upper slope subsequently failed. Slope stability analyses indicate_that-the slope was in critical conditions prior to the 1965 slide. The results also demonstrate_that the inherent weakness withinjthe rock mass was sufficient to explain the occurrence of failure_without external influences. iii TABLE OF CONTENTS Page No. ABSTRACT ii LIST OF TABLES vi LIST OF FIGURES vii ACKNOWLEDGEMENTS xiii 1 INTRODUCTION 1.1 Background 1 1.2 Previous Work 2 1.3 Purpose and Scope 2 2 REGIONAL GEOLOGICAL SETTING 2.1 Physiographical Setting 10 2.2 Climate in Southwestern British Columbia 11 2.2.1 Present Conditions 11 2.2.2 Paleoclimate 11 2.3 Tectonic Setting 12 2.4 Regional Geology 14 2.5 Regional Structures 16 2.6 Mass Movements 16 3 SITE CHARACTERIZATION OF THE HOPE SLIDE 3.1 Field Program 25 3.1.1 Mapping Method and Traverses 25 3.2 Site Geology 26 3.2.1 Lithology 27 3.2.2 Discontinuities 31 3.2.2.1 Joints 31 3.2.2.2 Pre-existing Shear Zones 36 3.2.2.3 Tension Cracks 39 3.3 Laboratory Analysis 40 3.3.1 Direct Shear Tests 40 3.3.2 Grain Size Analysis 41 3.3.3 Atterberg Li mits 42 3.3.4 X-Ray Diffraction Analysis 44 3.4 Climatic Conditions at the Slide Site for December, 1964 and January, 1965 45 3.5 Drainage Conditions 46 4 MECHANISIMS OF THE SLOPE FAILURE 4.1 Prehistoric Landslide 69 4.2 1965 Rock Avalanche 72 4.2.1 Source Area 72 4.2.2 Trenches 74 4.2.3 Volume 75 4.2.4 Geometry and Geological Control of the 1965 Failure Surface 78 4.2.5 Rockslide Kinematics 80 4.2.6 Possibility of Two Separate Slide Events 82 4.2.7 Seismic Energy Considerations from the Two Slide Events 84 iv TABLE OF CONTENTS (continued) Page No. 5 STABILITY ANALYSIS OF THE 1965 HOPE SLIDE 5.1 Overview of Slope Stability 96 5.2 Method of Analysis 97 5.2.1 Limit Equilibrium Analysis 97 5.2.2 SG-Slope, Method of Slices 98 5.3 Analytical Considerations 100 5.3.1 Assumptions 101 5.3.2 Parameters for Analysis 103 5.3.2.1 Basal Failure Plane 103 5.3.2.2 Unit Weight 103 5.3.2.3 Cohesion 104 5.3.2.4 Water Table Ratio 105 5.3.2.5 Angle of Internal Friction and Strength Considerations 106 5.4 Static Back Analysis 110 5.5 Stability Analysis and Discussion 111 5.5.1 Conclusions of Stability Analysis 120 5.6 Seismic Trigger 122 6 HAZARD ASSESSMENT 6.1 General 142 6.2 Hazard Evaluation 143 7 CONCLUSIONS AND SUGGESTIONS FOR FUTURE STUDIES 154 REFERENCES 159 APPENDICES 165 Appendix A: Topographic maps (in pocket) 166 Appendix A-l: 1:2500 scale Topographic Map of the 1965 Detachment Surface and part of Johnson Ridge (Domains 1 and 2, study area) Appendix A-2: 1:1000 scale Topographic Map the upper part of the 1965 Detachment Surface (Domain 1 of the study area) Appendix A-3: Index Map of the study area Appendix B: X-Ray Diffraction Analysis 167 Appendix B-l: Diffraction Patterns of untreated sample HSCF 170 Appendix B-2: Diffraction Patterns of heated sample HSCF 171 Appendix B-3: Diffraction Patterns of glycolated sample HSCF 172 Appendix B-4: Diffraction Patterns of heated and glycolated sample HSCF 173 Appendix B-5: Diffraction Patterns of glycolated sample HSCL 174 V TABLE OF CONTENT (continued) Page No. Appendix C: Local Weather Conditions at the Slide Site 175 Appendix C-l: Monthly Precipitation Averages at the Hope Slide Site 179 Appendix C-2: Monthly Temperature Averages at the Hope Slide Site 181 Appendix C-3: Comparison of the Monthly Precipitation and Temperatures at the Hope Airport, Allison Pass and Hope Slide sites 184 Appendix C-4: Daily Precipitation and Temperature Records for December 16, 1964 to January 9, 1965 for the Hope Airport, Allison Pass and Hope Slide sites 186 Appendix D: Summarized Results of Stability Analyses 188 Appendix D-l: Scenario 5 189 Appendix D-2: Scenario 6 191 Appendix D-3: Scenario 7 192 Appendix D-4: Scenario 8 194 vi LIST OF TABLES Table 3.1 Joint roughness measurements of SURF 11-12 and SURF07-08. Table 3.2 Atterberg limits of samples HSCF and HSCL. Table 4.1 Listing of the four known rock avalanches and radiocarbon dates in the Hope area. Table 4.2 Comparison of volume estimates of the 1965 Hope Slide. Table 4.3 Comparison of the potential energy released during the slide events and the associated seismic energy at the Hope Slide and Brenda Mine sites. Table 5.1 Possible residual strength (<£r) values for the gouge from sample HSCF. Table 5.2 Static back analysis of profile AA'. Table 5.3 Parameters used for stability analyses of scenarios 1 to 4. Table 5.4 Parameters used for stability analyses of scenarios 5 to 8. Table 5.5 Ranges of </>g values for a critical slope (F=l) and given parameters, based on analysis results of scenarios 5 to 8. Table 5.6 Peak acceleration and corresponding K<. values for different epicentral distances of the ML=3.2 earthquake at the slide site. Page No. 35 43 69 77 85 109 112 113 114 120 124 LIST OF FIGURES Fig. 1.1 Location map of the Hope Slide. Fig. 1.2 Oblique aerial photograph of the 1965 Hope Slide, looking east (B.C. Government Photograph BC(0)444, January 1965). Fig. 1.3 Pre-1965 vertical aerial photograph, showing approximate outlines of the prehistoric and 1965 landslide scars, and location of incipient landslide noted by Mathews and McTaggart (1969) (B.C. Government Photograph BC 4014-24, 1961). Fig. 1.4 Oblique photograph showing exposure of the lower northwestern flank of the 1965 slide (lighter tone), which deepened the prehistoric scarp (darker tone above). SC=scree cones, BF=buff felsite. (Photography by K.W. Savigny, 1989). Fig.2.1 Pre- and post-1965 cross section based on 1965 topographic maps, taken parallel the direction of runout (238° azimuth). Note the minimal removal of rock in the lower slope, and the steeper inclination (37°) in the upper part and shallower slope angle (25°) in the lower portion of the post-1965 profile. Fig.2.2 Topographic map of the slide site (50 m contour interval) prepared by photogrammetry. The study area (Domains 1 and 2), reference locations, temporary bench marks, two hiking trails, and the average direction of slide movement are shown. Fig.2.3 Five northwest trending morphogeological belts in the Canadian Cordillera (after Gabrielse and Yorath, 1989). The black solid circle indicates the location of the Hope Slide. Fig.2.4 Modern tectonic setting of the Pacific northwest region (after Monger, 1989b). The black solid circle labelled as HS indicates the location of the Hope Slide. Fig.2.5 Lineaments in the Hope-Fraser-Coquihalla corridors (after Savigny, in prep.). Fig.2.6 Simplified regional geology of the Hope area, showing major fault bounded lithotectonic terranes and plutonic complexes (after Monger, 1989a). The black solid circle indicates the location of the Hope Slide. vii Page No. 6 7 8 9 18 19 20 21 22 23 LIST OF FIGURES Fig.2.7 Regional faults and lineaments in the Hope-Fraser-Coquihalla corridors (after Monger, 1989a). The black solid circle shows the location of the Hope Slide. Fig.3.1 Local geology of the 1965 Hope Slide detachment surface. Outcrops of the greenstone, the two types of felsite and granodiorite, and the two faults within the study area are shown. Fig.3.2 Structural and morphological features at the slide site. Fig.3.3 Oblique photograph of the middle exposure of a thick felsite sheet in contact with greenstone, looking south (see Fig.2.2 for location, photography by K.W. Savigny, 1989). Fig.3.4 Oblique photograph looking east across Eleven Mile Creek. Note the distinct contact between the greenstone and granodiorite. Fig.3.5 Lower hemisphere projection of the three major joint sets, Jl, J2 and J3 in Domain 1 (see Fig.2.2 for location) on an equal area stereonet, based on a total number of 1910 data points and 1 % counting area. Fig.3.6 Lower hemisphere projection of the three major joint sets, Jl, J2 and J3 in Domain 2 (see Fig.2.2 for location) on an equal area stereonet, based on 5904 data points and 1% counting area. Fig.3.7 Vertical stereo-photographs showing the 1965 slide surface (B.C. Government Photograph BC 5124-023,024,025, 1965). Fig.3.8 Oblique photograph looking northeast along the upper near vertical fault (UF). Note also the younger buff felsite (BF) crossing the older greyish-white felsite (WF). SC=scree cones, solid square indicates the sampling location of the fault gouge, (see also Fig.2.2, Photography by K.W. Savigny, 1989). Fig.3.9 Oblique photograph looking north, showing the two faults identified on the slide surface. The upper fault (UF) is prominent on any photograph; the lower fault (LF) is easier to identify in an oblique view such as this one. The solid squares indicate the sampling locations of gouge (see also Fig.2.2, photography by K.W. Savigny, 1989). viii Page No. 24 48 49 50 51 52 53 54 55 56 ix LIST OF FIGURES Fig.3.10 An example of a brecciated zone, located at approximately 1600 m along the northwestern flank. Note the highly disturbed rock mass in the centre to the left of the photograph, and the more massive greenstone to the right. Page No. 57 Fig.3.11 Topographic map indicating positions of the open cracks and movement hub locations (5 m contour interval). The arrow indicates the location of a boulder laying across a crack (see Fig.5.12). 58 Fig.3.12 Oblique photographs taken in 1988 and 1990, illustrating opening of the tension cracks north of the headscarp apex (see Fig.2.2 for location). Note the different (arrow) positions of the boulder laying across one of the larger cracks in 1988 (photograph A) and 1990 (photograph B). (Photography by author, 1988 and K.W. Savigny, 1990). 59 Fig.3.13 Strength envelope for the massive greenstone based on direct shear tests. 60 Fig.3.14 Strength envelope for the schistose greenstone based on direct shear test. 61 Fig.3.15 Strength envelope for the buff felsite based on direct shear test. 62 Fig.3.16 Strength envelope for the granodiorite based on direct shear test. 63 Fig.3.17 Grain size distribution of sample HSCF. 64 Fig.3.18 Grain size distribution of sample HSCL. 65 Fig.3.19 Plasticity chart showing relative positions of the two gouge samples (HSCF and HSCL). 66 Fig.3.20 Bar graph of daily mean temperature for the period of December 16, 1964 to January 9, 1965. The 1964 and 1965 values are estimations based on comparisons of the 30 year normalized (1951-1980) records from the Hope Airport, Allison Pass and short term averages (1979-1988) from the Hope Slide stations. 67 Fig.3.21 Bar graph of daily precipitation for the period of December 16, 1964 to January 9, 1965. 1964 and 1965 values are estimations based on comparisons of the 30 year normalized (1951-1980) records from the Hope Airport, Allison Pass and short term averages (1979-1988) from the Hope Slide stations. 68 Vertical aerial photograph of the pre-1965 slope showing the pre-historic slide scar, which can be clearly delineated. Numerous pre-existing lineaments and trenches can be identified. Current exposure of the buff felsite and greenstone is located at L2 and L3; CI and C2 are pre-existing chutes; LS=series of lineaments (tension cracks?); SD=a small pre-existing displacement (after Croasdale, 1988). (B.C. Government Photograph BC4014-24, 1961). The geometry at the 1965 headscarp resembles a "pie shape" configuration, possibly controlled by joint sets Jl, J2 and J3 from Domain 1. Similar dislodgement geometry was noted along the ridge. Isopach map of the 1965 Hope Slide (after Mathews and McTaggart, 1969). The approximate limit of the prehistoric slide and the upper fault are shown. Oblique photograph showing the continuation of a trench into the rock mass at the headscarp apex (looking south, photography by K.W. Savigny, 1989). Index map showing the locations of cross sections in Fig.4.5 (XX') and Fig.4.6 (YY'). An example of "step-like" profile (XX1) sub-parallel to the dip direction of Jl (270°) on the upper slope of the Hope Slide (within Domain 1), based on 1:1000 topographic map. The steeper and shallower surfaces in this profile range from 42° to 51° and 20° to 31°, respectively. Another example of "step-like" profile (YY') sub-parallel to 270° on the upper slope of the Hope Slide (within Domain 1), based on 1:1000 topographic map. The steeper and shallower surfaces in this profile dip at relatively consistent angles of 45° and 30°, respectively. A schematic diagram illustrating the possible sequence of deposition of travelling debris, channelled by the pre-existing chute, against the pre-existing kame (not to scale). A schematic diagram (plan and vertical views) illustrating the possible two phases of sliding, in which the first phase controlled by the buff felsite sheets and the upper fault (partially after Mathews and McTaggart, 1969; not to scale). A hypothetical slope profile divided into n number of slices (after Sperling, 1991). Body forces (after Sperling, 1991) and water table ratio (R=h/z) for each slice of the profile. Location map for the cross sections AA' and BB' (Figs.5.4 and 5.5). Profile AA1 consisting entirely of greenstone, parallel to the direction of longest runout, selected for stability analysis. Profile BB' includes buff felsite sheets in contact with greenstone, parallel to AA', selected for stability analysis. Relative influence of cohesion (c) on the shear strength of greenstone (4>), a sensitivity analysis. Relative effects of the water table ratio (R) on the shear strength of greenstone (</>), a senstivity analysis. A typical stress strain deformation curve. Flow chart describing the approach and break down of the stability analysis of eight different scenarios. Stability analysis of scenario 1 (profile AA'). Stability analysis of scenario 2 (profile AA'). Stability analysis of scenario 3 (profile BB'). Stability analysis of scenario 4 (profile BB'). Stability analysis of scenario 5 (profile BB' containing gouge on the failure surface, a worst case scenario). Stability analysis of scenario 6 (profile BB' withoug gouge on the failure surface). Stability analysis of scenario 7 (the assumed first phase of sliding, profile BB'). Stability analysis of scenario 8 (the assumed second phase of sliding, profile BB'). Location map for profiles C C and DD' (Figs.6.2 and 6.3) LIST OF FIGURES Fig.6.2 Profile C C parallel to 270° selected for hazard evaluation. Fig.6.3 Profile DD' parallel to 270° selected for hazard evaluation. Fig.6.4 Oblique photograph of the headscarp apex (solid circle) area, showing the large tension crack (TC) at approximately 1725 m elevation separating the toe of profile DD1 and the intact slope above. J=joints, t=trench, the stars illustrate the location of mapped tension cracks (Fig.3.11). (Photography by K.W. Savigny, 1989) Fig.6.5 Stereoplot indicating the possible direction of sliding along Jl (280° to 315°) in the upper northwestern flank, but movement will be limited by the buttressing rock mass in lower elevations. Fig.6.6 Stereoplot indicating the possible toppling failure mechanism facilitated by J3 in the upper northwestern flank (after Goodman and Bray, 1976). xii Page No. 149 150 151 152 153 xiii ACKNOWLEDGEMENTS Many people helped to make this thesis possible, but I would like to first thank my advisor, Dr.K. Wayne Savigny, who provided not only academic guidance, but also the understanding for my "mother perogatives", continual encouragement and financial assistance. I am grateful to the members of the thesis committee: Dr.S.G. Evans, who was the one to introduce me to landslide studies, Dr.W.H. Mathews and Dr.L.J. Smith. Thanks are due to the Dept. of Mineral and Mining Processing Engineering, U.B.C., who provided the shear box for laboratory direct shear testing; Mr. Tony Sperling who provided the SG-Slope program for stability analyses; Mr. Ken Gyser of British Columbia Institute of Technology, for the.loan of the survey equipment; Mr. Don Pollock of the Prairie Farm Rehabilitation Administration for the unpublished chlorite data; Dr.J.W.H. Monger of the Geological Survey of Canada for his editing of the tectonics and geology sections of this thesis; Dr.K.C. McTaggart and Dr.H. Greenwood for their insights on the petrographic descriptions; and all my field assistants throughout this study. This work was funded by the Geological Survey of Canada, Terrain Sciences Division (E.M.R. Research Agreements 105 (1988/89 and 124 (1989/90), and Contract No. EMR-MMD-90-0043 (1990/91)); Natural Sciences and Engineering Research Council of Canada (Operating Grant A1923); the Science Council of British Columbia (Grant 57RC-18); University of British Columbia Grant No.5-56492, to which the following agencies contributed: B.C. Hydro and Power Authority, B.C. Ministry of Transportation and Highways, C.N. Rail, CP Rail, Regional District of Fraser and Cheam, Trans Mountain Pipe Line Co. Ltd, and Westcoast Energy Inc. Direct Support from Dr.S.G. Evans of the Terrain Sciences Division, Geological Survey of Canada for the preparation of two topographic maps of the Hope Slide is also gratefully acknowledged. Last but not least, I would like to thank my husband, Ulrich, for his never ending patience and understanding, especially at times of stress and frustration. Without his support, this challenge would never have been undertaken. 1 1 INTRODUCTION 1.1 BACKGROUND In the early morning of January 9, 1965, a catastrophic rock avalanche having a total volume of 47.3xl06 m3 descended the southwest slope of Johnson Peak in the Cascade Mountains (Mathews and McTaggart, 1969). The site is located about 17 km east of Hope, along British Columbia Highway 3 (also known as the Hope-Princeton Highway), approximately 160 km east of Vancouver, British Columbia (Fig. 1.1). It became known as the Hope Slide and is one of the largest historical rock avalanches in Canada (Fig. 1.2). The rock avalanche inundated a small lake (Beaver or Outram Lake), travelled up the opposite valley before spreading north and south along the valley. It stripped nearly all the vegetation in its path and buried several kilometres of the Hope-Princeton Highway, three vehicles and four people. Based on interviews with families and friends of the casualties, rescue crew members and highway personnel, Anderson (1965) compiled a chronological record of the disaster. It appears that the highway was still open to eastbound traffic just before 4:00 a.m., but was blocked shortly after when westbound traffic encountered by what appeared to be a snowslide. Whether it was indeed a snowslide, or the periphery of a precursor landslide, is unknown. Around 7:00 a.m., a landslide descended the mountain side burying the highway, vehicles and people. It is important to note that the times of the two slides were coincident with two small seismic events of magnitudes of 3.2 and 3.1, recorded at 3:56 a.m. and 6:58 a.m, respectively. It has therefore been suggested that seismicity may have triggered the mass movement (Mathews and McTaggart, 1969, 1978; Wetmiller and Evans, 1989). The 1965 event took place at a site where a prehistoric landslide of similar size had occurred approximately 9700 years B.P. (Dawson, 1879; Mathews and McTaggart, 1978). 2 The prehistoric slide scar can be delineated on aerial photographs taken prior to 1965 (Fig. 1.3). The 1965 event covered a larger area of the same slope as the prehistoric landslide, and scoured deeper into the mountain side, leaving a fresher looking scar below that of the earlier slide (Fig. 1.4). 1.2 PREVIOUS WORK The catastrophic rock avalanche in 1965 caused sensational headlines at the time. A description of the slope failure based on a study conducted shortly after the slide was reported by Mathews and McTaggart (1969), who later revised and abridged their report (Mathews and McTaggart, 1978). Bruce and Cruden (1977) briefly analysed the stability of the slope, but did not undertake further site investigation. Wetmiller and Evans (1989) recently re-evaluated the seismograph data recorded on the day of the 1965 Hope Slide. No comprehensive investigation has been undertaken to determine: the deformation patterns of the discontinuities and rock mass, the geometric and geotechnical characteristics of the failure surface, the relationship, if any, between the prehistoric landslide and the 1965 event, the influence of seismic loading with respect to other possible triggering mechanisms, and the extent of the landslide hazard. 1.3 PURPOSE AND SCOPE In response to continuing expansion of greater Vancouver and the Fraser Lowland, development pressure toward the east in the Fraser Valley and the adjacent mountain slopes is increasing. With the attendant increased exposure to landslides, there are, however, virtually no zoning guidelines that address natural slope hazards. 3 A catastrophic mass movement at a strategic location can have considerable impact on society, even though there may be no direct losses of lives and properties. Recent examples are the Squamish Highway rock slides that occurred on October 20 and 24, 1990, north of Vancouver. After the first incident (October 20), six scalers on site to clear the problematic slope of loose rock and debris were injured by a second rock slide (October 24). Direct road access between Vancouver and Squamish was completely cut off; alternate and more expensive routes, such as helicopters and free ferries, had to be put in place. The highway was closed for fourteen days until assessment on its safety was confirmed. If a large landslide, like the Hope Slide, inundated a main transportation corridor, for instance the Fraser River Canyon, major highways, transmission lines for communication, hydro, gas and oil, railway lines, and the river itself may all be blocked or damaged. The resulting direct and indirect economic and social impacts could be very significant. It is, therefore, important to further our understanding of landslide hazards affecting strategic transportation corridors. The Hope Slide was chosen for investigation because of its strategic location along one of the main highways in the province, its size, recurring activity, recent history and existing documentation. It is part of a regional study of landslides affecting major transportation corridors in southwestern British Columbia (Savigny, 1990). The objective of this thesis is to begin detailed investigation of the 1965 Hope Slide. Many aspects of the slide were examined in this study, which provide improved insights into the nature of the rock mass fabric, and the geometric and geotechnical characteristics of the failure surface, factors not elucidated as part of earlier studies. Chapter two describes the regional geological and tectonic setting of the Hope area. Two new topographic maps of the study area were produced: one at a scale of 1:2500 covering the entire detachment surface and part of Johnson Ridge (Appendix A-l , in pocket) and the other covering the upper half of the detachment surface at a scale of 4 1:1000 (Appendix A-2, in pocket). Geological, structural and geomorphological features were mapped, and extensive joint surveys were carried out to determine the relationship between the discontinuities and their control of the failure surface. Rock samples were collected for direct shear tests to determine the shear resistance of the rock mass. Gouge materials from pre-existing shear zones were analysed in the laboratory for grain size distribution, Atterberg limits and clay mineralogy. These are described in chapter three. Previous estimates of the volume of the 1965 rock avalanche indicated that the volume of slide debris in the valley was less than the volume of rock removed from the mountain. Since the bulk is expected to increase, the slide volume was re-calculated. Evaluations of the geometry, kinematics and failure plane control of the 1965 slide were undertaken based on the history of instability, discontinuity patterns and local geology. In light of recent seismic data generated by a slope failure at Brenda Mine in south-central British Columbia, the seismicity associated with the Hope Slide is now being re-evaluated by Weichert et al. (1990) as effects rather than the cause of the 1965 mass movement. Also, a rough estimate was made to compare the released potential energy and seismic energy of the Hope Slide and Brenda Mine events. These discussions are presented in chapter four. In chapter five, the assumptions and strength parameters required for slope stability analyses using SG-Slope (Sperling, 1991), which is based on Sarma's method of slices, are discussed. Eight separate scenarios were analysed. The purpose of the first four scenarios was to ascertain whether the chosen parameters were appropriate. These results were then used to narrow the range of strength parameters for the last four scenarios. The stability analyses provide insights into the strength of the slope prior to the 1965 mass movement, in particular the greenstone. Hazard assessment of the 1965 slide perimeter is discussed in chapter six. Chapter seven presents the findings of this study and provides recommendations for future work at the Hope Slide. Definitive conclusions cannot yet be drawn due to the complexity and the 5 scale of the landslide, and the limited scope of this thesis. The enhanced understanding of the Hope Slide, which this research program provides, is however an essential first step in developing the framework of more focused research programs with the ultimate objective of understanding how and why the slide occurred. Fig.1.1 Location map of the Hope Slide. Fig. 1.2 Oblique aerial photograph of the 1965 Hope Slide, looking east (B.C. Government Photograph BC(0)444, taken January 1965). 8 Fig. 1.3 Pre-1965 vertical aerial photograph, showing approximate outlines of the prehistoric and 1965 landslide scars, and location of incipient landslide noted by Mathews and McTaggart (1969) (B.C. Government Photograph BC 4014-24, 1961). Fig. 1.4 Oblique photograph showing exposure of the lower northwestern flank of the 1965 slide (lighter tone), which deepened the prehistoric scarp (darker tone above). SC=scree cones, BF=buff felsite. (photography by K.W. Savigny, 1989). 10 2 REGIONAL SETTING 2.1 PHYSIOGRAPHIC AL SETTING The Hope Slide is located about 17 km southeast of Hope along the Hope-Princeton Highway in southwestern British Columbia (Fig. 1.1). The mass movement affected the southwestern slope of Johnson Peak and inundated several kilometers of the valley below. Johnson Peak reaches an elevation of 1950 m; the valley floor at the slide site is currently at an elevation of about 700 m, up to 60 m higher than before the 1965 landslide. The 1965 slide scar extends to 1800 m in elevation at the headscarp. Based on a cross section parallel to the direction of longest runout (238° azimuth), the averaged inclination of the overall pre-1965 slope was 37°; at present the average slope inclination is 37° above 1450 m elevation, and 25° between 1450 m and 1000 m; the angle of reach (fahrboschung, Hsu, 1975) is 24° (Fig.2.1). Two marked trails provide access from the highway to the headscarp and Johnson Ridge. One trail begins at the base of the slide, crossing the colluvium and the lower section, then follows along the southern edge of the slide scar; the other trail starts from the abandoned highway, and from there it follows the heavily forested area roughly parallel to the northwest slide margin (Fig.2.2). The latter trail is overgrown with vegetation and difficult to follow at lower elevations, but it is well defined from approximately midway to the top. Throughout this report, several parts of the slide site will be referred to repeatedly, these reference locations are illustrated and labelled in Fig.2.2 and Appendix A-3 (in pocket). The areas covered in this study are Domains 1 and 2 as outlined in Fig.2.2. Although dangerous in places, much of the study area is accessible. The slide debris consists mostly of large angular blocks which provide fairly secure hand and footholds. The inaccessible parts are the precipitous cliffs, the steep scree slopes along the entire northern slide scar and the middle exposures between 1150 m and 1400 m elevation 11 (Fig.2.2). Spalling of the rocks is common from the headscarp and the northwest slide margin. The effects from heavy precipitation, surface runoff, the southwest aspect and the dark colour of the rock mass and repeated freeze-thaw cycles are the main causes for spalling of the near-surface rocks. 2.2 CLIMATE IN SOUTHWESTERN BRITISH COLUMBIA 2.2.1 PRESENT CONDITIONS The climate of the Hope area is wetter and slightly cooler than Vancouver. This is mainly due to orographic effects of the Coast and Cascade Mountains. Based on the thirty year normalized meteorological records for the Hope Airport weather station, the mean annual total precipitation in the Hope area is about 1716 mm, 193 cm of which fall as snow; Vancouver Airport receives a mean annual precipitation of 1050 mm (1951-1980 normalized values, Environment Canada, 1982). At Hope, mean daily temperatures can range from -0.4°C in the winter to 15.5°C in the summer. 2.2.2 PALEOCLEVIATE The climate of southwestern British Columbia immediately after deglaciation was warmer and drier than present. Paleotemperature and paleoprecipitation curves, applicable to southwestern British Columbia and northwestern Washington, U.S.A., were constructed by Mathewes and Heusser (1981). The paleotemperature curve shows low July temperatures near 14°C at about 12,000 years B.P., rising rapidly to maximum values of slightly greater than 16°C between approximately 10,400 and 10,000 years B.P., then declining steadily to 14.5°C between 6,000 and 4,000 years B.P. Maximum temperatures 12 are found clustering between 10,000 years B.P. and approximately 7,500 years B.P. There has apparently been little change in temperatures for the past 4,000 years. The paleoprecipitation curve shows moderately high values of 1900 mm between 12,000 and 10,400 years B.P., then dropping rapidly to minimum values of around 1450 mm between 10,000 and 7,500 years B.P., from which time precipitation rose to modern values around 6000 years ago and has remained at similar levels since then. The post-glacial period prior to the deposition of Mazama ash (approximately 6,600 years ago; Bacon, 1983), when the climate was warmer and drier, is informally termed by Mathewes and Heusser (1981) as the "early Holocene xerothermic intervaT. 2.3 TECTONIC SETTING The overall tectonic history of the Canadian Cordillera is complex. Many studies have been conducted in the past few decades to gain an understanding of tectonism and its effects on the Cordillera (Coney et al., 1980; Gabrielse and Yorath, 1989; Kleinshpehn, 1985; McTaggart, 1970; Monger, 1970, 1986, 1989a, 1989b; Monger etal., 1972, 1982; White, 1959). The Canadian Cordillera consists of five northwest trending morphogeological belts. From east to west, they are called: Foreland Belt, Omineca Belt, Intermontane Belt, Coast Belt and the Insular Belt (Fig.2.3). In the evolution of the Canadian Cordillera, rifting was the initial dominant tectonic process which began in mid-Proterozoic. The ancient western craton margin was passive until mid-Jurassic, when it became a convergent margin. From Jurassic time onward, collisions occurred in an overall setting where the North American Plate moved westward relative to the east-dipping subducting oceanic lithosphere. 13 Collisions and/or accretions between two large composite allochthonous terranes, "Terrane I" and "Terrane II" (Monger et al., 1982), and the North American Plate resulted in crustal thickening. "Terrane I", the eastern or inner composite terrane was largely four smaller terranes that were amalgamated by late Triassic before accretion to the ancestral craton in Jurassic time. The western or outer composite terrane, "Terrane U", which mainly consisted of two terranes, was established by Late Jurassic, but was probably not attached to "Terrane I", the new western margin of North America, until Cretaceous time (Monger et al., 1972, 1982). These terranes with fault related boundaries, largely originated from island arcs and basin deposits. Several episodes of deformation are recognized in the Hope area. Each new deformation overprinted or obscured earlier ones. Deformation during Jurassic and Cretaceous times was generally under a contractional and/or transpressive regime. Crustal thicknening characterized the area in the Coast-Cascade belt during this time and became non-marine by mid-Cretaceous (Monger, 1989a). Deformation by mid-Cretaceous was associated with thrusting, folding, metamorphism and plutonism, and is now characterized by regional northwest trending elements. In contrast, crustal thinning predominated in early Tertiary (ca. 50-40 Ma), during which transtension was the dominant mode of deformation (Monger, 1989a). The major structures in the area resulting from this episode of deformation are the north-south trending Fraser River Fault system, which extends into Washington State as the Straight Creek Fault; younger, northeast trending structural elements, including the Coquihalla and Vedder Faults, are Neogene(?) in age. All.of these show evidence of dextral and/or vertical displacements, indicating that they are right lateral strike-slip or normal faults (McTaggart and Thompson, 1967; Monger, 1970, 1989a). Southwestern British Columbia is part of the leading continent edge of the North American Plate and is an active subduction zone. To the west is the Juan de Fuca Plate, a younger and smaller oceanic plate bounded by transform faults (Fig.2.4). Subduction of 14 Pacific oceanic plates beneath the west coast of the continent continues in the Cascadia subduction zone and gives rise to the Cascade volcano chain that extends from northern California into southern British Columbia. Current relative plate motions in this subduction zone suggest a northeasterly or easterly maximum compressive stress direction. Numerous lineaments and structures that are related to tectonic activity show a regional trend of northwest to northeast. Savigny (in prep.) recently compiled an inventory of these lineaments in the Hope-Fraser-Coquihalla corridors from aerial photograph interpretation (Fig.2.5). A large number of landslides were also identified in these corridors, most of them were found associated with lineaments. 2.4 REGIONAL GEOLOGY Earlier reports by Daly (1912) and Cairnes (1923, 1924), and more recently by McTaggart and Thompson (1967) and Monger (1970, 1989a, 1989b) describe the geological history of the Hope area. The regional geology is summarized in the following paragraphs. The Hope area is situated near the junction of Jhe northern Cascade and Coast Mountains, also known as the Coast-Cascade Belt The geology of the Cascade Mountains in this vicinity is comprised of metavolcanic and metasedimentary rocks flanking the east and westThe core_narro_ws_andjnerges in the north and northwest with predominantly granitic and Wghergrade metamorphic rocks of the Coast Mountains. The Coast-Cascade Belt in the Hope area consists of numerous lithotectonic terranes of Permian to Cretaceous ages (Fig.2.6). From west to east, these are the Chilliwack, Harrison Lake, Bridge River and Tyaughton-Methow terranes (Fig.2.6) (Monger, 1989a). The Hozameen Complex and the Bridge River Complex make up the Bridge River terrane that is Carboniferous to Jurassic in age. The Hozameen rocks in the Hope area 15 were_emplaced between Permian and Jurassic times and probably represent deposits of a former ocean basin environment-(Monger, 1989b). East of the Bridge River terrane, rocks belonging to the Tyaughton-Methow terrane include the Triassic Spider Peak Formation, the Jurassic Dewdney Creek Formation and Ladner Group, and the Late Jurassic and Cretaceous Relay Mountain Group to the north. Conglomerates derived from the Bridge River terrane form the Cretaceous Jackass Mountain and Pasayten Group rocks. To the west of Hope, metamorphic rocks between Harrison Lake and the Fraser River have uncertain origins. They appear to have been derived from equivalent metamorphosed Bridge River, Methow and Shuksan terranes, together with granitic rocks during early Late Cretaceous (ca. 90 Ma) and early Tertiary (50 Ma) times (Monger, 1989a), contemporaneous with plutonism and regional metamorphism. In Eocene (or earlier?) time, a conglomerate and sandstone unit was laid down in the area (McTaggart and Thompson, 1967; McTaggart, 1970). Plutonic rocks in the region comprise the Late Cretaceous Spuzzum Intrusions, Yale Intrusions and isolated granitic/tonalitic intrusive units; the Oligocene/Miocene (ca. 35-18 Ma and earlier) Chilliwack and Mount Barr Batholiths and the Hell's Gate unit (Fig.2.6) (Monger, 1989a). The^ l965^ Hope Slide involved the second division (Greenstone) of the Hozameen ComplexwithinJhe_Bridge_Riyjer_terrane.. and intrusive units of felsite.of.an .unknownjLge^  (Mathews and McTaggart, 1969, 1978; Monger, 1989a). These will be described in more detail in section 3.2. 16 2.5 REGIONAL STRUCTURES Structural elements in Jurassic and early Cretaceous are predominantly northwest trending, indicating a regional northeast-southwest compressive stress direction. Associated with dextral transpression, deformations characterized by northwest trending, west-verging folds and thrust faults were established by mid-Cretaceous, probably contemporaneous with development of the Hozameen Fault in the area (Monger, 1989a). The main Tertiary structure is the conspicuous north-south trending Fraser River fault zone (Hope and Yale Faults) which developed between late Eocene (ca. 46 Ma) and Oligocene (ca. 35 Ma) times. Two main northeast trending faults in the Hope area are the Coquihalla and Vedder Faults, which were probably formed in the Neogene(?) Period (Monger, 1989a). Major faults and lineaments in the Hope area are illustrated in Figs.2.6 (Monger, 1989a) and 2.7 (Savigny, in prep.). Both the Cascade and Coast mountain systems were subsequently uplifted during Pliocene-Pleistocene time (Parrish, 1982). 2.6 MASS MOVEMENTS Numerous mass movements have been documented along the Fraser Canyon. Between Hope and Lytton, over seventy post-glacial or interglacial landslides have been identified (Piteau, 1977). Ryder et al. (1990) recently reported two Holocene age rock avalanches at the confluence of Texas Creek and the Fraser River, about 300 km from Vancouver. The younger of the two avalanches had an estimated volume of 7.2xl06 m3 and occurred about 1200 C 1 4 years ago (Ryder et al., 1990). Savigny (in prep.) completed an extensive regional inventory of landslides covering the Fraser, Coquihalla and Hope-Princeton valleys and a corridor extending 20 to 30 km on each side of these transportation arteries. The inventory, based on aerial photograph interpretation, includes linear features. 17 More than thirty large landslides were identified with estimated volumes in the range of lxlO6 to 500xl06 m3 (Fig.2.5). Among these are four major rock avalanches: Hope Slide, Cheam Slide, Lake of the Woods Slide and Katz Slide. 18 horizontal distance (m) 0 305 610 915 1220 1525 1830 2156 2467 2776 6500" 5500 ^ 4500" c = 3500 CO > o o 2500" 1500 500 NE sw N. x x pre-1965 surface 1965 surface ^ ' ' S ^ / / 2005 h 1696 - 1388~ c 1079 = OJ > <D -771 "463 154 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 horizontal distance (ft) Fig.2.1 Pre- and post-1965 cross section based on 1965 topographic maps, taken parallel the direction of runout (238° azimuth). Note the minimal removal of rock in the lower slope, and the steeper inclination (37°) in the upper part and shallower slope angle (25°) in the lower portion of the post-1965 profile. F i g 2 2 Topographic map of the s l i d e s i t e (50 m contour i n t e r v a l ) prepared by photogrammery. The study area (Domains 1 and 2), reference loca t i o n s , temporary bench marks, two h i k i n g t r a i l s , and the average d i r e c t i o n of s l i d e movement are shown. 20 Fig.2.3 Five northwest trending morphogeological belts in the Canadian Cordillera (after Gabrielse and Yorath, 1989). The black solid circle indicates the location of the Hope Slide. 21 Fig.2.4 Modern tectonic setting of the Pacific northwest region (after Monger, 1989b). The black solid circle labelled as HS indicates the location of the Hope Slide. Fig.2.5 prep.)-Lineaments in the Hope-Fraser-Coquihalla corridors (after Savigny, in 23 Fig.2.6 Simplified regional geology of the Hope area, showing major fault bounded lithotectonic terranes and plutonic complexes (after Monger, 1989a). The black solid circle indicates the location of the Hope Slide. Fig.2.7 Regional faults and lineaments in the Hope-Fraser-Coquihalla corridors (after Monger, 1989a). The black solid circle shows the location of the Hope Slide. 25 3 SITE CHARACTERIZATION AT THE HOPE SLIDE 3.1 FIELD PROGRAM Field work was initiated in 1988, when the structures and geology at the headscarp and ridge areas (Domain 1 in Fig.2.2) were mapped along survey controlled traverse lines. Temporary benchmarks (Fig.2.2) were established at appropriate sites along some of these traverses and a survey was later carried out using an AGA model 14A electronic distance measuring device (EDM), a Wild T2 theodolite and multiple mirror targets. A 1:1000 scale, 5 m contour interval topographic map of this domain of the slide was prepared by McElhanney Geosurveys Ltd. using survey controlled ground data (Appendix A-l in pocket). Samples of rock and shear materials were also collected for laboratory testing. Phase two of the field work was undertaken under the direction of Dr.K.W. Savigny in 1989, when the structures in the lower part of the slide below the upper fault (Domain 2 in Fig.2.2) were mapped. Also, movement hubs across several open cracks behind the northwest headwall (Fig.2.2) were installed to monitor the rate and direction of movement (Fig.2.2 and section 3.2.3). A 1:2500 scale, 5 m contour interval topographic map of the whole slide surface, based on survey controlled data points, was prepared (Appendix A-2 in pocket). Field mapping was finalized in 1990, along with additional sampling of gouge materials. The areas covered in this study consists of Domains 1 and 2 in Fig.2.2. 3.1.1 MAPPING METHOD AND TRAVERSES A continuous and unbiased line mapping method, known as DISCODAT, was adapted for recording the discontinuities in the rock mass. DISCODAT (Herget, 1977) was 26 first developed for mining practices by Canada Centre for Mineral and Energy Technology (CANMET). DISCODAT provides a method to simplify and organize the collection, retrieval and manipulation of discontinuity data for a rock mass. The methodology used in this study was slightly modified from DISCODAT to eliminate the inflexibility of the mnemonic codes in the original version of DISCODAT (Cruden et al., 1977). Continuous traverses were made wherever possible. All discontinuities encountered along each traverse line were measured and recorded. Survey stations were chosen for their accessibility, relative positions within the study area and visibility from the slide base. Temporary survey ground control was set up at the base of the slide. Stereonet analyses of the discontinuity data were performed using a software package developed by Golder Associates1. The results will be discussed in section 3.2.2.1. 3.2 S I T E G E O L O G Y The 1965 Hope Slide occurred in the Permian to Jurassic age Greenstone Member (second division) of the Hozameen Complex (Monger, 1989a), and an intrusive unit described as felsite by Mathews and McTaggart (1969, 1978). Fig.3.1 shows the geology at the slide site (see also Appendix A-3 in pocket). The Hozameen Complex consists of four lithologic divisions (Cairnes, 1923, 1924; McTaggart and Thompson, 1967). The first (lowest) division is composed of mainly ribbon chert, greenstone and local bodies of limestone. wMch.are.joverlam-by-a-division„composed almost entirely of greenstone, with local lenses of chert and limestone. This division is more than 1300_m_thick in me area of die Hope Slide, but thins to[jboutjjOOmJowards the Cana(k-United,,Sjates border approximately 30 km south of the slide. The third division 1 Software package, called Stereo Analysis Package, is developed by Golder Associates, 224 W.8th Ave. Vancouver, B.C. 27 consists of a belt of ribbon chert and argillite that thins towards the north. Monger (1970) described the argillite as a pelite, referring to fine grained clastic rocks that may be shale, argillite, slate or phyllite. The uppermost division of the Hozameen Complex consists of greenstone, chert, limestone and argillite. The Hozameen rocks were affected by general, low grade regional metamorphism in the subgreenschist to low greenschist facies, with locally higher grade metamorphism near the contacts with gneissic and granitic rocks (McTaggart and Thompson, 1967; Monger, personal communication, 1990). There are two granitic intrusions in the general area of the slide site. One of these is the Chilliwack Batholith; the other consists of smaller plutons (Mount Outram Pluton, Monger, 1989a) that lie across or near the Hozameen Fault. The granitic plutons found at or near the Hope Slide have been described as tonalites2 by Caimes (1923) and McTaggart and Thompson (1967). 3.2.1 LITHOLOGY (a) GREENSTONE The greenstone (the second division of the Hozameen Complex) at the slide site is a weakly metamorphosed volcanic rock, varying in composition between basaltic and andesitic and is locally associated with serpentinite (McTaggart and Thompson, 1967). No limestone lenses were observed within the study area (Fig.2.2) and only rare pieces of chert and what appeared to be serpentinite were found among the slide debris. Outcrops of greenstone are widespread at the headscarp, along the mountain ridge to the south, and along the northwest flank (Fig.3.1). 2 Based on the volume % of plagioclase and quartz, tonalite should be considered as a granodiorite (Streckeisen, 1976; Monger, 1989a). 28 The greenstone is generally dense, massive, closely fractured, aphanitic and in some place finely veined by quartz and/or calcite. It is dark green to greenish grey in colour on freshly broken surfaces. On weathered surfaces, however, greenstone is rusty in colour. The greenstone is very fine grained even when examined under a petrographic microscope. It can be described as intergranular, sub-ophitic and porphyrinic. In descending order of abundance, the principal mineral constituents are actinolite (approx. 60%), quartz and feldspar (approx. 15%), chlorite (15-20%) and an opaque mineral, probably magnetite (^ 5%). Actinolite occurs mostly as blades and needier, that are interwoven in very fine, dense aggregates of quartz and feldspar. It is difficult to determine the composition of the feldspar because of its small grain size. Very slender veinlets of quartz and feldspar (probably plagioclase) within the greenstone are commonly iron stained along the edges; chlorite and actinolite laths grow perpendicular to the vein walls. In relatively larger veins, chlorite and actinolite tend to lie oblique or subparallel to the vein walls, resulting in slight schistosity of the greenstone. Minor calcite occasionally occurs as vein filling. The greenstone locally varies subtly from the above description to a sub-trachytic, finer grained rock with slight schistosity and/or foliation. Hand specimens of this greenstone have a slightly greasy or soapy feel on fresh surfaces. In the field, no distinct boundaries or consistent patterns can be found between this type of greenstone and the massive variety. Occurrences of the schistose greenstone are found mostiy below the headscarp to the south at approximate elevations of 1725 m and 1675 m (Fig.3.2). At these locations, near vertical exposures of this foliated-schistose greenstone are wavy, with wavelengths of approximately 5 to 10 metres, and amplitudes of approximately 0.3 to 1 m (marked as wavy greenstone in Fig.3.2). At the middle exposures on the current slope (Fig.2.2), the wavy surface of the massive greenstone dips sub-parallel to the slope with apparent shallow amplitudes. 29 A thin section prepared from the schistose greenstone indicates that composition is very similar to that of the massive greenstone. The minerals are finer grained with very indistinct boundaries, especially between the silicate minerals. A slight schistosity can be seen due to the fibrous chlorite and needle or lath like actinolite. (b) FELSITE Two varieties of felsite are found at the Hope Slide: a buff coloured variety and a pjnk to_^yish-white variety.-Both occur mostly as sills or "sheets" which can be up to 6 m thick (Fig.3.3), the latter term was coined by Mathews and McTaggart (1969, 1978). Like the greenstone, they are aphanitic in hand specimens. Both types of felsite tend to occur subparallel to the_slope. The subparallelism between the buff felsite and the greenstone is particularly pervasive along the lower northwestern flank and in the middle exposures of the detachment surface (see Fig.2.2 for locations). The buff felsite is also younger than the white felsit^ tinsijs^eyjdentjby a near vertical buff felsite unit crossing a white felsite sheet at elevations between 1500 and 1600 m (Fig.3.1). At the middle exposures (Figs.2.2 and 3.1), the buff felsite sheets presently occupy the troughs of the apparently wavy, massive greenstone mentioned in the above paragraphs. The felsite sheets here may have been continuous at one time, but the parts occurring at the wave crests were probably removed during the 1965 event. Spot measurements of the buff felsite-greenstone contacts indicate consistent dip angles between 30° and 35°. Measurement of the lithologic contacts at the middle exposures and along the precipitous cliffs was impossible due to unsafe access. Contacts between white felsite and greenstone vary over a wide range of dip angles between 30° and 80° (von Sacken et al., 1989). Whi^ ejiehitejshee^ the higher elevations of the study area. Comparisons of the two types of felsite contacts with greenstone show that shearing was-more-a)mmon--at"the--buff--felsite-greenstone 30 contacts, consistent with the higher gouge content, whereas white felsite-greenstone contacts are relatively sharp and clean. It was difficult to obtain rock samples of reasonable sizes containing contacts of greenstone and either of the felsites. The rock tended to break along the contact if the contact was relatively clean of gouge. Where gouge material was sufficiently thick, samples of the gouge were collected for analysis. Thin section examinations indicate that the two types of felsite contain very similar principal minerals: quartz and feldspar (plagioclase), actinolite, chlorite and biotite and magnetite. The quartz and feldspar constitute approximately 70 to 75% of the rock; they make up an equigranular, fine to very fine grained groundmass, with occasional mosaic intergrain boundaries. Actinolite (approx. 20%) largely occurs as blades, mixed with chlorite and biotite (10-15%). A minute quantity of magnetite is scattered throughout in both felsite samples. Relic phenocrysts of plagioclase can be identified based on distinct Carlsbad twins. The main differences between the two kinds of felsite are that the buff variety contains minor subrounded garnet crystals and very slightly schistose muscovite, which suggest a slightly higher grade of metamorphism. (c) GRANODIORITE Although not involved in the 1965 slide, outcrops of granodiorite are found along Johnson Ridge just south of the slide (Fig.3.1). Its contact with greenstone can be traced along Johnson Ridge, but it can be clearly seen on the opposite (to the east) mountain side across Eleven Mile Creek (Fig.3.4). It is a white, coarse to medium grained granodiorite (Monger, 1989a; Streckeisen, 1976). Physical weathering of the rock tends to produce a sugary texture (grus) and/or sub-rounded pits on the surface. Based on petrographic examination, it contains mostly plagioclase (approx. 65-70%) and quartz (approx. 20-30%), biotite (10-15%) and hornblende (3-5%). Minor accessory minerals include 31 scattered euhedral to subhedral crystals of zircon (<3%) and apatite (minute in quantity). Plagioclase is characterized by transecting and parallel twins, some are zoned concentrically, while others have very minor granophyric texture. Quartz is generally irregular in sizes and shapes, some with serrated edges. Mafic constituents consist of flakes of biotite and green to light brown hornblende. 3.2.2 DISCONTOtTJITIES 3.2.2.1 JOINTS In the headscarp and ridge areas (Domain 1 in Fig.2.2), the orientation of rock mass discontinuities were measured at over 1900 points during the detailed field mapping using the modified DISCODAT method. Stereonet analyses of these structural data were performed using a computer package1, in which the contouring method is based on 1% counting area. Stereonet analyses of the data show that there are three major sets of joints within Domain 1 (Fig.2.2). These joint sets are labelled as Jl, J2 and J3 with average dip angles and dip directions of 457273°, 857305° and 807018° 3 , respectively. Fig.3.5 shows a lower hemisphere projection of the pole concentrations of these three dominant joint sets from Domain 1 on an equal area stereonet. It should be pointed out that within Domain 1, approximately 10% of the discontinuities have similar orientations as Jl, but with shallower dip angles as low as 20° (Fig.3.5). Alternating surfaces of Jl and shallower joints give a step-like geometry, which is best illustrated in cross sections of the slope taken parallel to the dip direction of Jl (see 1 Software package, called Stereo Analysis Package, developed by Golder Associates, 224 W. 8th Ave. Vancouver, B.C. 3 All joint orientations are given in dip angle/dip direction (azimuth), unless otherwise stated. 32 Figs.4.6 and 4.7, and section 4.2.4). Unfortunately they are masked by the dominant Jl set when presented in a stereonet (Fig.3.5). The same three joint sets are found in the lower portion of the slide (Domain 2 in Fig.2.2). However, their orientations are shifted relative to those in Domain 1 (Fig.3.6). Jl dips at a slightly shallower angle of 40° in the direction of 253° azimuth, instead of 45 7273°. J2 also dips less steeply at 80° and in a more northwesterly direction than J2 in Domain 1. J3 maintains its average dip angle of 80°, but appears to have switched its dip direction from northeast to northwest. During field work, the dip direction of the J3 joints in Domain 1 was also observed to vary between northeast and southwest, with a consistent mean dip angle of 80°. Hence, this change of dip direction in Domain 2 was not surprising. Possibly another variation of J3 is indicated by a separate concentration of poles (labelled as J3? in Fig.3.6) with a mean orientation of 607036°. The rocks along the entire northwestern flank of the slide (see outcrop number 5, 6 and 8 in Appendix A-3, in pocket) are much more fractured and brecciated than the outcrops found above the headscarp (see outcrop number 1 and 2 in Appendix A-3, in pocket); hence discontinuity measurements along the northwestern flank are likely distorted and therefore not as representative. Also, for reasons of safety and inaccessibility, discontinuity measurements of the middle exposures (outcrop number 7 in Appendix A-3, in pocket) and the steep cliffs of the northwestern flank were limited. The difference in orientation of the joint sets illustrated in Figs.3.4 and 3.5 cannot be explained simply by an episode of surface displacement. This can be demonstrated by considering that the Jl joints from Domain 1 (Fig.3.5) were undisturbed, but those in Domain 2 (Fig.3.6) experienced displacement. By superimposing the two stereonets (Figs.3.5 and 3.6), one can see that the two Jl pole concentrations in Figs.3.5 and 3.6 do not match. Counterclockwise rotation of Fig.3.5 (to the north) and keeping Fig.3.6 stationary, so that the Jl pole concentrations coincide as much as possible, it can then be seen that the pole concentrations of J2 and J3 are not only mis-matched, but they are 33 shifted even further apart. If Jl, J2 and J3 were all produced at the same time and experienced the same displacement, all three joint sets should match after the rotation, but the difference in the Jl joints between Domain 1 and Domain 2 is not the same as those in J2 or J3 between the two domains, neither in magnitude nor direction. A clockwise rotation of Fig.3.5 (to the south), keeping Fig.3.6 stationary, does not cause any of the three pole concentrations to coincide. The nature of difference in joint orientations is speculated to be related to tectonic activity rather than only gravitational displacement. Discontinuity data from the upper slide area (Domain 1) indicate average spacings of 94 cm, 58 cm and 72 cm for joint sets Jl, J2 and J3, respectively. These are estimates based on selected traverses that contain a relatively large number of joints and are oblique to the strikes of the joints. The best in-situ representation of intersecting joint surfaces are those along the mountain ridge a short distance south of the slide. Jl in particular appears to have very smooth surfaces, on which asperities (second order irregularities, after Patton, 1966) are found to be very minor. Following the method of Bruce and Cruden (1980), a joint roughness survey was undertaken on two exposed Jl joint surfaces in the ridge area (which will be referred to as SURF11-12 and SURF07-08, see Fig.2.2 for location), on which the measurable surfaces span approximately 7x7 m2. The method is a modification of that described by Kerrich (1974). It was chosen because it requires only a Brunton compass and an exposed joint surface. This technique for evaluating surface roughness is based on the orientation of the fabric element of an outcrop. Following Kerrich (1974) and Bruce and Cruden (1980): d = e + z + w + e [3.1] where d = a dip measurement of a discontinuity, which is part of a set of sub-parallel discontinuities that form the rock mass fabric. 9 = the true dip of the fabric element. 34 z = the error due to differences in orientation among the different discontinuities at the outcrop. w = the error due to differences in orientation at various locations on a single discontinuity surface. e = the observational and operational error due to repeated measurements of dip at the same location on the same discontinuity. Using fundamental statistics, for a large population of dip measurements, say n, of the same discontinuity, the true dip (9) of the fabric element can be estimated by a mean value, E(d) (these are the same notations as in Bruce and Cruden (1980)). The variance of this mean is (^ (d), which can be estimated by S2(d). Similarly, z, w and e for the population of dip measurements can be characterized by their means and variances. Assuming that z, w and e are random and independent variables, then E(d) * 9 [3.2] S2(d) - o^ d) = (^ (z) + (^ (w) + (^e) [3.3] The procedure for measuring the surface roughness of discontinuities are outlined by Bruce and Cruden (1980). The following steps were taken in the field: (1) A cross with 0.3 m arms parallel to the strike and dip of the discontinuity was marked at approximately the centre of the desired discontinuity surface. (2) A total of fifteen measurements were taken: three measurements of the dip at the centre and at the end of each arm of the cross were recorded to estimate the standard deviation of the true dip, S(d). (3) To complete the survey, steps (1) and (2) at two larger scales of 3 m and 30 m should be repeated. For this study, the 30 m scale was not used due to the limited exposures of the available Jl surfaces, only scales of 0.3 m and 3 m were used. Although the 30 m scale was not included, which would give an indication of the 35 waviness of the surface, small scale roughness of the discontinuities (second order irregularities, Patton (1966)) can still be characterized. Readers are referred to equations (1) to (4) in Bruce and Cruden (1980) for the estimations of S2(d), S2(e), S2(w) and S2(z). S2(z) (equation (4) in Bruce and Cruden (1980)), which estimates the error due to different orientation of different discontinuity, was not necessary for the selected surfaces, because they were of a single fabric element. Based on the results of a large sample population and at 30 locations, Bruce and Cruden (1980) stated that the topographic component of shear resistance, i, can be approximated by the standard deviation of w, S(w). For SURF11-12 and SURF07-08, the topographic component of shear resistance, i, were estimated to be 2.4° and 1.3° on the 0.3 m scale which averaged 1.9°, and 1.8° and 4.4° on the 3 m scale which averaged 3.1°, respectively. These i values are considered to be very minor, and the two Jl surfaces are considered planar. The mean dip angles from these measurements correlate well with those obtained from the stereonet (45° for Jl, see Fig.3.5). These results are tabulated in Table 3.1. Surface Scale E(d) S(w) «*> i S(d) S(e) SURF11-12 0.3 m 47° 2.4° 0.6° 1.2° 3.0 m 46° 1.8° 0.5° i . r SURF07-08 0.3 m 40° 1.3° 1.2° 1.4° 3.0 m 40° 4.4° 1.8° 3.1° where (E)d *» mean dip angle S(w) * i, discontinuity roughness S(d) = deviation of dip measurement of the discontinuity S(e) = deviation due to operational and observational error Table 3.1 Joint roughness estimates (angle i), based on the method of Bruce and Cruden (1980) 36 Larger scale irregularities (waviness) were observed in the greenstone, where they are marked as wavy greenstone in Fig.3.2 and Appendix A-3 (in pocket), and at the middle exposures where the felsite sheets apparently fill the troughs of the wavy greenstone (this was not labelled in Fig.3.2, section 3.2.1). Three of the five wavy greenstone outcrops, located around the headscarp (outcrop number 4, 5 and 6 in Appendix A-3 and Fig.3.2) are near vertical, and four of the five outcrops involve the schistose greenstone (outcrop number 4 and 5 in Appendix A-3). The characteristics of these large scale waves could only be visually estimated due to their upright occurrences and/or inaccessibility. 3.2.2.2 PRE-EXISTING SHEAR ZONES Two categories of pre-existing shear zones are referred to in this report: (1) faults and (2) shear zones. The latter is subdivided into (2a) brecciated zones that often contain fine grained and/or gouge materials and (2b) tectonic shear zones that are gouge filled lithologic (felsite-greenstone) contacts. (1) FAULTS Two faults were identified at the Hope Slide, both appear as lineaments across the slide on aerial or oblique photographs. A set of vertical stereo photographs in Fig.3.7 show the rupture surface including the faults. The upper fault is especially conspicuous, it trends between 030° and 035° across the slide surface with a near vertical dip. Fig.3.8 is a view of the slide looking along the strike of the upper fault, which outcrops at the base of the precipitous cliff along the northwestern flank at approximately 1550 m elevation (solid square in Figs.3.8 and 2.2), where highly slickensided fault gouge up to 25 cm in thickness was found. A sample (sample #HSCF) of the gouge material was collected at this location 37 (Fig.2.2) for laboratory testing. The lower fault is less obvious, but is easily identifiable on oblique photographs. It can be seen in Fig.3.9 that the lower fault continues toward a shear zone (located above SC in Fig.3.9) at the northwestern flank. Both faults correlate with the trends of regional lineaments found in the area (Fig.2.5). (2) SHEAR ZONES Aside from the two faults, a number of tectonic shear zones of various widths can be found around the perimeter of the slide (Fig.3.2). These pre-existing shear zones are present in two forms: (2a) shear zones characterized by large zones of disturbed rocks and (2b) shear zones located at felsite and greenstone contacts. (2a) BRECCIATED ZONES The best clue to the existence of this type of shear zones exist is the well developed scree cones which coalesce with the slide debris or adjacent scree cones on the slide surface (Figs.3.2 and 3.10). The width of the brecciated zones are 10 to 30 m, where the rocks are highly fractured, jointed and disturbed. Fig.3.10 shows part of a brecciated zone, note that the rock on the right side of the photograph becomes progressively more massive and competent. Brecciated zones often contain seams of finer grained materials. These seams, variable in thickness (from a few millimetres to 0.5 m), are generally composed of silty sand to clay sized particles; they are indicative of shearing and represent immature shear zones. The accumulation of these fine grained particles at the base of the cliffs make up the scree cones. Sorted by gravity during the descent of the particles, the scree cones are well graded and fine upwards. These fine to very fine grained materials have a high dry strength, making access to the disturbed rock zones along the steep cliffs very difficult. 38 The surrounding rock of these brecciated zones is generally very fragmented, loose and highly susceptible to dislodgement (Fig.3.10). Brecciated zones are found nearly everywhere along the northern half of the slide margin and around the headscarp apex (Fig.2.2). At least six such zones can be identified. (2b) LITHOLOGIC SHEAR ZONES Buff felsite and greenstone contacts are most commonly found below the upper fault (Appendix A-3 in pocket and Fig.3.1) along the lower northwestern flank and in the middle exposures (Fig.2.2). Shearing occurs locally at both the top and bottom contacts with greenstone. These shear zones consist of a central seam, containing gouge in a matrix of fines and small greenstone chips, varying from a few millimetres to 1.5 metres in thickness. The gouge in the central seams of these shear zones is finer than that found in the brecciated shear zones (type 2a above). A sample of the lithologic shear material (sample HSCL) was collected for laboratory analysis (see Fig.2.2 for location). This type of shear zone was also observed by Mathews and McTaggart (1978) on the valley wall opposite (southwest) to the slide site, but they did not specify whether the observed shear zones were located at buff or white felsite contacts with greenstone. In contrast, at higher elevations, white felsite sheets are welded to greenstone in a comparatively sharp, clean manner with little to no sign of shearing. Fig.2.2 shows the sampling locations for the two shear samples (HSCF and HSCL), which were analysed in detail to determine the grain size distribution, Atterberg limits and clay mineralogy. 39 3.2.2.3 TENSION CRACKS The existence of open tension cracks just north of the headscarp apex (Fig.2.2) has been known since 1975 (Horcoff, 1975). However, no measurement or monitoring of displacements has ever been undertaken. In the 1988 field season, the location of the open cracks at approximately 1800 m elevation were mapped (Fig.3.11), spot measurements were made and oblique photographs were taken. One of the larger open cracks measures up to 30 cm wide and 16 m deep (von Sacken et al., 1989). In 1989, notable displacement was observed across several of the cracks. Movement hubs were then installed to monitor the rate and direction of movement. Comparison of oblique photographs taken in 1988 and 1990 shows the evidence of movement (Fig.3.12). No significant movement was recorded across any of the monitored cracks between 1989 and 1990. It is crucial that the monitoring program be continued, and annual or semi-annual measurements across these cracks should be made. Some of the tension cracks at the headscarp approximate the orientations of joint sets J2 and J3 (85°/305° and 807018°, respectively) in Domain 1 and that of the local slide margin (Figs.2.2, 3.2 and 3.11). From approximately 1805 m to 1775 m elevation, these tension cracks occur continuously, but below approximately 1775 m elevation, ground vegetation begins to cover and mask their appearance and continuity. Along the upper northwestern flank where open tension cracks are most prominent (Fig.2.2), the mode of failure appears to be toppling (see section 6.2). Large blocks of rock in this area are bounded by tension cracks and rockfalls are common. Numerous rockfalls were noted during field work, most originated from the northwestern flank and headscarp areas of the slide. Some created so much dust that the view to the lookout in the valley from the mountain top was obscured for several minutes. 40 3.3 LABORATORY ANALYSES 3.3.1 DIRECT SHEAR TESTS Direct shear tests were conducted on saw-cut surfaces of greenstone, buff felsite and granodiorite to determine their basic friction angles (<pb). Procedures and apparatus used for testing follow the guidelines by Brown (1981), Hoek and Bray (1977) and Bruce (1978). Rock samples were first cut with a diamond saw into appropriate sizes to fit the direct shear box, which allows a maximum sample size of approximately 10x10 cm2. Saw-cut surfaces of the two halves of each sample were then sanded with 80 grit sandpaper to remove any large protrusions. Samples were then cast inside the shear box with a cement and sand mixture to prevent movement of the samples during shearing. The cut surfaces of a rock sample, which represent the failure plane, were aligned so that they were parallel to and level with the direction of shear. Care was taken to keep these surfaces clean prior to testing. From the isopach map of Mathews and McTaggart (1969), the maximum depth of rock removed during the 1965 rock avalanche was 147 m. Based on this overburden thickness and an average rock density of 2.8 g/cm3, maximum normal stress due to overburden would be approximately 4100 kPa (1500 psi). Therefore, the range of normal stresses for the direct shear tests was extended to approximately 4700 kPa (1750 psi) to ensure thorough coverage of overburden loading. Each sample was sheared twice (forward and reverse shear) under each of the six different normal stresses at approximately: 300 kPa (100 psi), 1400 kPa (500 psi), 2800 kPa (1000 psi), 3400 kPa (1250 psi), 4100 kPa (1500 psi) and 4700 kPa (1750 psi). Five greenstone samples, one buff felsite and one granodiorite samples were tested. Resulting basic friction angles for three of the four massive greenstone samples were 34° (Fig.3.13). The schistose greenstone sample was tested to have a basic friction angle of 41 31° (Fig.3.14). Buff felsite and granodiorite showed basic angles of friction of 35° and 31°, respectively (Figs.3.15 and 3.16). The results for the greenstone samples are consistent with the literature values of similar rock types (basalt, amphibolite and metavolcanic rocks) which are reported to vary between 30° and 40° (Deere and Miller, 1966; Hoek and Bray, 1977). Bruce and Cruden (1977) also performed direct shear tests on felsite samples, though the felsite type was not specified. They reported a basic friction angle of 30°, based on fifteen samples, each of which was sheared only once. Assuming that they tested the same type of felsite (buff variety), this fa value is 5° less than that obtained in this study, which represents a 17% difference in strength. If failure took place through intact felsite, as assumed by Bruce and Cruden (1977), the factor of safety they calculated would been under-estimated. Cohesion values on the order of 100 kPa were obtained from the direct shear test results (Figs.3.13 to 3.16), however, these are probably artifacts due to data fitting since no curvature is noted in the strength envelopes and no cohesion is expected for the smooth sawn rock surfaces. 3.3.2 GRAIN SIZE ANALYSES Combined sieve and hydrometer tests were performed for samples HSCL and HSCF. Samples were first oven dried, then broken down by hand or by using a mortar and pestle. Sieve analyses were performed and the portions passing through the #200 mesh sieve were later used for hydrometer tests (Lambe, 1951). The natural shear sample from the lithologic contact (HSCL) was gritty and contained chips of greenstone. Its natural water content by weight was 5%. The fault sample (HSCF) was more clayey and more plastic, and the natural moisture content was 31%. 42 Results of the grain size analyses for the two gouge samples are illustrated in Figs. 3.17 and 3.18. It can be seen that in sample HSCL, obtained from the buff felsite-greenstone contact, only 5% are clay-size particles, whereas the fault gouge sample (HSCF) contains a clay fraction of 17%. The 5% clay fraction of sample HSCL should be viewed as a lower limit, because during sieve processing of the natural sample, some very fine powder sized particles were lost. Some of the fines were retained as little "balls" in the coarser sieves, because the moisture picked up during the sieving process under the ambient air caused the particles to clump together. They were later pressed through the sieves by hand. It is possible that although only small amounts were lost from the handling, they may be significant. 3.3.3 A T T E R B E R G L IMITS Following the method described by Lambe (1951), samples passing through the #40 mesh or #20 mesh (Dr.R.G. Campanella, personal communication, 1990) sieve were used to determine the Atterberg limits through the hydrometer tests. The results are tabulated in Table 3.2. Sample HSCF has a plastic limit (wP) and liquid limit (wjj of 32% and 45%, respectively; the corresponding values for sample HSCL are 17% and 19%. Fig.3.19 illustrates the relative positions of the two clay samples (HSCL and HSCF) on a plasticity chart. Both samples could be classified as CL or ML (Lambe and Whitman, 1968). Activity of a clay is defined as the ratio of the plasticity index to clay fraction (Grim, 1962; Lambe, 1951). Activity values were determined to be 0.76, and 0.38 for samples HSCF and HSCL respectively. Based on Skempton's classification, the gouge HSCF is considered normal and HSCL inactive (Grim, 1962; Skempton, 1953). Sample wP(%) wL(%) wn(%) IP CF(%) A HSCF 32.29 45.3 31.29 13.01 17 0.76 HSCL 17.39 19.5 5.26 1.91 5 0.38 where wp = plasticity limit wL = liquid limit wn = natural water content by weight Ip = plasticity index = wL-Wp CF = clay fraction A = activity = IP/CF Table 3.2 Atterberg Limits for Samples HSCF and HSCL 44 3.3.4 X-RAY DIFFRACTION ANALYSIS X-ray diffraction (XRD) techniques were used to determine the clay mineralogy of shear zone materials. A total of eleven samples of shear zone fillings collected at various locations were tested4. Seven of these contained mixtures of smectite, chlorite and amphibole. Three of the eleven samples contained talc. To identify the smectite mineral, two samples (HSCF and HSCL) were selected for more detailed XRD analyses (See Appendix B for XRD procedure used). The X-ray diffraction pattern for sample HSCL indicates that the gouge is largely clinochlore, a chlorite mineral, mixed with fibrous actinolite, which is not unexpected since it is present in the host rock, greenstone (see section 3.2.1). Sample HSCF, however, contains a mixture of clinochlore, kaolinite, montmorillonite and actinolite. Appendices B-1 to B-5 show the diffraction patterns of the two samples using different treatments. The identity of the clay minerals helps to explain the activity values obtained in the Atterberg limit tests. Chlorite and kaolinite are considered inactive clays, whereas montmorillonite and chlorite mixed-layer assemblages with montmorillonite are considered normal to active (Grim, 1962). Although chlorite is considered to be inactive, its behaviour as a shear material is not fully understood, since virtually no work on its shear strength or determination thereof has been documented. 4 X-ray diffraction analyses were performed by the Geological Survey of Canada, Ottawa. Results for samples HSCF and HSCL were confirmed by additional XRD analyses performed by the author at U.B.C. 45 3.4 CLIMATIC CONDITIONS AT THE SLIDE SITE FOR DECEMBER, 1964 AND JANUARY, 1965 Mathews and McTaggart (1969, 1978) suggested that the weather conditions at the time of the 1965 rock avalanche were not a contributing factor to slope failure. Nevertheless, the local meteorological conditions at the Hope Slide were re-examined. There was no permanent weather station^n operation at Uie^ Hope Slide (HS) location (currently located at Lat. 4jj^l7WiJ^g^J^LtJdLW,j674 m elevation) until 1973. The two closest weather stations in 1965 w e r e A i r p o r t (HA, 39 m in ejevatioj^ tocated 18 km. normwesLQ_fjhe„site at elevation 1341 m)^ 32tomJo the southeast. Neither HA nor AP is truly rejiresentative of die slide site, which is higher in elevation than Hope Airport, and much lower than at Allison Pass. In order to establish representative meteorological conditions at the time of the 1965 slide, short term averages for precipitation and temperatures for the slide site were determined based on records between 1979 and 1988. Based on the 10 year averages, the slide site receives a total annual precipitation of 1160 mm, of which 931 mm fall as rain and 241 cm as snow. By comparing the short term averages (1979-1988) for HS and the thirty year mean values (1951-1980) for HA and AP, estimates of rainfall, snowfall, and mean daily temperatures for the two months of December 1964 and January 1965 could be made. Appendix C summarizes the comparisons of precipitation and temperatures at all three locations (HA, AP and HS). For the two months of December, 1964 and January, 1965, an estimated total of 156 mm and 142.9 mm of precipitation, respectively, would have fallen at the base of the Hope Slide (Appendix C). Mean daily temperatures were -2.5°C and -4.2°C; mean daily maximum temperatures were 0.6°C and -0.9°C; and mean daily minimum temperatures were -5.5°C and -7.4°C, respectively for December, 1964 and January, 1965. 46 It was confirmed that unusually cold temperatures prevailed for more than three weeks prior to the 1965 slide event (Fig.3.20). 33% of the precipitation received in December, 1964 and January, 1965 fell on the twenty five days prior to the slope failure (Dec. 16, 1964 to Jan.9, 1965) (Fig.3.21). It is reasonable to assume that a minimum of 20% of the precipitation fell as snow at the base of the slide site (weather station location), and much more at higher elevations on the mountain side. Mean daily temperatures were consistently well below freezing during the twenty five day period from December 16, 1964 to January 9, 1965. Judging from the weather conditions in 1964/65, it is possible that the ground just below the surface could have been frozen, but the thick snow cover on the slope would have acted as insulation, preventing the ground from freezing deeply. The prevailing cold conditions would also have prevented appreciable thawing during the period of December 16, 1964 to January 9, 1965. 3.5 DRAINAGE CONDITIONS Seepage observed during field investigations was minimal, but this is not unexpected for several reasons: (a) the rock mass is highly fractured, which facilitates good drainage, particularly if the fractures are throughgoing and inter-connected; (b) the ridge and headscarp areas are topographically high, where recharge rather than discharge occurs; (c) field investigations were conducted in the warmest months of the year, when recharge is at a seasonal low. Seepage was found at only two locations in Domain 1, both were located at sheared discontinuities. At the middle felsite and greenstone exposures, a number of springs can be found at or near the lithologic contacts, indicating that the water table is close to the surface (Fig.3.2). Water flow from these springs appeared to be continuous, although the 47 discharge volume seemed small at the time of field investigations. Water is channeled toward a pre-existing chute, which disappears beneath the debris at lower elevations. Other minor seepage can be found, usually around brecciated zones or lithologic contacts (type 2 in section 3.2.2.2) along the northwest flank of the slide scar. The volume of discharge at these locations was also small during the months of July through September. ' • * *» * V.' - ^. <•» V* . •* ./ >'"9iJr, •• LEGEND greenstone (Hozameen Complex) granodiorne (Mount Outram Pkiton) buff felsHe white felsite quartz vein Ithologic contact fault (deflried,approxtmated,assumed) 1865 side imit 500 LOOKOUT ..B.EHSSE3: Fig.3.1 Local geology of the 1965 Hope Slide detachment surface. Outcrops of the greenstone, the two ty.pes.Tof f e l s i t e and granodiorite, and the two f a u l t s within the study area are shown. The white areas are covered by debris or vegetation. ."KOHSEX.:: Fig.3.2 S t r u c t u r a l and morphological features at the s l i d e s i t e . Fig.3.3 Oblique photograph of the middle exposure (see Fig.2.2 for location) of a thick felsite sheet in contact with greenstone, looking south (photography by K.W. Savigny, 1989). 52 W — E Legend (%) 5 to 10 10 to 15 I | 15 to 20 l^Hli 20 to max. Fig.3.5 Lower hemisphere projection of the three major joint sets, Jl, J2 and J3 in Domain 1 (see Fig.2.2 for location) on an equal area stereonet, based on a total number of 1910 data points and 1% counting area. Fig.3.6 Lower hemisphere projection of the three major joint sets, Jl, J2 and J3 in Domain 2 (see Fig.2.2 for location) on an equal area stereonet, based on 5904 data points and 1 % counting area. Fig.3.7 Vertical stereo triplet of the 1965 Hope rock avalanche, showing the slide surface (B.C. Government Photographs BC 5124-023,024,025, 1965). Fig.3.8 Oblique photograph looking northeast along the upper near vertical fault (UF). Note also the younger buff felsite (BF) crossing the older greyish-white felsite (WF). SC=scree cones, solid square indicates the sampling location of the fault gouge, (see also Fig.2.2, Photography by K.W. Savigny, 1989). Ul Fig.3.9 Oblique photograph looking north, showing the two faults identified on the slide surface. The upper fault (UF) is prominent on any photograph; the lower fault (LF) is easier to identify in an oblique view such as this one. The solid squares indicate the sampling locations of gouge (see also Fig.2.2, photography by K.W. Savigny, 1989). Fig.3.10 An example of a brecciated zone, located at approximately 1600 m along the northwestern flank. Note the highly disturbed rock mass in the centre to the left of the photograph, and the more massive greenstone to the right. Fig.3.11 Topographic map indicating positions of the tension cracks (darker lines) and movement hub locations (5 m contour interval). The arrow indicates the location of a boulder laying across a crack (see Fig.5.12). LA OO 59 Photograph A Fig.3.12 Oblique photographs taken in 1988 and 1990, illustrating opening of the tension cracks north of the headscarp apex (see Fig.2.2 for location). Note the different (arrow) positions of the boulder laying across one of the larger cracks in 1988 (photograph A) and 1990 (photograph B). (Photography by author, 1988 and K.W. Savigny, 1990). 59 a Photograph B Fig.3.12 Oblique photographs taken in 1988 and 1990, illustrating opening of the tension cracks north of the headscarp apex (see Fig.2.2 for location). Note the different (arrow) positions of the boulder laying across one of the larger cracks in 1988 (photograph A) and 1990 (photograph B). (Photography by author, 1988 and K.W. Savigny, 1990). 60 8000.00 -i O 6000.00 Q_ O 4000.00 O CO 2000.00 -0.00 i i ) i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i 0.00 2000.00 4000.00 6000.00 800C Normal Load (kPa) Fig.3.13 Strength envelope for the massive greenstone based on direct shear tests. 61 5000.00 H O 0- 4000.00 Q < 3000.00 O < 2000.00 LU 1000.00 H 0 0 0-00 11111111111111111111111111111111111111111111111111111111 0.00 1000.00 2000.00 3000.00 4000.00 5000.00 NORMAL LOAD (kPa) Fig.3.14 Strength envelope for the schistose greenstone based on direct shear test. 62 Fig.3.15 Strength envelope for the buff felsite based on direct shear test. 63 7000-6000-"o" -^ 5000 : -^ 4000 : o : i • SHEAR [ O O O O o o i i i 1 i i i i 1 i 1000-0 -0 1 1 1 ' 1 I I 1 1 I 1 1 1 1 I 1 1000 2000 3000 • NORMAL i i i [• i i 4000 LOAD 1 1 1 1 1 1 1 | 1 1 1 1 | 5000 6000 7000 (kPa) Fig.3.16 Strength envelope for the granodiorite based on direct shear test. Sample HSCF GRAIN 8IZE (nun) Fig.3.17 Grain size distribution of sample HSCF. Sample HSCL Fig.3.18 Grain size distribution of sample HSCL. Fig.3.19 Plasticity chart showing relative positions of the two gouge samples (HSCF and HSCL). 67 x 3 2 -22 COMPARISON OF MEAN DAILY TEMPERATURE AT THE HOPE SLIDE SITE 0 -Ar ~ -6 - . -8 -10 -12 -f -14 -16 -18 H -20 iff T i i i i i i—i—i 1 —i—i—i—i—i—i—i—i—i— r Dec.ia7 18 19 20 21 22 23 24 25 26 27 28 29 30 3Uan.l2 3 4 5 6 [//\ HS 84/65 VALUES 1964/65 DATE HS 10 YR MEAN VALUE Fig 3.20 Bar graph of'daily mean temperature for the period of December 16, 1964 to January 9, 1965 The 1964 and 1965 values are estimations based on comparisons of the 30 year normalized (1951-1980) records from the Hope Airport, Allison Pass and short term averages (1979-1988) from the Hope Slide stations 68 COMPARISON OF DAILY PRECIPITATION (mm) 24 AT THE HOPE SLIDE SITE a a z o i M Pm M u 3! o 1 Dec. 137 18 10 20 21 22 23 24 25 26 27 28 20 30 3Uan.l2 3 4 5 8 7 a \7~A HS 1864/85 VALUES 1064/65 DATE HS 10 YR MEAN VALUE Fig.3.21 Bar graph of daily precipitation for the period of December 16, 1964 to January 9, 1965. 1964 and 1965 values are estimations based on comparisons of the 30 year normalized (1951-1980) records from the Hope Airport, Allison Pass and short term averages (1979-1988) from the Hope Slide stations. 69 4 MECHANISMS OF THE SLOPE FAILURE 4.1 PREHISTORIC LANDSLIDE Numerous postglacial landslides have been documented in southwestern British Columbia. In the landslide inventory of the Hope-Fraser-Coquihalla corridors (Savigny, in prep.), more than thirty landslides were identified. In the Hope area alone, there are at least four major rock avalanches (Table 4.1), the Hope Slide being the only site with a confirmed record of multiple landslides. Location Hope Slide #1 Hope Slide #2 Cheam Slide Katz Slide date of occurrence based on available C 1 4 dates (years B.P.) unless otherwise stated 9680+320 (Mathews and McTaggart, 1978) Jan. 9, 1965 (Mathews and McTaggart, 1969) 4350±70 (Naumann, 1990) 4690+80 (Naumann, 1990) 5010+70 (Clague, personal communication, 1991) 3260+70 (Naumann, 1990) Lake of the Woods Slide 8260±70 8430+60 (Naumann, 1990) (Naumann, 1990) Table 4.1 A list of the four known rock avalanches and radiocarbon dates in the Hope area. Except for the 1965 Hope Slide, all of them occurred during prehistoric time. The 1965 Hope Slide was not a singular event at the site; a prehistoric landslide occurred at the same location. This prehistoric landslide was first recognized by Dawson in 1877 (1879). On aerial photographs taken before 1965, the prehistoric landslide scar can be clearly delineated (Fig. 1.3). 70 Based on interpretation of the pre-1965 aerial photographs, several important geomorphic features are revealed (Fig.4.1), which probably influenced the failure behaviour of the prehistoric and/or the 1965 slide: (i) The source of the prehistoric event was derived mainly from the northeast (Mathews and McTaggart, 1969), the same direction as the 1965 event; (ii) There were two pre-existing chutes within the zone of detachment, one originating from the north above the fault (CI in Fig.4.1), which continued downslope and still occurs as a chute today (below the middle exposures in Fig.2.2). The other originated along the lower northwestern flank (C2 in Fig.4.1) and has remained, relatively unchanged; (iii) The upper fault identified in this study (Figs.3.1 and 3.7) was a prominent lineament before 1965 in the same way that it is today; (iv) A large number of lineaments can be identified on the pre-1965 slope (i.e. the prehistoric rupture surface (Fig.4.1), most of them trend northwest or northeast. Several of these warrant special attention: • The pre-existing chute originating above the fault (CI) continued down the slope along the lineament labelled as LI. If LI is extended upslope toward the 1965 headscarp apex (see Fig.2.2 and Appendix A-3), it reaches a location where a (type 2a) shear zone was identified during this study (Fig.3.2), suggesting that LI is likely a tectonic feature; • Below (to the south of) a rock outcrop at the intersection of the lineaments labelled as L2 and L3 (Fig.4.1), located below the main fault near the centre of the prehistoric slide surface, some displacement or severe weathering (indicated by a deep notch) appears to have occurred. This outcrop coincides with the location of the current exposure of buff felsite and greenstone (middle exposures in Fig.2.2). This location coincidence and lineament L2 71 suggests that a felsite unit may have daylighted on the slope here before 1965; • A series of lineaments (labelled as LS, Fig.4.1) were identified between the prehistoric headscarp and the upper northwestern margin of the 1965 slide. They appeared to continue along the ridge to the south and southeast. These may have been tension cracks, just as there are tension cracks currently found near the 1965 headscarp apex (section 3.4); (v) Large transverse linear features, which will be referred to as "trenches" hereafter, can be seen on the upper slope of the mountain (trenches will be described in the following section); (vi) The headscarp of the prehistoric landslide can be seen crossing one such trench (Fig.4.1); and (vii) A small displacement, similar to the one described by Croasdale (1989) who suggested that it occurred during or after the 1965 event, can be found near the same location at the fault (labelled as SD on Fig.4.1). Mathews and McTaggart (1978) collected fragments of tree roots for radiocarbon (C 1 4 ) dating to estimate the time of occurrence of the prehistoric slide, which gave a date of 9680+320 years B.P. They suggested that the 1965 slide involved the collapse of the headwall of the prehistoric landslide, and that the prehistoric rock avalanche was probably very similar in volume to that of the 1965 event. An incipient landslide, located north of the 1965 slide scar, between approximately 850 m and 1100 m in elevation, was reported by Mathews and McTaggart (1969) based on the evidence of disturbance of massive to slightly schistose greenstone, irregular topography and series of open fractures that were filled with talus. Although this particular incipient landslide was not thoroughly investigated in this study, its description by Mathews and McTaggart (1969) contains similarities of small scale local disturbances of 72 the rock mass in the area bounded by outcrops number 1 and 3 (see Appendix A-3 for location), located at elevations 1775 m to 1800 m southeast of the headscarp (von Sacken et al., 1989). Here, blocks of massive greenstone bounded by discrete joints are dislodged in a "pie-shaped" configuration, in which two steeply dipping joints (J2 and J3) and one dipping sub-parallel to the slope (Jl or the related shallower joint, referred to in section 3.2.2.1) intersect. A similar configuration can be discerned on a larger scale at the 1965 headscarp apex (see Fig.2.2 for reference location, Fig.4.2). 4.2 1965 ROCK AVALANCHE 4.2.1 SOURCE AREA Mathews and McTaggart (1969) constructed an isopach map for the 1965 Hope Slide by dividing the slide surface on the pre- and post-slide topographic maps5 into square grids of 30x30 m2 (100x100 ft2). Fig.4.3 shows the isopach map indicating the areas of debris lost and gained at the site, the upper fault that crosses the slide surface and the limit of the prehistoric landslide. The upper fault and the limit of the prehistoric landslide are included in Fig.4.3 for comparison. Much of the material lost during the 1965 event came from the northeastern quarter of the slide, encompassing and extending beyond the headscarp of the prehistoric slide (Figs. 1.3, 4.1 and 4.3). A maximum depth of rock removed from the mountain side was 147 m (480 feet, Mathews and McTaggart, 1969). Along the lower northwestern flank below the upper fault, it is obvious that the 1965 slide scoured deeper into the mountain side, up to 50 m in places, leaving behind a fresher looking cliff (Fig. 1.4). More than half the material (approximately 26x106 m3, based on 5 Pre-1965 and post-1965 slide topographic maps were prepared by McElhanney Geosurveys Ltd., Vancouver, B.C. for the B.C. Ministry of Highways in 1965. 73 volume calculations in this study, see section 4.2.3) in the 1965 event came from between elevations of 1500 m and 1800 m, at and above the upper fault (Figs.4.3). The headscarp apex (see Fig.2.2 for location) of the 1965 slide, like the prehistoric slide scar, crosses and is partially bounded by a trench (section 4.1), which is located at approximately 1800 m on the north side of the headscarp apex (Figs.2.2, 3.2 and Appendix A-3 in pocket). Traverses made on the lower southern half of the slide (Fig.2.2) indicate that relatively thin layers of rock, or no rock at all, were removed from this part of the slope. It was here that Mathews and McTaggart (1969) suggested that the travelling debris was "clearly ... locally airborne", based on the evidence that original tree trunks were cut off several metres above the ground. The southern limit of the 1965 Hope Slide is bounded by a pre-existing snow avalanche track, which also remained relatively undisturbed. The two pre-existing chutes on the prehistoric slide surface described in section 4.1 are still discernible after the 1965 event. The larger of the two (labelled as pre-existing chute in Fig.2.2 and CI in Fig.4.1) located near the middle of the 1965 rupture surface probably played a role in diverting the slide debris during the landslide. There are many physical similarities between the prehistoric and 1965 slides: the general shape and outline of the headscarps of the 1965 and prehistoric events are very similar (Figs.4.1 and 4.3); in both events, the source material came from the northeast quarter of the slide surface (Mathews and McTaggart, 1969, 1978); felsite appears to have occurred near the centre of and daylight on both detachment surfaces. All these suggest that the prehistoric and 1965 slides may have failed in much the same way: rupture occurred along intrusive contacts in the lower part of the slope, the slide mass was possibly separated by the near vertical upper fault, and the upper slope was largely controlled by the orientation of joints. 74 4.2.2 TRENCHES Several trenches are found adjacent to the 1965 headscarp near the mountain top. They typically trend sub-parallel to the ridge axis (Figs.3.2 and 3.5, and Appendix A-3) and have been known to exist since 1969 (Mathews and McTaggart, 1969; Mollard, 1976). These large linear features have also been termed as "antislope scarps", "ridge top depressions" or "sackung" (Bovis, 1982; Jahn, 1964; Radbruch-Hall et al., 1978). The trenches at the slide site are sparsely vegetated with small trees and bushes. The trees within these trenches appear to be stunted compared to surrounding trees, and almost all show signs of growth compensation, possibly as a result of slope movements. Underlying the vegetation is till-like sediment, locally up to 6 m thick (Mathews and McTaggart, 1969), although it is often difficult to measure the thickness because of the ground cover. The thickness of sediment in the trench which the 1965 headscarp crosses is approximately half a metre. Trenches may have originated from tension cracks, attributable to ridge-spreading or stress relaxation caused by cyclical glacial loading and unloading, steepening of the mountain sides by glacial erosion, and/or mass wasting processes (Mollard, 1977). The Hope Slide site is a unique location for verification of this theory, because the trenches seen above the 1965 headscarp continue towards the southeast and merge with moraines about 3 km away from the slide site, between Johnson Peak and Mount Outram. Research is currently ongoing to determine whether the trenches are related to glacial processes. Radiocarbon dating of organic material from a lake impounded by the moraine will be used to determine the age of ice that created the moraine, which is presently assumed to represent the maximum elevation of ice during the last glaciation, hence, a long-standing elevation control. Numerical simulation of the crustal effect of differential ice loading across the mountain slope are also part of these ongoing studies (Ryder, J.M. and Savigny, K.W., personal communication, 1991). 75 Close examination of the Hope Slide indicates that discernible and substantial deformation of the rock mass appears to be restricted to portions of the slope lying (topographically) below these large trenches. The trench located north of the headscarp apex (Fig.3.2) at approximately 1800 m elevation, appears to coincide with the upslope limit of deformation. The rock mass (topographically) above this trench and to the south appears to be relatively intact6. The relationship between these trenches and the upslope limit of deformation is unclear. A similar association between slope deformation and trenches has also been recognized at Affliction Creek (Bovis, 1982) and at the Mystery Creek Slide, north of Whistler, British Columbia (Evans, personal communication, 1991). As mentioned before, the 1965 headscarp also cuts a trench (Figs.3.2 and 3.5; this is the same one located at 1800 m elevation north of the headscarp apex previously mentioned). An oblique view of this trench to the south suggests that it may continue and follow existing discontinuities (Fig.4.4). Due to limited vertical exposure, it cannot be confirmed how the trenches are related to the discontinuities. 4.2.3 VOLUME Two previous independent volume calculations have been made on the 1965 Hope Slide. Mathews and McTaggart (1969) estimated the volume by summing the volumes of each grid square of the isopach map they produced (Mathews, W.M., personal communication, 1989). They calculated the volume lost to be 47.3xl06 m3 with no change in the estimated volume after the slide. 6 "Intact" is used, hereafter, to refer to rock mass that presently displays relatively minor to no evidence of displacement. 76 Bruce and Cruden (1977) estimated a volume of 47.7x10° nr for material lost from the upper slope, but the volume of material gained at the bottom of the slide was estimated to be 43.3xl06 m3, a 9% volume reduction. It is puzzling to not have a net volume increase after a landslide, since the break-up of the rock mass should theoretically increase its volume, typically by a bulking factor of about 15% to 40% (Savigny, K.W., personal communication, 1990). For this reason, another attempt was made to estimate the volume of the 1965 landslide in this study. The new topographic maps prepared during this study were not used in the volume calculations, because an updated pre-1965 topographic map at an appropriate scale was needed. Since the comparisons of volume should be consistent in units, the original pre-and post-slide topographic maps made in 1965 were used instead. These were also the ones used by Mathews and McTaggart (1969) and Bruce and Cruden (1977). Two attempts to estimate the volumes were made. In the first trial, the pre-slide and post-slide topographic maps were digitized. Using a graphics software package7 and a grid of 100x100, three dimensional grid surfaces were produced. Then the volume difference between the pre- and post-slide surfaces were calculated. The results indicate that a total of 33xl06 m3 of rock were lost. This value is 30% less than either estimates of Mathews and McTaggart (1969) or Bruce and Cruden (1977). The volume gained at the base of the slide was approximated at 28xl06 m3, a net decrease of 16%. In the second attempt, the overall accuracy of the isopach map (Mathews and McTaggart, 1969) was first verified by its reconstruction using a coarser (120x120 m2) grid size. It was then digitized, and the volumes lost and gained were calculated using the same computer software. The volume of the source rock was calculated to be 48.3xl06 m3 and the volume of material deposited at the bottom of the slide totaled 43.3xl06 m3. These and previous results are tabulated in Table 4.2: 7 Software package, QUICKSURF 2.91, was used for volume calculations. It is developed by Schreiber Instruments, Denver, Colorado. 77 Volume lost Volume eained (xlO6 m3) (xlO6 m3) calculated by Case 1 Case 2 Case 3 Case 4 47.3 47.7 33.0 48.3 47.3 43.3 27.8 43.3 Mathews and McTaggart(1969) Bruce and Cruden (1977) trial 1, this study trial 2, this study Table 4.2 Comparison of volume estimates. Cases 1, 2, and 4 were based on the isopach map by Mathews and McTaggart (1969). Case 3 was based on pre- and post-slide topographic maps prepared in 1965. The results from the second attempt closely resemble those by Bruce and Cruden (1977). The net volume loss in both cases is approximately 10%. The estimates of volume of material lost in all cases, except for trial 1 of this study (Case 3 in Table 4.2), are all very close. The 30% difference in volume between trial 1 and the others can be attributed to the choice of grid size, which could be refined and improved by using finer grids. Grids of up to 400x400 can be used with QUICKSURF. However, available computer memory space and compatibility, process and execution times became major limitations. All of the calculations, except trial 1 (Case 3), were based on the same isopach map (Mathews and McTaggart, 1969), which was in turn based on the same original topographic maps. It seems more than coincidental that no net volume increase is observed. Possible sources of errors or uncertainties for the volume calculations are: (1) Aerial photographs taken prior to the 1965 slide did not contain sufficient number and distribution of ground controls. (2) Heavy coniferous tree cover in much of the lower southern half of the detachment zone before 1965 obscured the ground surface of the pre-slide topography. (3) Outram (Beaver) Lake in the valley (Fig. 1.3) was thought to be a shallow pond (Mathews and McTaggart, 1969), but its depth was unknown. If it was much deeper than expected, it could have accepted a greater volume of debris. 78 (4) The slide debris may have displaced a large (and unknown) volume of soft sediments which were lost as mudflows in the valley. Neither the displaced mud nor the volume of debris below the old valley floor are accounted for by the volume calculations. (5) In the first attempt of volume estimation (trial 1 in this study), three dimensional surfaces were generated for both the pre- and post-slide which consisted of 10,000 grid squares. QUICKSURF was used to calculate the volume under each grid of each surface, and then the grid volumes were summed to give a total volume. The volumes lost and gained were taken as the difference between the pre- and post-slide surfaces. Errors due to the smoothing effect and the approximation of the irregular edges of the slide margin by rectangular columns could be significant. It is suspected that the inaccuracy of the pre-slide topography and the unaccounted volume of displaced mud flows as discussed (points 1,2 and 4) above are the primary causes of the apparent net volume loss. 4.2.4 GEOMETRY AND GEOLOGICAL CONTROL OF THE 1965 FAILURE SURFACE The mode of failure of the 1965 Hope Slide was translational. The failure surface is believed to have been controlled by two types of discontinuities: joints and lithologic contacts containing gouge. Of the three dominant joint sets found at the study site (Figs.3.5 and 3.6), the only one that would favour translation is Jl (45 7273°), which does not daylight on the upper slope (Domain 1, Fig.2.2). Therefore, some other factor or factors must be involved in 79 order for translational movement to occur. At lower elevations in the slide area (Domain 2, Fig.2.2), Jl has a mean orientation of 40°/253°, which approximates more closely the overall direction of the natural slope gradient (Fig.2.2) and the average pre-1965 slope angle of 37°. The line of intersection between Jl from Domain 1 (45 7273°) and Jl from Domain 2 (407253°) trends at 236°, which correlates extremely well with the direction of longest runout (238°, Figs.3.5 and 3.6). Any shallower joint, dipping less than 40° in a southwesterly direction would result in a line of intersection trending in the same general direction as 238°. The steeply dipping joints, J2 and J3, facilitate tensional or dilatational displacement that is compatible with movement along Jl toward the southwest. Close examination of the joint data in Domain 1 (Fig.2.2 and section 3.2.2.1) indicates that approximately 10% of the measured joints have shallow dip angles averaging approximately 30°. These joints would daylight on the slope and could facilitate sliding of the rock mass. Profiles parallel to the dip direction of Jl (270°) based on 5 m contour intervals suggest that the upper slope (Domain 1) has a "step-like" geometry, in which steeper (Jl) and shallower surfaces alternate (Figs.4.5 to 4.7). Fig.4.5 shows the locations of the cross sections in Figs.4.6 and 4.7. In Fig.4.6, the shallower surfaces vary between 20° and 31°, the steeper surfaces vary between 42° and 51°, It can be seen in this figure that the step-like geometry continues upslope in the undisturbed rock beyond the headscarp. In Fig.4.7, the shallower surfaces appear to be fairly consistent, dipping at 30°; the steeper surfaces are parallel to Jl, dipping at 45°. The surface of the 1965 detachment is steeper in the upper portion (above approximately 1450 m, Fig.2.1) than the lower portion (Fig.2.1). The shallower inclination of the lower slope approximates the orientation of the contacts between the buff felsite sheets and greenstone. The orientation of these contacts and the local shearing 80 observed by Mathews and McTaggart (1969) led them to conclude that they partly contributed to the slope failure. In summary, the 1965 failure surface was controlled by two mechanisms: the upper part of the slope was largely influenced by the pervasive jointing and attendant step geometry, and the lower portion of the slope was controlled by the weakness of the pre-existing gouge filled contacts between the buff felsite and greenstone. Lateral control of the 1965 landslide was most likely due to the pre-existing brecciated zones (type 2a of the pre-existing shear zones in section 3.2.2.2) common along the northwestern flank. The southern margin of the slide surface was influenced by a pre-existing avalanche chute (Fig.4.1). 4.2.5 ROCKSLroE KINEMATICS As pointed out by Mathews and McTaggart (1969), much of the debris was deposited around a pre-existing kame ridge at the base of the slide, which may have momentarily slowed or halted some of the debris. As illustrated in Fig.4.8, travelling rock fragments would initially pile up on the stoss (east) side of the kame until the "trough" was filled, the moving mass would then travel over the top of the kame. If the momentum of the debris was sufficiently great, debris could be launched over the former Outram Lake, creating a shadow zone. This theory is supported by the present topographically low area, where the former lake was located. The pre-existing chute in the middle of the slide below the middle exposures (as mentioned in section 4.1, Figs.4.1 and 2.2) probably diverted the colluvium slightly towards the west and/or caused the fast moving debris descending from the headscarp area to be partially airborne over the lower southern portion of the slide surface (outcrop 81 number 9, Appendix A-3), which was relatively undisturbed (Mathews and McTaggart, 1969). The pre-existing chute and kame must have affected, at least partially, the distribution of the debris in the valley. Because of the way the slide debris was deposited (roughly sorted by rock type and hence by colour, in roughly concentric ridges), Mathews and McTaggart (1969, 1978) suggested that much of the momentum may have been effectively transferred to the voluminous muddy sediment in the valley floor. Shreve (1966) described other large rock slides which had exhibited similar behaviour, where the slide mass acquired a very high speed in its descent, and upon encountering a projecting shelf of rock (such as outcrop number 9, Appendix A-3), the slide mass launched itself as a relatively coherent unit on a cushion of trapped compressed air. Momentum was rapidly dissipated when the leading edge of the airborne mass hit the ground causing comminution and deposition of the rock (Shreve, 1966), which caused the displacement of the soft valley sediment. The distribution and deposition of the debris as roughly concentric ridges also reflect some of the original layering of the felsites and greenstones, and the influence of the pre-existing landforms. The kame and chute not only affected the distribution of the colluvium, they could also have altered the direction of movement of the sliding mass, perhaps significant enough to override the effects of the joint orientations. The direction of movement of the 1965 event has been assumed to be parallel to the longest ruout direction (238°). If the detachment occurred along Jl joint surfaces, movement should be along 273° (the mean dip direction of Jl and the related shallower joints, Fig.3.5). However, the natural gradient of the detachment surface is toward the southwest (approximately 240°), which also resembles the longest runout direction. Thus, a direction of movement between approximately 240° and 270° could be expected. The direction of movement of the centre of mass, which can be estimated from the isopach map (Mathews and McTaggart, 1969), gives approximately 250°. Several factors could have influenced the movement direction: 82 (1) The prehistoric slide scarp would deflect at least some of the debris toward the south and southwest. (2) The pre-existing chute in the middle of the slope would channel moving debris toward the west-southwest at lower elevations. (3) The pre-existing kame in the valley and the outcrop in the lower elevations (outcrop number 9, Appendix A-3) would influence the flow and hence the distribution of debris at the foot of the slide, as discussed above. 4.2.6 POSSIBILITY OF TWO SEPARATE SLIDE EVENTS The seismic data of the two small earthquakes associated with the 1965 Hope Slide were recently re-evaluated by Wetmiller and Evans (1989), who suggested a seismic trigger could not be ruled out. It remained unclear what the seismic cause and effect relationship was, if any. Keefer (1984) studied 40 earthquake induced landslides throughout the world. He concluded that there was no clear evidence of earthquakes having magnitudes of less than 4.0 causing any type of landslide, and that the Hope Slide may be a rare exception. In all previous publications, the 1965 Hope Slide was considered to have occurred as one event, a consequence of the two seismic events. New evidence outlined in the preceding sections leads to an alternate hypothesis that the Hope Slide was two events occurring several hours apart, each of which caused a measurable seismic event. The first earthquake at the Hope Slide supposedly caused a snowslide which blocked the Hope-Princeton Highway and westbound vehicles at around 4:00 a.m (Anderson, 1965). However it is possible that this blockage was due to a rock slide, rather than a snowslide. Since it was still dark and a rock slide would have incorporated the snow 83 cover from the mountain side, it would have been difficult to distinguish one from the other. The buff felsite sheets dip sub-parallel to the slide surface at approximately 30° to 35° (Fig.2.1 and section 3.2.1). At least parts of their contacts with greenstone should have daylighted on the mountain side and interpretation of high level aerial photographs suggests that the felsite may have daylighted near the current middle exposures (Figs.2.2 and 4.1). The hypothesis is advanced here that the first phase of detachment occurred along these litho-tectonic contacts. The headscarp of the first phase of movement is believed to have been controlled by the upper near vertical fault trending northeast across the slope (Fig.2.1 and UF in Fig.3.8), just as it formed the headscarp of the prehistoric slide. With systematic debuttressing of the upper part of the slope since the prehistoric event, additional oversteepening of the toe caused by the first phase of detachment, coupled with the existence of pervasive joints, the upper portion of the slope subsequently failed a few hours later (Fig.4.9). In April, 1990, Brenda Mine in south-central British Columbia experienced a collapse of an open mine pit wall, in which 8xl06 tons of material were removed (Weichert et al., 1990). This slope failure was registered as a regional seismic event of a local magnitude of 2.3 (Weichert et al., 1990). In light of this data together with the structural data summarized in the foregoing paragraphs, it is proposed that the 1965 Hope Slide occurred in two stages and that the gravity displacements induced seismic shocks equivalent to earthquakes of ML=3.2 and ML=3.1, respectively (Weichert et al., 1990). Although there remains the possibility that there was only one rock avalanche which occurred approximately three hours after a small seismic event (ML=3.2, which caused a snow avalanche), and that the mass movement induced a seismic signature similar to a ML=3.1 earthquake. Based on the available evidence, this hypothesis seems too conjectural and coincidental to be put forward. 84 4.2.7 SEISMIC ENERGY CONSIDERATIONS FROM THE TWO SLIDE EVENTS Energy dissipated from a large landslide can be transferred into seismic energy, but topographic amplification, geology, pattern and path of disintegration of the failure block can complicate this transfer process (Dr.R.M. Ellis, Dept. of Geophysics and Astronomy, U.B.C., personal communication, 1991). Coupling effects between landslides and earthquakes and associated energy transfer efficiency are of great interest in areas where earthquake activity is common (e.g. the whole west coast of North America). Investigation and comparison of energy and seismic signal are currently ongoing to determine the possible links between landslides and earthquakes (Weichert et al., 1990). Although a detailed analysis of this type is beyond the scope of this thesis, rough estimates can be made by comparing the potential energy released during the Hope Slide with the seismic energy of the associated earthquakes. It is assumed that the proposed two events (in section 4.2.6) were equal in volume, which produced similar seismic events. The change in potential energy is given by: mass (m) = = density (p) x Volume (V) unit weight (7) = density (p) x gravity (g) h = height (elevation) difference between initial and final centres of gravity of the slide mass. Using equation [4.3] assuming the same 7=28.6 kN/m3 and a total volume V=24.15xl06m3 (from this study) and h=365 m (half of 730 m, Mathews and McTaggart, 1969), the potential energy released by the first stage of the mass movement is estimated to be 2.5xl021 ergs. The seismic energy transmitted by an earthquake can be calculated using the formula (Richter, 1958): A PE = mgh A PE = pVgh A PE = yVh [4.1] [4.2] [4.3] where 85 log E = 5.8 + 2.4 mb [4.4] where mb = unified magnitude derived from body waves E = energy of seismic event (ergs) Weichert suggests using equation [4.4] with the assumption of m b=ML. The commonly used energy formula: M s = magnitude derived from surface waves was not used here, because the poor seismograph data gave only the local magnitude (MjJ, and there is no general direct relationship between M L and M 8 (Weichert, D., personal communication, 1991). So, energy for the earthquake of local magnitude of 3.2 was calculated to be 3xl013 ergs, this is nine orders of magnitude lower than the potential energy released during the Hope Slide. Similarly, the released potential energy and seismic energy (for ML=3.1) for the second phase of the slope failure can be calculated. Knowing the volume, and assuming a vertical displacement of 300 m (1000 ft, Weichert, D., personal communication, 1991) and the same rock density as for the Hope Slide, the collapse at the Brenda Mine released approximately 2.37X1020 ergs in potential energy. The energy of the associated seismic event is estimated to be 2.1xlOn ergs. The following table summarizes this energy comparison: logE = 11.8 + 1.5 M, s [4.5] where volume from [4.4] M L APE (ergs) E ^ ^ (ergs) transfer efficiency - Eseismic / a P E Location (xlO* m3) Hope Slide first phase 24.15 second phase 24.15 3.2 2.5X1021 3.1 2.5xl021 2.3 2.4xl020 3xl013 6.9xl012 Brenda Mine 2.76 2.1xlOn 8.9X10'10 Table 4.3 Comparison of the potential energy released during the slide events and the associated seismic energy at the Hope Slide site and the Brenda Mine sites. 86 As demonstrated by this comparison, if a seismic signature equivalent to a ML=2.3 earthquake was produced by the slope failure at the Brenda Mine, which was significantly smaller in volume than the Hope Slide (Table 4.3), the Hope Slide must have produced a seismic shock(s) comparable to, if not stronger, than that produced by the Brenda Mine failure. Hence, the two small earthquakes recorded at the time of the 1965 Hope Slide are viewed as effects rather than causes of the two failures. Also from Table 4.3, it can be seen that the only a very small proportion (less than one millionth) of the potential energy was transferred into seismic energy in both cases. The efficiency in the transfer process was much greater at the Hope Slide than at the Brenda Mine site, meaning that more of the potential energy issued from the mass movements at the Hope Slide was transferred into seismic energy that produced the signals recorded, and less was dissipated through other means. 87 Fig.4.1 Vertical aerial photograph of the pre-1965 slope showing the pre-historic slide scar, which can be clearly delineated. Numerous pre-existing lineaments and trenches can be identified. Current exposure of the buff felsite and greenstone is located at L2 and L 3 ; CI and C2 are pre-existing chutes; LS=series of lineaments (tension cracks?); SD=a small pre-existing displacement (after Croasdale, 1988). (B.C. Government Photograph BC4014-24, 1961). 88 Fig.4.2 The geometry at the 1965 headscarp resembles a "pie shape" configuration, possibly controlled by joint sets Jl, J2 and J3 from Domain 1. Similar dislodgement geometry was noted along the ridge. 89 Fig.4.3 Isopach map of the 1965 Hope Slide (after Mathews and McTaggart, 1969). The approximate limit of the prehistoric slide and the upper fault are shown. 90 Fig.4.4 Oblique photograph showing the continuation of a trench into the rock mass at the headscarp apex (looking south, photography by K.W. Savigny, 1989). Fig.4.5 Index map showing the locations of cross sections in Fig.4.5 (XX') and Fig.4.6 (YY'). 92 1950-. Fig.4.6 An example of "step-like" profile (XX') sub-parallel to the dip direction of Jl (270°) on the upper slope of the Hope Slide (within Domain 1), based on 1:1000 topographic map. The steeper and shallower surfaces in this profile range from 42° to 51° and 20° to 31°, respectively. 93 1 9 0 0 - , '1800 H C o o > JL> 1 700 OJ 1600 intact slope outcrop below headscarp on slide surface w ~ i — i — r i i i i i i I ' ' 100 200 h o r i z o n t a l d i s t a n c e ( m ) i—i—i—r 300 Fig.4.7 Another example of "step-like" profile (YY') sub-parallel to 270° on the upper slope of the Hope Slide (within Domain 1), based on 1:1000 topographic map. The steeper and shallower surfaces in this profile dip at relatively consistent angles of 45° and 30°, respectively. 94 Fig.4.8 A schematic diagram illustrating the possible sequence of deposition of travelling debris, channelled by the pre-existing chute, against the pre-existing kame (not to scale). sw Fig.4.9 A schematic diagram (plan and vertical views) illustrating the possible two phases of sliding, in which the first phase controlled by the buff felsite sheets and the upper fault (partially after Mathews and McTaggart, 1969; not to scale). VO U i 96 5 STABILITY ANALYSIS OF THE 1965 HOPE SLIDE 5.1 OVERVIEW OF SLOPE STABILITY Slope stability is largely dependant on the local geology, material properties, extent and orientation of discontinuities, pore water pressures, and stress deformation characteristics of the site. External influences such as erosion, seismic loading, heavy and prolonged precipitation and human activities can also affect the stability of a slope. Numerous methods of analysis have been developed to assess the risk of slope failure. Most of these fall into two broad categories: limit equilibrium or stress-strain analysis. For this study, a two dimensional, limit equilibrium method based on Sarma's method of slices (1973, 1979) was chosen. Slope stability analysis is generally undertaken before failure occurs, on the basis of measured shear strength parameters, pore pressures and geometric configuration. In the case of the Hope Slide, many of the parameters are unknown and cannot be ascertained. The water table was never measured prior to the 1965 event, and furthermore, there is no precipitation or temperature record for the site until 1973. The cohesion on the 1965 failure surface cannot be measured or confirmed, since much of the actual detachment surface is covered by debris and any gouge would have been removed or highly altered by the occurrence of the slide and subsequent weathering. Because of the many unknown factors which affected the strength of the 1965 sliding surface, stability analyses were performed first by using reasonably wide ranges of parameters. As will be seen in later sections, up to five variables were considered in the analysis, so it was impractical to examine all the permutations of the parameters. More detailed analyses were performed using those combinations of parameters that gave marginally stable slopes (where F = 1). 97 5.2 METHOD OF ANALYSIS 5.2.1 LIMIT EQUILIBRIUM ANALYSIS The limit equilibrium method was chosen for this study because it is a relatively simple method based on the shape of the slip surface, which is known, and requires modest data input. Various limit equilibrium techniques have been developed to evaluate problems involving the safety of slopes. The problem of slope stability is often statically d^eterminate (i.e. the number of unknowns is greater than the number of available equations). However, if simplifying assumptions are made, the limit equilibrium approach enables one to find a solution of a slope stability problem by simple statics. Some of the common assumptions include a simple failure surface; the material above the slip surface is considered to be a homogeneous "free-body"; the disturbing and resisting forces above the failure surface are estimated to formulate equations of force and/or moment equilibrium (Chowdhury, 1978). The solution of the equations provides quantitative information regarding the risk of slope failure. The slip surface associated with a certain set of conditions that gives a minimum factor of safety, F, is theoretically the critical slip surface, where F is a measure or an indicator of stability. The factor of safety (F) depends on: (a) the total mobilized shear stress, T (i.e. driving forces) and (b) the total available shear strength, s (i.e. resisting forces). F is defined as: F = E resisting forces / E driving forces [5.1] F = s / T [5.2] 98 A value of F > 1 indicates a stable slope, in contrast, F < 1 indicates an unstable slope. If F = l , the slope in question is in a critical state of equilibrium (i.e. S=T). Based on the linear Mohr-Coulomb failure criterion, s can be expressed as a function of normal stress: s = c + a tan <p [5.3] where c = cohesion a = normal stress 4> = angle of internal friction. 5.2.2 SG-SLOPE, METHOD OF SLICES In most rock slopes, the sliding surface is controlled by existing discontinuities. The Hope Slide is no exception. Once the failure surface and the geometric configuration of the moving mass are identified, they can be represented by a two dimensional cross section, which is divided into a number of slices (Fig.5.1). The forces on each slice are identified in Fig.5.2. Using the linear Mohr-Coulomb failure criterion (equation [5.3]), the unknown force and moment equilibriums are solved. A computer program called SG-Slope (developed by Sperling, 1991) which is based on the method of slices (Sarma, 1973, 1979; Hoek, 1986) was used to evaluate the stability conditions of the 1965 Hope landslide. Sperling (1991) compares and summarizes the advantages and disadvantages of different limit equilibrium techniques. He concludes that Sarma's method is a more powerful two dimensional analytical tool than the others because this method not only incorporates the desirable features of older methods which allow non-circular slip surfaces and heterogeneous slope materials within the profile, but also includes advantageous features that permit inclined slice boundaries; the usage of another dimensionless stability indicator (critical horizontal acceleration); and the fact that it has a matching number of unknowns and equations renders it a rigorous method. 99 In addition to the factor of safety (F), SG-Slope also quantifies slope stability in terms of critical horizontal acceleration (KJ (Sarma, 1979). A value of K^<0 indicates an unstable slope, whereas, K c>0 indicates that a slope is stable. A slope with 1^ =0 is in a state of limiting equilibrium (i.e. F = l). Both K<. and F are valid indicators for the stability of a slope, although F is probably more familiar to many practitioners, since most of the existing case studies use this parameter. The solution strategy used in SG-Slope is briefly described as follows (Sperling, 1991): • The slope is divided into a maximum of 10 slices, which can have inclined sides; • Body forces and their points of application are computed based on slice geometries; • A system of equations for horizontal and vertical equilibrium, and interstice relationships based on the Mohr-Coulomb failure criterion are combined to solve for K c ; • Once Kj. is known, other unknown forces are calculated; • Moment equilibrium and the corresponding factor of safety, F, are computed; and finally • The solution should be checked to ensure that it is kinematically acceptable. The fundamentals of using SG-Slope, as in any stability analysis, are that the strength parameters, number of slices to be analysed in a selected profile, and the failure surface geometry should all be represented as accurately as possible. The advantages of using SG-Slope include: (1) Although SG-Slope is based on the concepts and methodology introduced by Sarma (1979), it obtains its solution by directly solving the system of linear equations rather than the closed form solution approach used in Sarma (1979) or Hoek(1986); 100 (2) Hoek (1986) addressed, but did not solve, the problem of "severe numerical instability" or non-convergence of the critical horizontal acceleration (Kc) if inappropriate F values were used in the function relating the two stability indicators. For a given F, there is only one corresponding K value which satisfies all the equations, but for a given value of K , several roots of F may exist (Sperling, 1991). SG-Slope overcomes this problem by employing a new iterative procedure which systematically narrows the increment, dF, until the increment straddling K=0 is reached. (3) An additional check for moment equilibrium, as recommended by Hoek (1986), is also included in SG-Slope. (4) SG-Slope is very easy to use. The user can make the input as complex or simple as desired by following the prompts and menus on screen. ANALYTICAL CONSIDERATIONS Before analysing the Hope Slide, the following must be considered: (i) a cross section of the slope must be chosen to represent the intact slide mass; (ii) the choice of slices through the selected cross section must accurately represent the basal failure surface and ground surface topography; (iii) the slice geometry must conform to the rules outlined by Sperling (1991) in order to ensure acceptability criteria for SG-Slope are satisfied; (iv) the slices must be defined so that each can be assumed to be homogeneous; and (v) reasonable values or ranges of values for parameters such as cohesion (c), water table ratio (R), angle of internal friction (<p) must be selected for analysis. 101 Two cross sections of the 1965 Hope Slide were chosen for stability analyses. Fig.5.3 shows the locations of these cross sections. Profile AA' consists of a cross section from the 1965 headscarp to elevation 1000 m, parallel to the line of longest runout (238° azimuth, Fig.5.4). This cross section is also approximately the centre line of the detachment area. The exposed rock mass in this profile is comprised entirely of greenstone. It is assumed that the same lithological uniformity existed on the pre-1965 slope. Profile BB' (Fig.5.5) is parallel to AA'. This cross section is selected because it includes exposures of the buff felsite and greenstone, which are assumed to have comprised the 1965 slip surface. Based on the clayey gouge found at these contacts along the lower northwestern flank and the evidence of weathering seen on pre-1965 aerial photographs, it is further assumed that gouge material existed at the pre-slide felsite-greenstone contacts. Both cross sections contain the outcrop of the upper fault that trends northeast across the slide. The two selected cross sections end at 1000 m elevation because when the pre- and post-1965 profiles are compared (Fig.2.1), it can be seen that virtually no rock was removed below this part of the slope during the 1965 event (outcrop number 9 in Appendix A-3). Moreover, undisturbed tree trunks and pre-existing chutes suggest that the slide mass may have been locally airborne (Mathews and McTaggart, 1969) and/or diverted to the west (see section 4.2.1). 5.3.1 ASSUMPTIONS Several assumptions are required for Sarma's method of slices and SG-Slope: (1) the point of application of interslice normal force passes through the center of each slice side; (2) the shear strength on interslice boundaries is fully mobilized; and (3) the linear 102 Mohr-Coulomb failure criterion is followed. Additional assumptions have been made to further simplify the problem: (i) The mode of failure was translation along planes of weakness that were controlled by discontinuities; (ii) The discontinuities were through-going and continuous; (iii) Since there is no obvious boundary observed in the field between the two types of greenstone, and the <pb values of both the massive and schistose greenstones fall within the range of <t> used in the analyses (see also section 5.3.2.5), the massive and slightly schistose greenstone are not differentiated; (iv) All buff felsite-greenstone contacts below (topographically) and along the main fault crossing the slide contained clayey gouge; (v) Although the clay minerals identified from the lithologic contact (sample HSCL) and the fault (sample HSCF) are different, strength parameters based on the clay minerals from sample HSCF (mixture of montmorillonite, kaolinite and chlorite) were used in the analyses due to the lack of information (especially <f> values) for chloritic shear materials (see also section 5.3.2.5); (vi) The influence of second order irregularities (Parton, 1966) on available exposed Jl joint surfaces around the headscarp is considered to be negligible. If roughness of a larger scale (first order irregularities or waviness) affected the strength of the slope, it is taken into account by adding angle i to ^ in the analysis (see section 5.3.2.5) or by the existence of cohesion (c); (vii) Although a seismic trigger remains a possibility in the 1965 event, external forces due to earthquake loading are not considered in the stability analyses. 103 5.3.2 PARAMETERS FOR ANALYSIS 5.3.2.1 BASAL FAILURE SURFACE In an earlier report by Bruce and Cruden (1977), the 1965 failure surface was approximated as a plane dipping at 30° in the direction of slide motion. The actual bearing was not specified. The 1965 detachment surface cannot be adequately represented as one plane because it steepens up-slope (Fig.2.1). As mentioned in chapter four, there are strong indications that the 1965 event failed along intrusive contacts on the lower part of the slope, while the upper part failure slide surface appears to be mainly controlled by step-like discontinuities (Figs.4.5 and 4.6). For the slope stability calculations, the failure surface and the pre-1965 topography are represented by straight line segments based on detailed cross sections (Figs.5.4 and 5.5). The slice sides in profiles AA' and BB' are presumed to be largely controlled by the J2 or J3 discontinuities, which have a near vertical dip. All buff felsite-greenstone contacts are presumed to contain gouge infill, which is represented by thick black lines in Figs.5.4 and 5.5. 5.3.2.2 UNIT WEIGHT (7) Unit weights for the massive greenstone, buff felsite and schistose greenstone were calculated using y = pg (Brown, 1981), where p is the measured bulk density. The unit weights of the schistose and massive greenstones are very similar, 28.62 kN/m3 and 28.32 kN/m3, respectively, so the two types of greenstone were not differentiated in the stability analyses, and a value of 28.6 kN/m3 was used. The unit weight of buff felsite was determined to be 24.7 kN/m3. 104 5.3.2.3 COHESION (c) The results from direct shear tests of saw cut surfaces of greenstone samples gave cohesion intercepts on the order of 100 kPa (Figs.3.13 and 3.14), but these values are presumably artifacts of data fitting since smooth sawn surfaces should not give any cohesion. According to Hoek and Bray (1977), depending upon how much the natural rock has been disturbed, hard rock with discontinuities and minor clay content can have cohesion values of up to 150 kPa, whereas cohesion in rock with higher clay rmneral content can range from 0 to 50 kPa. Considering that the rock mass at the slide site is highly fractured, has undergone both tectonic and gravity displacement, and that the Jl joint surfaces are smooth, the overall cohesion is likely low. The sensitivity analysis results in Fig.5.6 shows that as cohesion increases from 0 to 50 kPa, the friction angle required to maintain F = 1 decreases by 1° to 2°. The factor of safety is more sensitive to changes in cohesion for a small slide or a relatively thin failure block. For thicker slide masses, such as the Hope Slide, the normal stress (a) term in equation [5.3], which is proportional to the thickness of overburden, is the dominant factor, thus greater uncertainty in cohesion can be tolerated. Cohesion effects for the 1965 Hope Slide would presumably be determined by the pre-existing discontinuities and the infill material, which may have contained gouge. Cohesion values are assumed to range from 0 to 200 kPa for discontinuities with rock to rock contacts and 0 to 50 kPa for contacts containing gouge (Hoek and Bray, 1977). 105 5.3.2.4 WATER TABLE RATIO (R) Pore pressures influence the frictional strength of the rock discontinuities. Increases in pore pressure reduce the effective normal stress on the failure surface (Lambe and Whitman, 1968): a' = a - u [5.4] where a' = effective stress a = total stress u = pore pressure = T w h where 7 W = unit weight of water h = height of water table in a column of rock or soil. When the effective stress is taken into account, the pore pressure, u, influences the available shear strength (equation [5.3]) according to the Terzaghi-Coulomb equation. This is found by substituting [5.4] into [5.3]. When F = l , S=T, therefore T = c' + (a - u) tan </>' [5.5] In SG-Slope a dimensionless variable, R, known as the water table ratio is introduced to represent the height of the water column in each slice of the failure block. R is defined as the ratio of the distance (h) from the base of a slice i to the water table intersection with slice side i to the total slice side length (z) (see Fig.5.2). R has values between 0 and 1.0, where 0 means a dry slope and 1 indicates a fully saturated slope. The relationship between u and R is best explained in terms of the conventional pore pressure ratio, ru, which is defined as the ratio between the total upward force due to water pressure to the total downward force due to overburden pressure acting on the failure surface (Hunt, 1986): r u = (7 w h)/( 7 s z) [5.6a] 106 ru = (Tw / 7s) (h / Z) ru = (Tw / 7s) R [5.6b] [5.6c] where 7 S = unit weight of overburden z = height of the overburden column For the rock types at the slide site, 7 S = 2.5 to 2.8 7 W r u • R [5.7a] [5.7b] The results from the sensitivity analysis in Fig.5.7 shows the effects of R on the friction angle (4>) required to maintain a stable slope (i.e. F = l). As the water table rises, the friction angle increases in a non-linear fashion. With no cohesion, increasing R from 0 to 0.25 is compensated for by a 1° increase in <p, while increasing R from 0.75 to 1.0 requires a 4° increase in <f>. Comparison of Figs.5.6 and 5.7 shows that R and c are of comparable importance. As detailed in section 3.4 and Appendix C, excess pore pressures can probably be ruled out in the case of the 1965 Hope Slide, because for more than three weeks before the disaster occurred, the temperatures remained well below freezing. Therefore, high R values can be overlooked. 5.3.2.5 ANGLE OF INTERNAL FRICTION (<f>) AND STRENGTH CONSIDERATIONS Deformation within a rock mass is induced by tectonic, gravitational or geological processes. External loading (e.g. earthquakes), creep (very slow progressive straining), excess pore pressures and weathering can also cause deformation. For a slope to fail catastrophically, the strength of a potential slide surface must be overcome. The concept of peak and residual strengths is illustrated in Fig.5.8. This typical 107 deformation curve for rocks can be divided into several regions: OA, AB, BC and CD. In region OA, the deformation behaviour of a rock is elastic. In the AB region, the slope of the stress-strain curve decreases to zero with increasing stress. Here irreversible deformation is induced in the rock. A quasi-stable condition occurs in which a slope may deform plastically but re-adjust to a new equilibrium until sufficient stress is applied to cause destabilization. This process is known as hysteresis (Jaeger and Cook, 1971). Any additional stresses in this zone such as seismic loading may help to surpass the peak strength of the potential slip surface. If the peak strength (point B) is overcome, catastrophic failure occurs. For first-time slides in undisturbed rock, peak strength along discontinuities should be operative. Zone BC is characterized by a negative slope, where deformation is generally brittle. In real situations, sudden failure would occur at B and strain softening probably could not be seen or traced. Re-orientation of platy minerals, if any, would take place as shear strength continues to decrease towards the residual value (Jaeger and Cook, 1971). Zone CD represents the residual strength condition where considerable displacement can occur. Since the slope at the Hope Slide site has undergone tectonic displacement along the discontinuities, indicated by the shear gouge, and has a history of previous failure(s), the angle of shearing resistance may already have been reduced to its residual value <£r over some portion of the ultimate failure surface. Peak strength would not have been operative on the entire 1965 failure surface at the Hope Slide since the existence of gouge and shear zones clearly demonstrates that shearing occurred before 1965. The linear features (tension cracks and trenches) above the headscarp of the prehistoric slide are also evidence of displacement prior to the 1965 event (Fig.4.1 and section 4.1). In regions of the slope where stresses could not be sustained, stresses would be repeatedly transferred to neighbouring stronger regions until the overall slope became unstable (Grivas and Souflis, 1985). The strength of a discontinuity can be greatly affected by the occurrence and extent of clayey filling. If the thickness of shear gouge is greater than the amplitude of the 108 asperities, then the strength of the rock is controlled by the strength of the clay content; otherwise, a slope is largely controlled by rock-to-rock contacts (Goodman, 1970). Residual strength values are assumed for the analysis since if any progressive shearing had occurred before the 1965 landslide, it was most likely seated along the gouge filled pre-existing shear zones, which include the fault and buff felsite-greenstone contacts. Since gouge was found at the greenstone-buff felsite contacts, an estimate of <j> for this type of shear material (<pc) is needed. A review of the literature indicates that very limited work has been published on the determination of strength parameters for gouge comprised of chlorite. The only reference found was unpublished data, generously provided by Mr. Don Pollock of the Prairie Farm Rehabilitation Administration (personal communication, 1990), in which evaluations of the residual friction angle (4>r) are based, not on clay mineralogy, but on the liquid limits of various chloritic clays. For clays with liquid limits around 40-45%, which are similar to the liquid limit of sample HSCF, <pr is estimated to be between 12° and 18°. Correlations between plasticity index (IP) and residual friction angle (<f>t) were made by Voight (1973) and Kanji (1974) for natural soils. Voight (1973) published an empirical relationship, which gives a <f>T of 27° for a clay with a plasticity index Ip of 13%, which is similar to sample HSCF. Kanji (1974) expanded on the work by Voight (1973) and showed that the basic shape of the curve (</>r versus Ip) remains the same for a variety of clayey soils (Fig.l in Kanji, 1974) and that there is a well defined relationship between <f>t and Ip which can be expressed as: <t>T = 46.6 / I p 0 4 4 6 [5.9] This formula gives a <pr value of 15° for sample HSCF. In contrast to chlorite, much more work has been done on the more commonly occurring clay minerals: montmorillonite, kaolinite, illite and bentonite. Kenney (1967) conducted extensive testing on these different clays to determine the correlation between 109 mineralogy and residual friction angle. His results show that for pure or nearly pure montmorillonite, 4>T ranges from 4° to 10°, which is in general agreement with Skempton's results (1985). For natural soils containing 5% montmorillonite or mixed layer clays containing montmorillonite, <j>T can range anywhere between 10° and 29°. Table 5.1 summarizes these findings: <pr based on from 12°-18° liquid limits* 27° plasticity index** 15° plasticity index * * 10°-29° montmorillonitic clays Don Pollock (personal Communication, 1990) Voight (1973) Kanji (1974) Kenney (1967) * based on chloritic clays ** based on natural soils that may contain some chlorite Table 5.1 Possible 4>r values for gouge sample HSCF. Because of the scarcity of references on the residual friction angles (<pr) for chlorite, and the fact that the gouge sample from the lithologic contact (sample HSCL) was relatively small and gritty, a mixed clay (like sample HSCF: mixed layers of montmorillonite, kaolinite and chlorite) was assumed to exist as the infill of discontinuities at the Hope Slide. From the foregoing discussion, the gouge sample HSCF could have a 4>t value ranging from 10° to 29°. However, these values are based on studies of soils containing far more than 20% clay fraction. Skempton (1985) shows that if the clay fraction is less than approximately 25% (as for samples HSCL and HSCF), then the material behaves more like a silt with <f>T typically greater than 20°. Therefore, the friction angles for the gouge were chosen to range from 20° to 28° for the analysis, which was arbitrarily divided into 4>c of 24° to 28° and 20° to 24° for clearer graphical presentation. As shown earlier in equation [5.3], tan <f> is directly proportional to the total available shear strength (s), which is a material property that can be measured by direct 110 shear tests of rock samples. From section 3.3.1, direct shear test results show that the basic friction angles (fa) for massive greenstone, schistose greenstone and buff felsite are 34°, 31° and 35°, respectively. As Patton (1966) has demonstrated, fa is not necessarily equal to <f>, rather: 4> ~ fa + i [5.8] where i = average angle of inclination of surface irregularities to the shear direction fa = basic friction angle of planar rock surface. In order to account for possible large scale roughness on greenstone joint surfaces ( i«2°) , a wider range of 30° < <p < 40° (Deere and Miller, 1966; Hoek and Bray, 1977) was used for the slope stability analyses. This range was arbitrarily divided into two ranges of <£g, 34° to 40° and 30° to 36°, mainly to clarify the presentation in this thesis. It should be noted that from the basic friction angles for greenstone (<pt,=31° and 34° obtained by laboratory testing), <pg<36° ((f>g=fa+i) was predicted even when i=2° is included. 5.4 STATIC BACK ANALYSIS When a slope is about to fail, it is considered to be in a critical state of limiting equilibrium where the factor of safety equals unity. Using this constraint, the average shearing resistance of a slope at the time of failure can be estimated. However, depending on the number of parameters involved, there can be numerous permutations that give F=1. By assigning and changing the value of one unknown parameter at a time, holding all others constant, it is possible to estimate a set of parameters which would satisfy the critical condition of F=1. I l l Static back analyses were performed for profile AA 1, assuming no gouge was present on the basal surface. The numerical results can be found listed in Table 5.2 and are presented as parts of Figs. 5.6 and 5.7. For the case of zero cohesion and a dry slope (R=0), the 0 g required for F = l was 31.2°. This correlates with the lower <pt, of 31° for greenstone, obtained from the direct shear tests (see section 5.3.2.5), and suggests that the lower limit chosen for 4>g is appropriate. The shaded areas in Fig.5.7 indicate the range of <pb measured by laboratory testing of the greenstone, and the resulting range of 0 g (<pD+i) if a roughness contribution of i=2° is included. The points in the shaded zone represent sets of parameters (R, c and 4>) that are consistent with the measured strength of the greenstone. Values of <£ F = 1 >36° and 0 F = 1 < 3 1 ° appear to represent strengths of the greenstone that are too high or too low (i.e. inconsistent with <pD or <pb+i). This suggests that unless the water table ratio is high (approximately >0.50), cohesion values of greater than 100 kPa are unreasonable. Conversely, for the expected range of cohesion (0 < c„ < 100 kPa), R should be less than 0.50. 5.5 STABILITY ANALYSES AND DISCUSSION This section presents stability analyses for eight different scenarios using the parameters summarized in Tables 5.3 and 5.4. About half of these scenarios use profile BB' where gouge was assumed to exist at the lithologic contacts. Although the static back analysis for the case of F=l are preferable in principle (as performed in the previous section for profile AA1), they are very laborious because F is being calculated for a given parameter, namely <pg while holding c and R constant. For each of the points shown in Figs.5.6 and 5.7, approximately six iterations were required in order to obtain F = l . The analyses presented in this section use a slightly different approach, in which F is calculated for various values of <p, c and R. Only those sets of 112 Run no. R c (kPa) CIA 0 0 31.15 C1R1 0.25 0 32.2 C1R2 0.5 0 34.1 C1R3 0.75 0 37.1 C1R4 1.0 0 41.15 C1B 0 50 29.75 C1RC1 0.25 50 30.75 C1RC1B 0.5 50 32.6 C1RC1C 0.75 50 35.5 CIRC ID 1.0 50 39.55 C1B2 0 100 28.3 C1RC2 0.25 100 29.25 C1RC2A 0.5 100 31.05 C1RC2B 0.75 100 33.85 C1RC2C 1.0 100 37.85 C1B4A 0 200 25.3 C1RC3 0.25 200 26.2 C1RC3A 0.50 200 27.8 C1RC3B 0.75 200 30.4 C1RC3C 1.0 200 34.25 Table 5.2 Summary of stability analysis results using profile AA* for F=l Scenario Figure Profile in text all greenstone except gouge on slice base(B) or slice side(S) STRENGTH RANGE rt (°) (iVa) (kPa) STRENGTH RANGE 4>& <f>c c„ cc tf (°) (gPa) (kPa) 1 5.10 AA' S3 34-40 24-28 0-200 0-50 A=3 A=2 A = 100 A=25 2 5.11 AA* S3 30-36 20-24 0 0 A=3 A=2 0 0 3 5.12 BB' B2,4,6;S7 (solid line) 34-40 24-28 0-200 0-50 S7 (dashed line) A=3 A=2 A =100 A=25 4 5.13 BB' B2,4,6;S7 (solid line) S7 (dashed line) 30-36 20-24 0 0 A=3 A=2 0 0 A = increments used in analyses Table 5.3 Summary of parameters used for Scenarios 1 to 4 Scenario Figure Profile in text all greenstone except gouge on slice base(B) or slice side(S) STRENGTH PARAMETERS <p„ c„ <f>c cc O (KPa) (°) (kPa) Comments 5 5.14 Case 2 B2,4,6;S7 32-36 0-100 22 A=2 A =100 0 cg=0 (dashed line) cj = 100 (solid line) 6 5.15 Case 2 S7 32-36 0-100 22 A=2 A =100 0 c„=0 (dashed line) cj = 100 (solid line) 7 5.16 Case 2 B2,4,6;S7 32-36 0-100 22 A=2 A = 100 0 cg=0 (dashed line) c^  = 100 (solid line) 8 5.17 Case 2 SI (upper 3 slices of Scenario 5) 32-36 0-100 22 A=2 A =100 0 cg=0 (dashed line) (•1 = 100 (soUd line) A = increments used in analyses Table 5.4 Summary of parameters for Scenarios 5 to 8 115 parameters that are consistent with a marginally stable slope (i.e. 1 ^ F ^ 1.1) are selected for further analyses.. This provides similar information but requires only one iteration per parameter set. Note that the surface roughness, represented by the angle i, is not explicitly dealt with, rather it can be considered as part of the <pg value (i.e. <t>g=<f>ys+ i) used in the analyses. The maximum^  reasonable range for each of the parameters, as determined in section 5.3.2, is applied to the first four scenarios. From the results of these four scenarios, more restricted ranges of parameters, which gave marginal stability, were chosen for the analyses of scenarios five to eight. The last two scenarios assume two separate slide events (see section 4.2.6 and Table 5.4), where the lower part of the cross section was mobilized during the first event (scenario 7). The flow chart in Fig.5.9 illustrates the approach of the analyses and the break down of the eight scenarios, which will be described in more detail in the following paragraphs. Given the limited knowledge of the cohesion, water table ratio, the nature and extent of the gouge on the 1965 failure surface and the possible inaccuracy of the pre-1965 topography, a detailed analysis of the slope failure is not warranted. Rather than covering all the possible combinations of parameters (more than 800 calculations would be necessary even if only 3 values for each parameter were attempted), the intent of analyses discussed in this section is to demonstrate that there are sets of parameters for which consistency can be found at the time of failure. Scenarios 1 and 2 deal with cross section AA' (Fig.5.3), consisting entirely of greenstone and including the gouge-filled sub-vertical fault. In this profile, the fault outcrops at slice side 3 (side 1 is at the toe of the profile) as shown in Fig.5.4. It was found that the presence of the fault gouge has virtually no effect on the stability of this profile. This is attributable to the non-critical orientation of the fault (i.e. near vertical). Figs.5.10 and 5.11 show the factor of safety (F) plotted against R for six combinations of <pg, <f>c, c„ and cc. The shaded region corresponds to the sets of parameters which give a 116 slope with 1 < F < 1.1. Strength conditions above the shaded band corresponds to relatively stable slopes, while the area below the shaded zone gives relatively unstable slopes. Clearly, for <£g>34°, values of R greater than 0.50 would be required to destabilize the slope. Increases in cohesion would obviously strengthen the slope. The results indicate that values of <£g>36° are not representative of those mobilized since the corresponding factors of safety are unreasonably high. In scenario 2, where cohesion was assumed to be zero for both rock-rock and rock-gouge-rock interfaces, results show that a minimum <pg of 31° was required to maintain stability in a dry slope (Fig.5.11). The case of cg=0 kPa, R=0 and <£g=30° is worth mentioning here and comparing with the results of Bruce and Cruden (1977). They analysed the Hope Slide under the assumptions that both cohesion and pore pressures were zero, and using <pi,=30o obtained from direct shear tests on sawn felsite blocks, calculated a factor of safety of 1.05. This is 10% higher than the value of 0.95 obtained in this study. However, there are significant differences between the two analyses: (1) The static back analysis in this study was performed with a profile consisting entirely of greenstone, along which failure occurred; whereas Bruce and Cruden (1977) assumed failure occurred through jointed felsite, for which they showed a lesser <£b value. (2) Bruce and Cruden (1977) also assumed the basal surface was a single plane dipping at 30° in the direction of motion, which they did not specify. As mentioned earlier, the failure surface should not be represented by one average plane, since it is steeper in the upper portion than lower elevations (section 5.3.2.1). Cross section BB' (Fig.5.5) was used for scenarios 3 and 4, each of which was attempted with gouge (solid lines in Figs.5.12 and 5.13) and without gouge (dashed lines in Figs.5.12 and 5.13) at the felsite-greenstone contacts (at the bases of slice number 2,4 117 and 6, Fig.5.5) and at the fault outcrop (slice side 7, Fig.5.5). Blocks with felsite sheets comprising the basal zone are represented in Fig.5.5 as hatched areas, and the gouge is represented by thick black lines. It would have been more realistic if the felsite sheets represented by slices 2,4 and 6 were thinner and overlain by greenstone, but one of the assumptions made was that the geology within each slice in the profile must be homogeneous. Although the felsite and greenstone have slightly different unit weights, the factor of safety was checked to vary by only 1 % or less depending on the choice of the unit weight for the three hatched slices in Fig.5.5. The results of scenarios 3 and 4 (Figs.5.12 and 5.13) indicate that the stability of the slope was sensitive to the existence of the gouge on the basal failure plane. This is indicated by the separation between dashed and solid lines for the same parameters in Figs.5.12 and 5.13. The factor of safety was reduced by a minimum of 5% where gouge was assumed to be present. This is significant when considering most engineered slopes are designed for F = 1.1 or 1.2. The permutations of variables considered in scenario 3 enable the elimination of the intermediate to high end parameters, since the corresponding factors of safety are greater than 1.2. Those considered to be reasonable in scenario 3 fall within the range of parameters used in scenario 4. The results in Fig.5.13 indicate that the low end parameters used in scenario 4 gave unstable slopes where the factors of safety were less than one. Hence, these strength parameters can also be eliminated. Appraisal of scenarios 3 and 4 suggests that the minimum frictional strength for greenstone (</>g) to maintain a stable slope must be greater than 33°. This is 2° higher than the results from scenarios 1 and 2. For calculations that assumed no gouge in scenarios 1 to 4, the 2° difference can only be due to the geometry of profile or the possibly inappropriate <f>c value at the fault outcrop. It is consistent with the expectation that profile BB' is a more critical profile than AA', and the occurrence of gouge which may have covered about one third of the basal surface also affects the overall strength of the slope. In view of these results, narrower ranges of parameters were chosen 118 (Table 5.4), and four more series of analyses were performed. R was limited to values ranging from 0 to 0.50 for these analyses, because higher water table ratio are unrealistic considering the seasonally low water table in the fractured rock mass and the weather conditions at the time did not promote snow thawing or infiltration. Scenario 5 is a worst case scenario, in which the stability of profile BB' with clayey gouge on the basal surface was evaluated using the narrower ranges of parameters. For R=0, this profile is in critical equilibrium for <pg=32° if cg=100 kPa and for c/>g=35° if cg=0 kPa (Fig.5.14). The latter is in accord with the laboratory determination of (pj,=34° for greenstone. It appears that a difference of 2° in the value of <f>c had small effect on the strength of the slope. This can be shown by comparing the lines representing </>g=36° and 0C=24° in scenario 4 (Fig.5.13) and the <£g=36° and <pc=22° line in scenario 5 (Fig.5.14). The line from scenario 4 is included in Fig.5.14 for easy visual comparison. The only difference between the two lines is the value of <pc. The calculated F values differ by only 0.01. This suggests that a 2° difference in the choice of 4>c did not change the stability significantly. From Fig.5.14, it appears that immediately before slope failure, for reasonable values of c and R, the most likely range of <pg values in the overall slope was approximately 35° to 36°. The water table ratio was probably low, between 0 and 0.25 and the overall cohesion along the rupture surface was small to negligible. Scenario 6 was the same as scenario 5, except gouge was excluded on the sliding surface. In the absence of the gouge, a cohesion value of 100 kPa in this scenario makes the slope very stable (dashed lines in Fig.5.15). If cohesion was zero, then <f>& values of approximately 33° to 35°, slightly lower than those suggested in scenario 5, appear to be representative of the mobilized strength. The results again confirm that the occurrence of gouge on the failure surface shifts the lines in Figs.5.14 and 5.15 vertically, reducing F by 6% to 14%. 119 As discussed in section 4.2.6, Weichert et al. (1990) suggested that the 1965 event may have occurred in two stages, which are supported by field evidence of two different controls on the development of the basal failure surface. For these reasons, stability analyses using the same parameters as scenarios 5 and 6 were also performed on the lower (scenario 7) and upper (scenario 8) portions of the slope separately. Fig.5.16 (scenario 7) illustrates the results of stability analyses on the slope containing only the lower 6 slices of profile BB' below the upper fault, which is assumed to represent the first phase of the slide (Fig.4.9). The occurrence of a number of springs in this part of the slide surface suggests a low to moderate water table should be considered, so, the water ratio, R, is presumed to range from approximately 0 to 0.35. Cohesion in this part of the slope was presumably low to negligible due to the gouge along the intrusive contacts. These considerations largely preclude the dashed lines in Fig.5.16. Hence, for a marginally stable slope, the estimates of <pg range from 32° to 36°. In scenario 8, only the top three slices of profile BB' were analysed using the same parameters for the last three scenarios. The results of the stability analyses are shown in Fig.5.17. The upper slope is considered a recharge area, no springs were observed during the field investigation, therefore, the water table ratio is likely low (R»0). Cohesion here could be higher than the lower slope because of the absence of gouge in the rock discontinuities. Even for R=0 and c=100 kPa, only the higher values of <pg>34° would give a marginally stable slope. Also, from the non linear behaviour of the curves in Fig.5.17, it is clear that the strength of this portion of the profile is sensitive to R>0. In contrast to scenario 7, only the upper dashed lines in Fig.5.17 need to be considered, since the solid lines represent strength parameters that yield extremely unstable slopes. 120 5.5.1 CONCLUSIONS OF STABILITY ANALYSES Since the purpose of the analyses of scenarios 1 to 4 was mainly to narrow down the sets of parameters chosen, their results are incorporated with those in scenarios 5 to 8 and will not be discussed separately in this section. It is clear that the analyses results, based on the selected combinations of parameters shown in Figs.5.14 to 5.17, occur as well behaved families of curves (or lines). Even though only a few sets of parameters were used for the analyses, interpolation of the results can easily be made. Since failure has already occurred, it is reasonable then to examine only those results that gave a critical slope (i.e. F » l ) . Therefore, parameters that gave F=l from scenarios 5 to 8 represent the limiting conditions of the overall strength of the slope and will be presented here. Table 5.5 tabulates the range of </>g values which give a factor of safety of one, for 0<R<0.50 and c=0 and 100 kPa: ForF=l, <pc=22°, R=0 to 0.50 Scenario description with c g = 0 kPa with cg — 100 kPa ^g 5 with gouge partially on basal surface 35°-37° 32° -344 and along the upper fault 6 with gouge along the upper fault only 33°-35° 29°-31 c 7 profile below the upper fault only 32°-39° 28°-35< with gouge partially on basal surface and along the fault 8 profile above the upper fault only >36.5° >34.5C with no gouge Table 5.5 Ranges of </>„ values for critical slopes (F=1) based on analysis results of scenarios 5 to 7, the end values of each range listed represent those at R=0 and R=0.50. R is assumed to be zero in scenario 8. 121 Comparison of the ranges of 4>g for scenarios 5 and 6 in Table 5.5 indicates that the presence of the gouge on the sliding surface reduces <f>g by about 2° to 3° for F=l. A similar effect can be obtained by increasing the cohesion of the greenstone from 0 to 100 kPa. In the absence of the gouge on the basal surface, the increase of cohesion has a slightly greater effect on <f>g (a difference of 4° rather than 3°). At the time of failure, for reasonable values of R and c, the mobilized <pg value for a slope containing gouge is approximately 32° to 37° (scenario 5), and 29° to 35° in a slope without the effect of gouge (scenario 6). For the most part, these values are very consistent with the measured strength of the greenstone (<pb + i = 33° and 36°), indicating that the stability of overall slope was critical even without additional influence. Although conditions of R>0.50 cannot be ruled out, R was probably less than approximately 0.35, the high end values corresponding to R=0.50 listed in Table 5.5 are considered less likely. Under the assumptions that <pb=34° and the gouge is absent, conditions of R>0.50 and i < l ° would be required for failure (Fig.5.15). In contrast, there are numerous sets of reasonable R, i and c values that result in F*»l (Fig.5.14). Since <£0 (31° or 34°) falls well within these ranges (the ranges of <pg in scenario 5, Table 5.5) which are consistent with failure, it can, therefore, be concluded that the existence of gouge on the detachment surface is much more likely. As was mentioned in section 5.5, the value of <pc of 22° or 22° ±2° has relatively small effects on F compared to a value of <pg±2° (Fig.5.14). So unless the shear strength of the gouge is (substantially) less than 22°, small differences in the choice of <pc appear insignificant. As indicated in the analyses, the mere occurrence of the gouge on about one third of the failure surface is significant enough to affect the overall resistance of the slope to failure. If the failure was divided into two events, then depending upon the choice of R and cg, the lower portion of the slope is predicted to detach separately for the range of <pg of 122 28° to 39°. After the removal of the lower slope, the <pg values required to sustain a stable upper slope are larger than suggesting that it was weak and would indeed be unstable, and a contribution of cohesion (>100kPa) or i>2° would be required to prevent the second stage of failure. 5.6 SEISMIC TRIGGER Wetmiller and Evans (1989) analysed the earthquakes associated with the 1965 Hope Slide and determined that the two seismic events were very similar with near identical seismic characteristics for an assumed focal depth of 18 km. They located the epicenter within a circle of radius of 20 km, centred 10 km northeast of the slide. Due to the close proximity of the epicenter and the slide site, they conclude that a seismic trigger for the 1965 event cannot be ruled out. SG-Slope uses a pseudostatic rather than a dynamic approach to the analysis. The pseudostatic analysis assumes that the effect of an earthquake can be replaced by a static horizontal seismic force. The effects on pore pressures are not considered (Hunt, 1986; Chowdhury, 1978). Pseudostatic analyses are relatively simple and are adequate as first estimates of stability. In dynamic analyses, the influences from ground acceleration, frequency, ground shaking duration and damping characteristics of the slope are also considered (Hunt, 1986). Although they better represent the in situ loading conditions, dynamic analyses are much more complex. The critical horizontal acceleration, which is one of the stability indicators calculated by SG-Slope (Sperling, 1991), can be used to evaluate the acceleration necessary to destabilize a slope. The product K -^W defines a horizontal force, such as that due to seismic loading (Sperling, 1991): 123 F = W K C [5.10] F=mg-K c [5.11] Force (F) = mass (m) • acceleration (a) [5.12] a = K c g [5.13] where W = weight of slide mass g = gravitational acceleration Following the attenuation relationship for earthquakes typical in western Canada as defined by Hasagawa etal. (1981): apk= l O e ^ R h 1 5 [5.14] where a-k = peak horizontal acceleration (cm/s2) M = magnitude of the earthquake Rh = hypocentral distance of the earthquake event (km) and knowing the magnitudes and approximate distances of the two earthquakes associated with the 1965 Hope Slide, comparisons can be made between ap^  generated by the possible causative earthquakes and K c which would cause slope failure as calculated from SG-Slope. A check was made by comparing K<. as determined by SG-Slope and the value calculated according to Seed and Goodman (1964): K c = [sin ci (F-l)] / {F [1 + (tan 2 <f>/ F2)]*} [5.15] where F = pre-earthquake factor of safety. Several different shearing resistances (c/>) were checked using profile AA', because equation [5.15] applies only to homogeneous rock masses (i.e. only one <f> value). The K<. values calculated by SG-Slope and equation [5.15] correlate very well, especially where F is close to one. Since the two earthquakes associated with the 1965 slide are similar in nearly every aspect (Wetmiller and Evans, 1989), only one of the two (ML=3.2) is considered here. 124 Depending on the choice of hypocentral distance, the peak horizontal acceleration can be calculated. The hypocentral distance is R h = [(focal depth)2 + (epicentral distance)2]^  [5.15] The epicenter was placed about 10 km northeast of the slide site, but with an uncertainty of 20 km (Wetmiller and Evans, 1989). Therefore, the epicenter could have been anywhere from directly under the slide mass (Milne's epicenter in Mathews and McTaggart, 1969; Wetmiller and Evans, 1989) to as much as 30 km away (Wetmiller and Evans, 1989). Using the minimum and maximum possible epicentral distances and assuming that the focal depth is 18 km (Wetmiller and Evans, 1989), the corresponding Rh, a^ and generated by the 3.2 earthquake can be determined from equations [5.13 and 5.14]: Epicentral distance (km) Rh(km) apk (cm/s2) Kcpk 0 18 8.4 0.0086 10 20.6 6.9 0.007 30 35 3.1 0.0032 Table 5.6 Peak acceleration and corresponding K,, values for different epicentral distances of the ML=3.2 earthquake at the 1965 Hope Slide. The stability of the slide mass would have to have been extremely marginal, arguably in a state of limiting equilibrium, for an earthquake of magnitude 3.2 to act as a trigger of the 1965 slide. Selecting from the analytical results in section 5.5 (see Appendices D-l and D-2), the two most critical cases (only these two cases gave a remotely comparable K c , all others were too high) that have associated values comparable to those in Table 5.6 are: 125 (a) a slope with gouge at slice bases 2,4, 6 and slice side 7; <pc=22°, <pg=36°, cohesions for both gouge and rock are zero, R=0.25; the calculated factor of safety before earthquake activity is 1.00 and 1^ =0.001; (b) a slope with same condition as (a) above that the gouge occurs only at the fault outcrop (slice side 7) and <pg=34°; the pre-quake factor of safety and K,, are 1.01 and 0.005, respectively. Magnitudes corresponding to the lower and upper limits of the epicentral location (ie. 0 and 30 km away) were calculated for both cases (a) and (b) above. If the epicentre was directly under the slide mass, almost imperceptible earthquakes of magnitudes 1.5 and 2.8 would be all that were necessary to trigger a mass movement for cases (a) and (b) respectively. If the seismic source was 30 km away, then events of respective magnitudes of 2.3 and 3.6 were needed to cause failure (see also Appendix D for summarized results). As previously suggested by Wetmiller and Evans (1989) unless the focal depth of the quake was much shallower than 18 km, the slope would have to be near critical equilibrium for an earthquake of ML=3.2 to trigger the failure. It should be added that the seismic source location and magnitude of the event are not the only important factors, the direction of seismic ground motion must also coincide with that of slide movement to cause significant impact. Furthermore, since the stability of the slope at the Hope Slide was marginal, any disturbances, not necessarily seismic loading, would be sufficient to induce failure. 126 Fig.5.2 Body forces and water table ratio (R=h/z) for each slice of the profile (after Sperling, 1991). PRDFILE AA 1000 600 800 1000 1200 1400 1600 1800 2000 HDRIZDNTAL DISTANCE (n) Fig.5.4 Profile AA' consisting entirely of greenstone, parallel to the direction of longest runout, selected for stability analysis. P R O F I L E B B 129 1800 900 1100 1300 1500 1700 1900 HORIZONTAL DISTANCE (n) Fig.5.5 Profile BB' includes buff felsite sheets in contact with greenstone, parallel to A A', selected for stability analysis. 45.00 -i 40.00 H ,35.00 -9~ Q_30.00 H 25.00 20.00 Sensitivity analysis using profile AA1 no gouge on sliding surface except at fault outcrop for F = 1 , R=0 to 1.0 * * * * * * R=0 OOOOO R=0.25 • • • • • R=0.50 A A A A A R=1.00 i i i 1111 i 11 11 i 1111 i i | 111 11 i 11 11 i 11 i i i i 111 i i i i i i i i 11 0.00 50 .00 100.00 150.00 200 .00 250 .00 c (kPa) Fig.5.6 Relative influence of cohesion (c) on the shear strength of greenstone (<p), a sensitivity analysis. o Sensitivity analysis using profile AA' no gouge on sliding surface except at fault outcrop for F=1, c=0 to 200 kPa Fig.5.7 Relative effects of the water table ratio (R) on the shear strength of greenstone (<p), a senstivity analysis. B (peak strength) in in OJ 7 ^ c L . D -P m / (residual strength) 0 s t r a i n Fig.5.8 A typical static stress-strain deformation curve. to RouQhness neo&ur»n«nt on joint surfaces Direct shear testing to deterrtne phKb) schistose greenstone 31 nasslve greenstone 34 buff felsite 33 gronodiorite 31 X Sensitivity Analyses to detenMne effects on phi c vs. ph* R vi. phi State bach analysis i staple slope (profile AA') assumptions' c=C R=0 phi for F- l m i 31.2 STABILITY ANALYSES STRENGTH PARAtCTERS phKQ»30-40 based on phKb>> I and references phKo-eo-ee based on scarce reference as < 30.33.36 24.2&2S c<o>-iuaasoo I ANALYSES P*<*0> ph*C> 33-34 31-33 84 20-28 (*KQ> phKc) Gauge 343-35.5 24-23 No gouo* Goug* 33.3-34.3 33-37 23-24 £3-24 No goug* 33-33 22-24 PRELIMINARY RESULTS pW(g> 33-36 ph<c> 22-24 K < 1X30 c(g) lo« 0-100 c<c> I ruRTHO) ANALYSES USING PROFILE BB' NARROWER PARAMETER RANGES  pMg>=32,K& phi(c>-22 c<g>-0.U>0 c(c)=0 R-0,025.050 without gaugv Sc.7 bviow fault — ScB above fault c«0 phl(g)=33-36 c=100 phl(g>.32-34 c=0 phKg>=33-34 c-100 too strong phKg>=32-33 c-100 too strong c»0 p h K o » 3 7 c"10O phiCg> = 3 5 - 3 6 Fig.5.9 Flow chart describing the approach and break down of the stability analysis of eight different scenarios. Scenario A, Stability analysis using profile AA' no gouge except at fault outcrop O D LL_ 0.80 -0.60 •= 0.40 IT frX * * * * * 40 28 200 50 QOOOO 37 26 100 25 • • • • • 34 24 0 0 1111 i i i i i i 11 i i i i 111 i 11 i i i i i i i i 11 i i 11 i i 11 i i i i i i i i i i 0.00 0.20 0.40 0.60 0.80 1.00 R Fig.5.10 Stability analysis of scenario 1 (profile AA'). 1.30 -n O 0.70 3 a 0.60 4 0.50 Scenario 2, Stability analysis using profile AA' c=0 kPa for both greenstone and gouge * * * * * 30 20 OOOOO 33 22 • • • • • 36 24 0.00 i i i i i i i i i i i i i i i i i | i i i i i i 0.20 0.40 i i | i i i i i i i i i | i i i i i i i i i | 0.60 0.80 1.00 R Fig.5.11 Stability analysis of scenario 2 (profile AA'). VI 1.60 -i Scenario 3, Stability analysis using profile BB1 solid lines=gouge on slice bases 2,4,6 and slice side 7, dashed lines=gouge on slice side 7 1.40 --4—' O ^ 1 . 2 0 o o -4—1 U 1.00 o 0.80 <P*„<Pc c ? c c (°) (kPa) f **_**_* 40 28 200 50 LL * * * * * 40 28 200 50 OQDQD37 26 100 25 QQQQD 37 26 100 25 • • D O D 34 24 0 0 "OrxiOOD 34 24 0 0 0.00 i 11 i i i i i | 11 i 11 i 11 11 0.20 0.40 0.60 i i i i i i i i i i i i i i i i i i 0.80 1.00 R Fig.5.12 Stability analysis of scenario 3 (profile BB'). ON 1.20 1.10 O O 0.70 -0.60 Scenario 4, Stability analysis using profile BB' c=0 kPa for both greenstone and gouge, solid lines = gouge on slice bases 2 ,4,6 and slice side 7 dashed lines = gouge on slice side 7 only %) * * * * * 36 24 * * * * * 36 24 OOOOO 3 3 22 OQDQD 3 3 22 annr jD 30 20 •QUOD 30 20 I I I I I I I I I I I I I I I I I I I I I I 0.00 0.20 0.40 11 i i 11 111 i 0.60 i i i i i i i i i i 0.80 -rrn 1.00 R Fig.5.13 Stability analysis of scenario 4 (profile BB'). 1.30 - i Scenario 5, Stability analysis using profile BB' Spc=22° and c c=0 kPa. Solid line: c g=0 kPa; dashed line: c g=100 kPa gouge on slice bases 2,4,6 and slice side 7 Li_ M — o 1.20 H 1.10 O O D 1.00 /<pc = 36/24 scenario 4 0.90 -1 32 «x. a> "7 O *T* "T* T- j^^ c • • • • • 34 •L3HQE1 34 AAAAA 36 AAAAA 36 cg(kPa) 0 100 0 100 0 100 Fig.5.14 Stability analysis of scenario 5 (profile BB' containing gouge on the failure surface, a worst case scenario). oo 1.30 Scenario 6, Stability analysis using profile BB' c/>c=22° and c c=0 kPa. Solid line: cg=0 kPa; dashed line: c g=100 kPa gouge on slice side 7 only fl».0 cg(kPaj * * * * * 32 . A . « i» . L . a. «i> ~Z Q ™ " T * ~ • • • • • 34 ••SOS 34 AAAAA 36 AAAAA 36 0.50 0 100 0 100 0 100 Fig.5.15 Stabibty analysis of scenario 6 (profile BB' withoug gouge on the failure surface). ^ VO 1.40 - i Scenario 7, Stability analysis on lower 6 slices of profile BB', gouge on bases 2,4,6 and side 7 =22° and c c=0 kPa, solid ine: cQ=0 kPa; dashed line: cQ=100 kPa <pgO cg(kPa) • * * * + * 32 0 * * * * * 32 100 • • • • • 34 0 •C3HL3H 34 100 AAA/SA 36 0 AAAAA 36 100 I I I I I I I I I I I I I I I I I I I I 0.00 0.10 0.20 111 i i | 1111 0.30 i l 11 i i i 0.40 TT-T-| 0.50 R Fig.5.16 Stability analysis of scenario 7 (the assumed first phase of sliding, profile BB'). 1.20 - i Scenario 8, Stability analysis on top 3 slices of profile BB', gouge on side 1 only (ie. side 7 in scenario 7 ) , c/>c=22 and c c=0 kPa, solid line: cg=0 kPa; dashed line: c g=100 kPa *•(") cg(kPa) * * * * * 32 mL, mL. X *L. ~Z O • • • • • 34 • H B O 34 AAAAA 36 0.60 I I I i i i i i | i i I I i I i i i | I I I I I I i i i | i i i i i i i i i | i i i i i i i i i | 0.00 0.10 0.20 0.30 0.40 0.50 0 100 0 100 0 100 R Fig.5.17 Stability analysis of scenario 8 (the assumed second phase of sliding, profile BB'). 142 6 ASSESSMENT OF FUTURE HAZARD FROM THE 1965 SLIDE PERIMETER 6.1 GENERAL The scar left by the 1965 slide is an impressive reminder of the power of nature. In view of the recurrent instability of the Hope Slide site, the perimeter of the 1965 slide scar must be regarded as a possible source of future displacement. Quantitative evaluation of rockfall hazards from the mid-northwestern flank was undertaken when the highway below the slide was re-aligned (Piteau, 1976). In light of the new data presented here, a brief assessment of possible landslide hazard from the 1965 slide margin is included as part of this study. In this chapter, a discussion of the stability of three general areas of the 1965 slide margin (Fig.6.1) is presented (see also Fig.2.2 and Appendix A-3 for locations): (1) The 1965 headscarp area east and south of the headscarp apex, including the undisturbed ridge above the 1965 slide scar (cross sections C C and DD' in Fig.6.1; outcrop number 2,3 and 4 in Appendix A-3), (2) The upper northwestern flank, from the trench at the elevation of 1800 m, north of the headscarp apex to elevation of approximately 1600 m (see Fig.6.1; outcrop number 5 in Appendix A-3), (3) The lower northwestern flank, below approximately 1600 m elevation (Fig.6.1; outcrop number 6 and 8 in Appendix A-3). The southern slide margin is considered to be a relatively stable area, since the prehistoric event did not affect this part of the slope and very little rock was removed during the 1965 event. 143 6.2 HAZARD EVALUATION (1) AREA EAST AND SOUTH OF THE HEADSCARP APEX The most likely mode of failure in this area is sliding along Jl and the less dominant shallow joints. From Fig.3.5, these shallower joints can have dip angles as low as 20°, with a similar strike orientation as Jl. Therefore, two cross sections parallel to the dip direction of Jl were selected for evaluation. These cross sections (CC and DD' in Fig.6.1) include an outcrop that remains partially on the 1965 detachment surface and continues upslope to the intact6 ridge. Figs.6.2 and 6.3 show the straight line segments used to represent profiles C C and DD1 for stability analyses. The actual profiles based on 5 m contours are shifted 10 m vertically for clarity and comparison. The step-like topography of these cross sections is presumably due to the discontinuity patterns in Jl and the related shallower joints. In Figs.6.2 and 6.3, the 1965 failure surface, presumably the surface of possible future failure in the rest of the slope, is extended from the slide surface (point O) to beneath the outcrop. Point S indicates the point of intersection between the extended failure surface (O-S) and an assumed Jl joint (S-R). R represents the highest point in the profile and Q is location of the 1965 headscarp. In both cross sections, the basal configuration (O-S-R) outlines a possible failure surface. The presumed sliding surface (O-S) dips less than Jl (45°) and daylights on the slope. (O-S) inclines at 37° and 30° for C C and DD', respectively, while the surface topography below the headscarp (P-Q) dips at 45° and 37°, respectively. O-S is sub-parallel to the shallower steps in region P-Q. 6 see footnote 6 previously defined in chapter 4. 144 Stability analyses were performed on profiles C C and DD' using SG-Slope. For profile C C , both the cohesion and water table ratio were initially assumed to be zero. This gave a <p F = 1 of 39.5°, which is considerably higher than the expected range of cpg of 32° to 34° for a slip surface containing no gouge (sections 5.5 and 5.5.1). The analysis result suggests that C C would not be stable unless there is significant cohesion or a lack of throughgoing shallow joints. As discussed in section 5.3.2.3, cohesion strongly influences the stability largely because the failure block in C C is thin. If cohesion was increased to 50 kPa, <pg required for F = l was reduced to 30°. Under the assumption that cohesion and water table ratio were zero, analysis results for profile D D ' gave a <p F = 1 value of 31.5°. This suggests that the strength of the failure surface in D D ' is similar to or slightly stronger than that of the 1965 slope prior to the mass movement under the same assumed values for c and R. A </>g of 34° yields a factor of safety of 1.1. Since the locations of the two cross sections (CC and DD') are close to each other, one would expect similar results from the stability analyses under the same assumptions. The geometry of the sliding surface and the overall profile in C C are steeper than in DD* and represent a more critical orientation than DD'. In both cases (CC and DD'), the shallower surfaces facilitated sliding, rather than the Jl joints, which in effect faciliated dilational displacment. The surface and basal roughness (i.e. the alternating planes on the surface topography (P-Q-R) and the concavity on the failure surface (O-S-R)) may have "caused part of the sliding mass to be significantly more stable than an adjacent part" (Hoek, 1986; Sperling, 1991). The close proximity of the two profiles allows one to influence the other, thus the relative instability in C C is balanced by the relative stability in DD'. A large tension crack at approximately 1725 m elevation in profile DD1 (Fig.6.4) separates the outcrop there from the intact rock mass above. This tension crack was probably formed during the 1965 event. Its existence suggests that stresses were either dissipated through tensional displacement or were effectively transferred to a neighbouring 145 portion of the slope where excess resistance was available, buttressing the sliding mass and establishing a new equilibrium. Given the chosen profiles and assumptions, the results of these analyses suggest that C C , if not suported by adjacent stronger rock mass (e.g. DD'), its strength is worse than or at best approximates that of the pre-1965 slope. The profile DD' is relatively stronger. The volume of mass movement which could occur would probably be on the order 104 to K^m 3 . (2) THE UPPER NORTHWEST FLANK AREA, FROM THE TRENCH NORTH OF THE HEADSCARP APEX (1800 m) TO APPROXIMATELY 1600 m ELEVATION Between elevations 1805 m and 1750 m, extensive tension cracks are found in the upper northwest flank north of the headscarp apex (Figs.2.2, 3.11 and 6.1). Many of the cracks are open and trend sub-parallel to the orientation of J2 (85°/305°) or J3 (80°/018°) (see also section 3.2.3). Fig.6.5 illustrates the great circle representing the northwesterly dipping slope here and the great circles representing the pole concentration of the Jl and shallower (JS in Fig.6.5) joints. A <£g angle of 34° is used to represent the strength of the slope. Clearly, sliding can only occur in the directions between 280° and 315°, but would be limited by the buttress effects of the intact rock mass below. Toppling is the dominant mode of movement in this area. J2 and J3 joints dip near vertically allowing dilation or tensional displacement. Jl joints or any downslope dipping discontinuities would facilitate overturning or pivoting of columnar blocks bounded by J2 and J3. The toppling failure mechanism in association with trenches has been previously reported by Bovis (1982), Evans (personal communication, 1991) and Mollard (1977). Toppling failures in the upper northwest flank are probably a response to the removal of lateral support during the 1965 landslide. Here blocks of rocks tend to topple off the precipitous cliff onto the 1965 detachment surface to the south and southwest. 146 Using the stereographic method of Goodman and Bray (1976) and the joint data (494 points) in this area (outcrop number 5 in Appendix A-3), it was verified that toppling failure is kinematically feasible. Fig.6.7 illustrates that the concentration of J3 poles lies within the shaded zone where toppling could occur. The direction of toppling is roughly perpendicular to the precipitous headwall trending between 280° and 300°, which is averaged and represented by the great circle (80°/290°). The angle of internal friction, <£g, is assumed to be 34°. Further widening of the existing open cracks in this area will certainly induce toppling failures, which may occur either as a number of small topples or a large scale failure. The volume of loose material is limited by the size of the columnar blocks, and will probably be in the order of 103 m3. (3) THE LOWER NORTHWESTERN FLANK, BELOW ELEVATION OF APPROXIMATELY 1600 m There are a number of tectonic shear zones (type 2a and 2b, see section 3.2.2.2) in the mid to lower northwest flank (Figs.3.2 and 6.1, and Appendix A-3). Rockfalls originating here should be expected from time to time, since the exposed rocks (along the entire northern slide scar) are highly fractured, loose, and are subjected to frequent temperatures changes, freeze-thaw cycles and the wedging effects of ice. Based on a computer model, Piteau (1976) predicted that the probability of rockfalls from this area reaching the highway or the lookout is low. Ravelling of the rock observed during field work was relatively small in size, but larger rockfalls should be expected occasionally. If the felsite sheets observed along the lower northwest flank and in the middle exposures (Fig.2.2) extend to the north of the slide scar, and if the lithologic contacts are filled with clayey gouge, further shearing and displacement are possible. The incipient 147 landslide, located between 850 m and 1100 m (Fig. 1.3), reported by Mathews and McTaggart (1969) may be a manifestation of such instability. Further work is needed to determine the extent and nature of the discontinuities on the slopes adjacent to the slide scar, including the felsite contacts with greenstone and the faults. 148 LEGEND Fig.6.2 Profile C C parallel to 270° selected for hazard evaluation. 150 Fig.6.3 Profile DD' parallel to 270° selected for hazard evaluation. Fig.6.4 Oblique photograph of the headscarp apex (solid circle) area, showing the large tension crack (TC) at approximately 1725 m elevation separating the toe of profile DD' and the intact slope above. J=joints, t=trench, the stars illustrate the location of mapped tension cracks (Fig.3.11). (Photography by K.W. Savigny, 1989) 152 Fig.6.5 Stereoplot indicating the possible direction of sliding along Jl (280° to 315°) in the upper northwestern flank, but movement will be limited by the buttressing rock mass in lower elevations. 153 154 7 CONCLUSIONS AND SUGGESTIONS FOR FUTURE STUDIES The fame of the 1965 Hope Slide leads many people to believe that the rock avalanche is well understood, but only four papers have dealt directly with it, including only one field investigation (Mathews and McTaggart, 1969, 1978; Bruce and Cruden, 1977 and Wetmiller and Evans, 1989). The objective of this study was to expand on the existing knowledge and provide more focused directions for future detailed studies of the Hope Slide. The data from this thesis provide enhanced insight into some of the geotechnical and geological characteristics of the detachment zone. Definitive conclusions of why and how the 1965 slide event occurred cannot yet be drawn due to the complexity of the slide, and many of the parameters contributing to the failure are still unknown. The geology was first confirmed at the study site. It was found that there are two varieties of each of the felsite and greenstone. The felsite was differentiated mainly by colour. The buff felsite variety contains minerals that are indicative of a slightly higher grade of metamorphism than the greyish-white felsite. The greenstones differ mainly by texture: one occurs in a massive form and the other as a slightly schistose to foliated rock. Evidence of shearing and gouge was consistently found along the buff felsite and massive greenstone contacts; less frequent and relatively less intense shearing was present j along the contacts of the white felsite-- - - - - - • . • - - ^ greenstone. Three joint sets are dominant in the study area. The mean orientations are 45"7273° for set Jl, 85°/305° for set J2 and 807018° for set J3. A fourth set, which may be associated with Jl, has a similar trend as Jl but shallower dip angles as low as 20°. Its spatial and geometric relationship with Jl appears to result in a step-like topography. It is believed that the 1965 failure surface was governed by two mechanisms: the shallower lower portion of the detachment surface was controlled by gouge filled litho-tectonic contacts (between buff felsite and greenstone), and the steeper upper slope was 155 controlled by step-like discontinuities (Jl and the shallower joints), where the shallower joints, rather than Jl, facilitated sliding. The overall direction of the mass movement was influenced by not only the orientations of the discontinuities, but also pre-existing geomorphic features, which include the prehistoric slide scarp, a pre-existing chute and an adjacent rock shelf at lower elevations (below 1000 m), and a kame ridge at the foot of the slide. Geomorphic and physical evidence, based on aerial photograph interpretation, strongly suggest that the prehistoric and the 1965 slides failed in much the same way. One of the common features is that the headscarps of the prehistoric and 1965 slides both appear to have been limited by trenches, which trend mainly sub-parallel to the mountain ridge. Rock mass deformation also appears to be limited upslope by these trenches, above which the rock mass is relatively undisturbed. The relationship between the trenches and slope deformation is unclear. The apprent contiguous nature of these trenches with a moraine feature about 3 km southeast of the slide site raises the suspicion that they may be related with glacial processes, this relationship is currently being investigated by others. Stability analyses were performed to estimate the condition of the slope prior to the 1965 failure. The shear strength parameters considered were the friction angles of the greenstone (c/>g) and gouge (cpj, cohesion for the greenstone (cg) and gouge (cj, and the water table ratio (R). <pg is considered the most reliable indicator of the overall strength of the slope at the time of the slide, since it was estimated by direct shear tests. The other four parameters can no longer be measured or confirmed on the sliding surface, thus only inference of their values can be made. Static back analysis of the 1965 slope indicates that the minimum angle of friction (0g and p=i) was 31.2° assuming zero cohesion and a dry slope (R=0). Because of the many unknown parameters, stability analyses were performed to determine the different combinations of the parameters that would have give a slope of marginal stability. For assumed </>c=22°, R=0 to 0.50 and c=0 kPa, and a slope containing gouge along parts of 156 the sliding surface, the most likely values for cpg consistent with the occurrence of failure were 32° to 37°; if gouge was excluded, the most likely <pg values were 29° to 35°. For c=100 kPa, <f>g required for mobilization increased by 3°. Weichert et al. (1990) suggested that the Hope Slide may have occurred as two separate events, this hypothesis is supported by new field evidence that relate to the two failure mechanisms mentioned above. Stability analyses were further performed on the basis of two separate slide events. Results show that for given sets of parameters, the lower part would have failed even if the upper portion did not, and the upper slope was or subsequently became unstable under the same sets of parameters. Resulting values of <pg are consistent with the mobilized shear strength of the slope, and with the hypothesis that two landslides occurred. A pit wall collapse at the Brenda Mine in April, 1990 created a seismic signature equivalent to that of an earthquake in a magnitude of 2.3. Comparison of the potential energy released during the landslides and the transmitted seismic energy from the associated earthquakes was made for both the Hope Slide and Brenda Mine locations. In both cases, the seismic energy is at least nine orders of magnitude smaller than the released potential energy (Table 4.3). The comparison also shows that the efficiency in the transfer of the energy was substantially more efficient at the Hope Slide than at the Brenda Mine site in order to produce seismic shocks equivalent to those recorded on January 9, 1965. This suggests that the stability of the slope immediately before the landslide(s) was in a sufficiently critical stage that seismic loading was not necessary to induce failure. This research has only begun to touch on some of the aspects of the Hope Slide, it has provided improved insights on the rock mass fabric, the failure surface, ranges of shear strength of the greenstone and the content of the two different gouge materials. However, the causative factors of the 1965 rock avalanche are not resolved. Much more detailed and focused investigations should be undertaken, especially issues related to the following: 157 (1) Monitoring of the development of the tension cracks at the headscarp should be continued. Annual or semi-annual measurements should be carried out between the movement hubs installed in 1989. The positions of the hubs should also be established each time by detailed horizontal and vertical positioning surveys so that the absolute magnitude and direction of movements can be determined. (2) Investigation of the rock mass fabric in the adjacent slopes should be undertaken to correlate with those determined in this study and on a regional scale. This should include survey of the discontinuities and the nature and extent of occurrence of felsite sheets. One area of interest would be the incipient landslide reported by Mathews and McTaggart (1969). (3) To accomplish (2) above, topographic coverage needs to be expanded or updated for both the pre- and post-1965 topography, especially since the pre-1965 topography may be inaccurate. With new and accurate topographic maps, the problem of the volume calculations may be solved. (4) In view of the lack of information on chlorite-rich gouge, further analysis of the characteristics of the gouge content, particularly the lithologic gouge, and determination its shear strength are required. (5) If seismic signatures generated by landslides (such as the Brenda Mine case) can be positively identified, it may then be possible to determine the relationship between the 1965 event and the associated seismic shocks. (6) In-situ or laboratory direct shear testing of the discontinuities, particularly between greenstone and felsite contacts and the Jl joint surfaces, should be undertaken. (7) Identification of the relationship between the two types of greenstones and/or the two types of felsite should be persued. (8) Determination of the characteristics and effects of the apparent waviness in the greenstone also requires more study. REFERENCES APPENDICES REFERENCES Anderson, F.W. 1965. The Hope Slide Story. Frontiers, Calgary, Alberta. 46 p. Bacon, C.R. 1983. Eruptive History of Mount Mazama and Crater Lake Caldera, Cascade Range, U.S.A. Journal of Volcanology and Geothermal Research, 18:57-115. Bovis, M.J. 1982. Uphill-facing (antislope) scarps in the Coast Mountains, southwest British Columbia. Geological Society of America Bulletin, 93, pp.804-812. Brown, E.T. (Editor). 1981. Rock characterization, Testing and Monitoring. ISRM Suggested Methods. Pergamon Press Ltd. Oxford, England. Bruce, I. 1978. The field estimation of shear strength on rock discontinuities. Ph.D. Thesis, Dept. of Civil Engineering, University of Alberta, Edmonton, Alberta. 309 p. Bruce, I. and Cruden, D. 1977. The dynamics of the Hope Slide. Bulletin, International Association of Engineering Geology, No. 16, pp.94-98. — 1980. Simple rock slides at Jonas Ridge, Alberta, Canada. International Symposium on Landslides, New Delhi, 1, pp. 185-190. Cairnes, C.E. 1923. Geological explorations in Yale and Similkameen mining divisions, southwestern British Columbia. Geological Survey of Canada, Report 1922, Part A, pp. 88-126. Cairnes, C.E. 1924. Coquihalla area, British Columbia. Geological Survey of Canada, Memoir 139. Carrol, D. 1970. Clay Minerals: A Guide to theri X-Ray Identification. Geological Society of America, Special Paper 126, 70 p. Chowdhury, R.N. 1978. Slope Analysis. Developments in Geotechnical Engineering. Vol.22. Elsevier North-Holland Inc. New York, New York. 423 p. Clague, J.J. 1985. Radiocarbon Dates. Geological Survey of Canada. Laboratory No. GSC-4004. Coney, P.J., Jones, D.L. and Monger, J.W.H. 1980. Cordilleran suspect terranes. Nature, 288, pp. 329-333. Croasdale, D. A. 1989. A small scale translation^ rock failure affecting the basal failure surface of the 1965 Hope Slide. B.A.Sc. thesis. University of British Columbia, Vancouver, B.C. Cruden, D., Ramsden, J. and Herget, G. 1977. Pit Slope Manual, Supplement 2-1 - DISCODAT Program Package. CANMET (Canada Centre for Mineral and Energy Technology), CANMET Report 77-18, 61 p. Daly, R.A. 1912. Geology of the North American Cordillera at the 49th parallel. Geological Survey of Canada, Memoir 38. Dawson, G.M. 1879. Report on exploration in the southern British Columbia. Geological Survey of Canada, Report of Progress, 1877-1878, B: 1-173. Deere, D.U. and Miller, R.P. 1966. Engineering Classification and Index Properties for Intact Rocks. Technical Report No. AFWL-TR-65-116, Air Force Weapons Laboratory Research and Technology Division, Air Force Systems Command, Kirtland Air Force Base, New Mexico. Department of Transport, Meteorological Branch. (1964) Meteorological Observations in Canada. Monthly Records. Department of Transport, Meteorological Branch. (1965) Meteorological Observations in Canada. Monthly Records. Environment Canada. 1982. Atmospheric Environment Service. Canadian Climate Normals (1951-80). Temeperature and Precipitation. Gabrielse, H. and Yorath, C.J. (in press) The Cordilleran Orogen in Canada in Geology and Economic Minerals Deposits of Canada. Geological Survey of Canada, Ottawa. Goodman, R.E. 1970. Deformability of Joints, in Symposium on the determination of the In Situ modulus of deformation or rock. American Society for Testing Materials, S.P.T. 477, pp. 1974-196. — 1989. Introduction to Rock Mechancis. 2nd Edition. John Wiley & Sons, New York. Goodman, R.E. and Bray, J.W. 1976. Toppling of Rock Slopes. Proceedings of a Specialty Conference on Rock Engineering for Foundations and Slopes, Boulder, Colorado, n, pp.201-234. Grim, R.E. 1962. Applied Clay Mineralogy. McGraw-Hill Book Company, Inc., New York. — 1968. Clay Mineralogy. Second Edition. McGraw-Hill Book Company, Inc., New York. 596 p. Grivas, D.A. and Souflis, C. 1985. Seismic stability of natural slopes subject to progressive failure. Proceedings, 11th International Conference on Soils Mechanics and Foundation Engineering. San Francisco, n, pp. 1951-1954. Hasegawa, H.S., Basham, P.W. and Berry, M.J. 1981. Attenuation Relations for Strong Seismic Ground Motion in Canada. Bulletin of the Seismological Society of America, 71, pp. 1943-1962. Herget, G. 1977. Pit Slope Manual, Chapter 2 - Structural Geology. CANMET (Canada Centre for Mineral and Energy Technology), CANMET Report 77-41, 123 p. Hoek, E. 1986. General two-dimensional slope stability analysis in Analytical and Computational Methods in Engineering Rock Mechanics. Edited by E.T. Brown. Allen and Unwin, London. 259 p. Hoek, E. and Bray, J.W. 1977. Rock Slope Engineering. 2nd Edition, Institute of Mining and Metallurgy, London, England. 402 p. Horcoff, J. 1975. Tension Cracks, Hope Slide. British Columbia Department of Highways, Geotechnical and Materials Branch, Victoria, B.C. Unpublished Paper. Hsu, K.J. 1975. Catastrophic debris streams (Sturzstroms) generated by rockfalls. Geological Society of America. Bulletin. 8 6 , pp. 129-140. Hunt, R.E. 1986. Geotechnical Engineering Analysis and Evaluation. McGraw-Hill Book Company, New York. 729 p. Jahn, A. 1964. Slope morphological features resulting from gravitation. Zeitschrift fur Geomorphologie, Supplement Band, 5, pp.59-72. Jaeger, J.C. and Cook, N.G.W. 1971. Fundamentals of Rock Mechanics. Chapman and Hall Ltd, London. 515 p. Keefer, D.K. 1984. Landslides caused by earthquakes. Geological Society of America Bulletin, 95, pp.406-421. Kenney, T.C. 1967. The influence of mineral composition on the residual strength of natural soils. Proceedings, Geotechnical Conference, Oslo, 1:123-129. Kerrich, J.E. 1974. Notes on the orientation of sets of joints. Internal Report, University of Witwatersrand, 45pp. Kleinspehn, K.L. 1985. Cretaceous sedimentation and tectonics, Tyaughton-Methow Basin, southwestern British Columbia. Canadian Journal of Earth Sciences, 22, pp. 154-174. Lambe, T.W. 1951. Soil testing for Engineers. John Wiley & Sons, Inc. New York. 165 p. Lambe, T.W. and Whitman, R.V. 1968. Soil Mechanics. John Wiley & Sons, Inc. New York. 553 p. Mathewes, R.W. and Heusser, L.E. 1981. A 12 000 year palynological record of temperature and precipitation trends in southwestern British Columbia. Canada Journal of Botany, 59, pp. 707-710. Mathews, W.H. and McTaggart, K.C. 1969. The Hope Landslide, British Columbia. Proceedings, The Geological Association of Canada, 20, pp. 65-75. — 1978. Hope Rockslides, British Columbia, Canada, in Rockslides and Avalanches, I. Edited by B. Voight, Elsevier, New York, pp. 259-275. McTaggart, K.C. 1970. Tectonic history of the Northern Cascade Mountains in Structure of the southern Canadian Cordillera. Edited by J.O. Wheeler. Geological Association of Canada, Special Paper 6. pp. 137-148. McTaggart, K.C. and Thompson, R.M. 1967. Geology of part of the northern Cascades in southern British Columbia. Canadian Journal of Earth Sciences, 4, pp. 1199-1228. Mollard, J.D. 1976. Landforms and surface materials of Canada. A stereographic atlas and glossary. 5th Edition. J.R. Mollard and Associates, Ltd., Regina, Saskatchewan. 366 p. — 1977. Regional landslide types in Canada, in Reviews in Engineering Geology. Geological Society of America. HI, 272 p. Monger, J.W.H. 1970. Hope map-area, West half, British Columbia. Geological Survey of Canada, Paper 69-47. — 1986. Geology between Harrison Lake and Fraser River, Hope map area, southwestern British Columbia, in Current Research, Part B, Geological Survey of Canada, Paper 86-1B, pp. 699-706. — 1989a. Geology of Hope and Ashcroft map areas, British Columbia. Geological Survey of Canada, Maps 41-1989 and 42-1989. — 1989b. Overview of Cordilleran Geology in Western Canada Sedimentary Basin, A Case History, Edited by B. Ricketts, Canadian Society of Petroleum Geologists. Calgary. 1989. Monger, J.W.H., Price, R.A. and Tempelman-Kluit, D.J. 1982. Tectonic accretion and the origin of the two major metamorphic and plutonic welts in the Canadian Cordillera. Geology, 10, pp. 70-75. Monger, J.W.H., Souther, J.G. and Gabrielse, H. 1972. Evolution of the Canadian Cordillera: a plate-tectonic model. American Journal of Science, 272, pp. 577-602. Naumann, C M . 1990. A study of the interrelationship of rock avalanches and seismicity. M.A.Sc. Thesis, Dept. of Geological Sciences, University of British Columbia, Vancouver, B.C. Parrish, R.R. 1982. Cenozoic thermal and tectonic history of the Coast Mountains of British Columbia as revealed by fission track and geological data and quantitative models, unpublished Ph.D. Thesis, University of British Columbia, Vancouver, B.C. Patton, F.D. 1966. Multiple modes of shear failure in rocks. Proceedings, 1st International Congress of Rock Mechanics. Lisbon. 1, pp.509-513. Piteau, D.R. 1976. Hope Slide - Analysis of potential rockfall hazard. British Columbia Ministry of Transportation and Highways, Geotechnical and Materials Branch, Victoria, B.C. Unpublished Report. 163 — 1977. Regional slope stability controls and engineering geology of the Fraser Canyon, British Columbia. Geological Society of America Reviews in Engineering Geology. 3:85-111. Pollock, D. (personal communication) Prairie Farm Rehabilitation Administrtion, Saskatchewan. Radbruch-Hall, D.H. 1978. Gravitational creep of rock masses on slopes, in Rockslides and Avalanches, I. Edited by B. Voight. New York, Elsevier Scientific, pp.607-657. Richter, C F . 1958. Elementary Seismology. W.H. Freeman and Company, Inc. San Francisco, CA. 768 p. Ryder, J.M., Bovis, M.J. and Church, M. 1990. Rock avalanches at Texas Creek, British Columbia. 27, pp. 1316-1329. Sarma, S.K. 1973. Stability Analysis of Embankments and Slopes. Geotechnique, 23, No. 3, pp. 423-433. — 1979. Stability Analysis of Embankments and Slopes. Journal of the Geotechnical Engineering Division, American Society of Civil Engineers, 105, pp. 1511-1524. Savigny, K.W. 1990. Engineering geology of large landslides in the Lower Fraser River valley area, southwestern Canadian Cordillera. Abstract, Geological Association of Canada and Mineral Association of Canada, Joint Annual Meeting, Vancouver, B.C. — (in prep.) Engineering geology of large landslides in the Lower Fraser River valley area, southwestern British Columbia, in Landslide Hazard in the Canadian Cordillera. Geological Association of Canada. Special Paper. Seed, H.B. and Goodman, R.E. 1964. Earthquake Stability of Slopes of Cohesionless Soils. Proceedings of the American Society of Civil Engineers, Journal of the Soil Mechanics and Foundations Division. 90, pp.43-73. Shreve, R.L. 1968. The Blackhawk Landslide. Special Paper 108, The Geological Society of America, Inc. Boulder, Colorado. Skempton, A.W. 1953. The Colloidal "Activity" of Clay, Proceedings. Third International Conference on Soil Mechanics, I, pp.57-61. — 1985. Residual strength of clays in landslides, folded strata and the laboratory. Geotechnique, 35, pp.3-18. Sperling, T. 1991. Risk-cost-benefit framework for the design of dewatering systems in open pit mines. Ph.D. thesis, University of British Columbia, Vancouver, B.C. Streckeisen, A.L. 1976. To each plutonic rock its proper name. Earth Sciences Reviews, 12. Taylor, D.W. 1937. Stability of Earth Slopes. Journal of the Boston Society of Civil Engineers. 24, No. 3. von Sacken, R.S., Savigny, K.W. and Evans, S.G. 1989. Joint patterns at the headscarp area of the 1965 Hope Slide. Geological Survey of Canada. Open File OF2012. Weichert, D.H., Horner, R.B. and Evans, S.G. 1990. Earthquakes and the 1965 Hope Landslide? Abstract, American Geophysical Union's Pacific Northwest Regional Meeting, September 13-14, 1990, Seattle, Washington. Wetmiller, R.T. and Evans, S.G. 1989. Analysis of the earthquakes associated with the 1965 Hope landslide and their effects on slope stability at the site. Canadian Geotechnical Journal. 26, pp.484-490. White, W.H. 1959. Cordilleran tectonics in British Columbia. Bulletin of American Association of Petroleum Geologists, 43, no.l, pp. 60-100. Wilson, M.J. 1987. A Handbook of Determinative Methods in Clay Mineralogy. Chapman and Hall, New York, N.Y. 247 p. 165 APPENDICES A. TOPOGRAPHIC MAPS AND AN INDEX MAP (IN MAP POCKET) B. X-RAY DIFFRACTION ANALYSIS C. EVALUATION OF WEATHER CONDITIONS AT THE SLIDE SITE IN 1964/1965 D. SUMMARY OF STABILITY ANALYSES 166 APPENDIX A (IN MAP POCKET) APPENDED A-l: 1:2500 SCALE TOPOGRAPHIC MAP OF THE 1965 DETACHMENT SURFACE AND PART OF JOHNSON RIDGE (DOMAINS 1 AND 2) APPENDIX A-2: 1:1000 SCALE TOPOGRAPHIC MAP OF THE UPPER PART OF THE 1965 DETACHMENT SURFACE (DOMAIN 1 OF THE STUDY AREA) APPENDIX A-3: INDEX MAP OF THE STUDY AREA 167 APPENDIX B X-RAY DIFFRACTION ANALYSIS APPENDIX B -1 : DIFFRACTION PATTERN FOR UNTREATED SAMPLE HSCF APPENDIX B-2: DIFFRACTION PATTERN FOR HEATED SAMPLE HSCF APPENDIX B-3: DIFFRACTION PATTERN FOR GLYCOLATED SAMPLE HSCF APPENDIX B-4: DIFFRACTION PATTERN FOR HEATED AND GLYCOLATED SAMPLE HSCF APPENDIX B-5: DIFFRACTION PATTERN FOR GLYCOLATED SAMPLE HSCL 168 APPENDIX B X-RAY DIFFRACTION ANALYSIS In order to identify the clay minerals in the shear gouge samples, HSCF and HSCL, X-ray diffraction (XRD) techniques were used. Small amounts of the fraction that passed through the #200 mesh from the sieve analyses for each sample were mixed with water in a beaker. The dispersed samples were then stirred and allowed to partially settle. Minute quantities of the supernatant liquid containing suspended clays were pipetted onto glass slides, which were then air dried. Three slides for each sample were prepared, totalling 6 samples. For each of HSCF and HSCL, one slide remained untreated, one was heated to 550°C for one hour, and the third slide was suspended in a desiccator over ethylene glycol for 24 hours. The different treatments permit more positive identification of certain clay minerals. Minerals from the chlorite, mica, kaolinite and montmorillonite groups are easily distinguished from the change in XRD patterns under the different treatments of the sample as described above (Carrol, 1970; Grim, 1962, 1968; Horsky, S., Dept. of Geological Sciences, U.B.C., personal communication, 1990; Wilson, 1987). Typical basal spacings for various clay mineral groups are summarized as follow: Order 1st 2nd 3rd d(A) 20(°) d(A) 20(°) d(A) 20(°) Chlorite 14 6.3 7 12.6 4.7 18.9 Mica 10 8.8 5 17.7 3.3 27.0 Kaolinite 7 12.4 3.5 25.4 2.3 39.2 Montmorillonite 12-15 7.4-5.9 6-7.5 14.8-11.8 4-5 22.2-17.7 169 Peaks for chlorite minerals are easily distinguished, unless kaolinite is also present. However, under heat treatment, the structure in kaolinite is destroyed and the 1st order peak disappears; chlorite reflection usually increases in intensity and may be reduced from d = 14A to d=13.8A; the first order peak in montmorillonite also collapses to about d=9.5A. Only montmorillonite or mixed-layer montmorillonite is affected by the glycolation treatment. The 1st order spacing will be expanded to d=17A from d=12-15A. Sample HSCL was identified to contain mainly chlorite and fibrous actinolite. Sample HSCF, however, contains mixtures of chlorite, kaolinite, montmorillonite and actinolite. As described above, the 1st order peak of kaolinite disappeared in the heated sample. Montmorillonite was also positively identified in the glycolated sample, although the primary peak was relatively small and was masked by the chlorite peak. Based on the XRD patterns, it seems that the HSCL sample was nearly pure chlorite with minor fibrous actinolite. HSCF, on the other hand, contains kaolinite and montmorillonite in addition to chlorite and actinolite. It is likely that the clay has mixed-layer structures. Depending on the quantity of montmorillonite in a sample, it is generally considered a normal to active clay and can have residual strength (0r) as low as 10° (Kenney, 1967; Skempton, 1953). 170 M HSCFRftW 55 ; ii, 8108 : .68 CuKal+ 2 c C K 13 29 5.880 x : 2 t h e t a u : 611. Linear> •0506 C <Hg5(U><Si .aD4O18<0H>8 C i i n o o h l o j p e I I H .1 I k i -0135 Ca0.a(ftl,Mc[)3Si4O18(OH)2.xH2O M o n t H o r i l l o n i t e fl -1488 A12Si205<0R>4 Kaol in i te IT HA RG 29.990> ? h e l p Z O O M Natch F i l e C l e a r Back. N u l l K a2 Peaks SHOO,. COMV>. W f i l e -> H Appendix B-l: Diffraction patterns of untreated sample HSCF M = montmorillonite C = clinochlore K = kaolinite A = actinolite 171 H S C F H E A T s s : 8 . 9 1 8 0 t « : 1 . 6 8 Cu.Kal + 2 5 . 0 0 0 x : 2 t h e t a y 662. Lineap 29.990> 2 4 - 0 5 OS C ( H c f S f i l ) <Si . s l ) 4 0 J L 6 < 0 K ) S C ! i n o e h I o r 9 I I s 1 i B SC ; :.•>••«•«..'.!• 13-0135 C a 0 . 2 ( r l i .Mff)2Si4Oi0(OH)2^xH26 Mon^ fl 29-1488 A12Si205<0H>4 Kaol in i te IT Hi RC ?}ieli> Z O O M Hatch F i l e Clear Back. Null K a2 Peaks S M O O . C O M P . Hf i l e -> H Appendix B-2: Diffraction patterns of heated sample HSCF M = montmorillonite C = clinochlore A = actinolite 172 Mi H S C F G L V 1 s s : 0.0100 tr 1.00 CuKal+2 C 0 3.008 2 t h e t a y 619. L i n e a r 29.990> 24-0506 C < M g 5 A l H S i . ftl )4018<0H)S CI i n o c l i l o r e IT M I I h RC I C a 2 F e 5 ( S i 8 5 2 2 > (OrOJ! fex»r>o a c t i n o l i t e i eiwoan 13-0135 29-1488 0.2<fll,M«) Si4 10(OH)2.xH2O MontMori1lonite R A12Si205<0H>4 Kaolinite IT Hd RC ?help ZOOM Hatch File Clear Back. Null K a2 Peaks SMOO. COMP. Ufile -> H Appendix B-3: Diffraction patterns of glycolated sample HSCF M = montmorillonite C = clinochlore A = actinolite 173 HSCFH&G s s ! 0 .8108 t n : 1 . 0 0 CuKal+2 A < 3.880 24-0506 C 13-0135 * 2theta y : 653. Linear I I,h RG ferroan 11.990> <Mg5AlXSi,f i I>4Oi0<OH>8 Cl inochlore IT Ca0.2(fll,Mcf)2Si4O10(OH)2,xH2O MontHori1lonite fl ? h e l p Z O O M Match F i l e C l e a r Back . N u l l K a2 Peaks S H O O . C O M P . U f i l e -> M Appendix B-4: Diffraction patterns of heated and glycolated sample HSCF M = montmorillonite C = clinochlore A = actinolite HSCLCLV 5 s : 8 . 0 1 8 0 t « : 1 . 9 0 CuK*i+2 /III < 5 . 0 0 0 x : 2 t h e t a u : 1471. L i n e a r S4-03fg C (Hg3AlM5iJj])401ifl<OH>3.Clinochlore II M 1 I 29.990> ?help Z O O M Match File Clear Back. Null K a2 Peaks S M O O . C O M P . Wfile -> H Appendix B-5: Diffraction patterns of glycolated sample HSCL C = clinochlore A = actinolite 175 APPENDIX C EVALUATION OF 1964/1965 WEATHER CONDITIONS AT THE HOPE SLIDE LOCATION APPENDIX C-l: MONTHLY PRECIPITATION AVERAGES BASED ON 1979-1988 RECORDS AT THE HOPE SLIDE STATION (HS) APPENDIX C-2: MONTHLY TEMPERATURE AVERAGES BASED ON 1979-1988 RECORDS AT THE HOPE SLIDE STATION (HS) APPENDIX C-3: COMPARISON OF MONTHLY PRECIPITATION AND TEMPERATURE AVERAGES BASED ON 30 YEAR NORMALS (1951-1980) FOR THE HOPE AIPORT (HA) AND ALLISON PASS (AP) STATIONS, AND 10 YEAR AVERAGES (1979-1988) FOR THE HOPE SLIDE (HS) STATION APPENDIX C-4: ESTIMATES OF THE DAILY TEMPERATURES AND PRECIPITATION FOR THE PERIOD OF DEC.16, 1964-JAN.9, 1965 AT THE HOPE SLIDE LOCATION 176 APPENDIX C EVALUATION OF WEATHER CONDITIONS AT THE HOPE SLIDE SITE IN 1964/1965 At the time of the Hope Slide in 1965, the two closest weather stations in operation were the Hope Airport and Allison Pass stations. The Hope Airport station (HA) is located approximately 18 km northwest of the slide site, at an elevation of 39 m; the Allison Pass station (AP) is located at an elevation of 1341 m 32 km southeast of the slide. Neither is entirely representative of the slide site. In 1967, a new weather station (HS) was set up in Sunshine Valley near the base of the slide. However, after only five years of operation, it was closed for a year and then moved to a new location in 1973 (Lat. 49°17'N., Long. 121 °14'W, elevation 674 m). Temperature and precipitation records from this station for the period between 1979 and 1988 were collected, and short term averages were determined (Department of Transport, Meteorological Branch, 1979-1988). They were then compared to the thirty year (1951-1980) normal values (Environment Canada, 1982) of the Hope Airport and Allison Pass stations in order to establish a better representation of the weather conditions at the base of slide. As expected, the Hope Slide site in general receives more rain and less snow than Allison Pass, and less rain and more snow than Hope Airport; and temperatures at the slide are in between those at Hope Airport and Allison Pass. Comparisons of the records show that the overall temperatures at the Hope Slide are approximately the average of those at HA and AP with small variations; and the total annual precipitation at HS is approximately 65% of that at HA and 75% of that at AP; 80% of the total annual precipitation fall as rain, the remainder as snow. Based on short term averages then, the weather station at the Hope Slide receives a total annual precipitation of 1160 mm, 931 mm of which are rainfall and 241 cm fall as 177 snow (Appendix C-3). The difference between the sum of rainfall and snowfall, and the total annual precipitation is due to occasional missing monthly records between 1979 and 1988 (see Appendices C-l and C-2). Using the 30 year mean values from the HA and AP stations and the following formula: P H S = (0.65 PHA + 0.75 P^) 2 [B. 1] where P = precipitation (mm), the HS location would have received an estimated 1130 mm of total annual precipitation in 1964/5, of which 904 mm (80%) fell as rain and 226 cm (20%) as snow. Specifically for the months of December and January, the slide site (HS) receives about 55 % of the monthly precipitation of that from either Hope Airport or Allison Pass. Hence, to estimate the daily precipitation for HS for the period of Dec. 16, 1964 to Jan.9, 1965 (or 25 days before the occurrence of the mass movement), the following formula was used: PHS = (PHA + PAP) * 0.55 + 2 [B.2] Rainfall at HS constitutes approximately 70% of the total monthly precipitation in December and 60% of that in January. The daily (maximum, minimum and mean) temperatures for the same period of time were calculated by taking the average of those from HS and AP: T H S = (T H A + TAP) - 2 [B.3] where T = temperature (°C) It should be noted that precipitation records in the 1960's were not separated into rainfall and snowfall. Using equations [B.2 and B.3] and the 30 year normal values from HA and Ap, for the month of December, 1964, HS received a total monthly precipitation of 156 mm, of 178 which 109.2 mm fell as rain, the rest as snow. Mean December maximum and minimum daily temperatures were estimated at 0.6°C and -2.5°C, respectively. Mean daily temperature for December was -5.5°C. Similarly for January, 1965, the total montly precipitation was 142.9 mm, of which 85.7 mm were rainfall.. Temperatures varied from a mean daily maximum of -0.9°C to a mean daily minimum of -4.2°C, with a mean daily temperature of -7.4°C. Appendix C-4 summarizes the thirty year mean monthly values for HA and AP and the ten year mean monthly values for HS. APPENDIX C-l MONTHLY PRECIPITATION AVERAGES BASED ON 1979-1988 WEATHER RECORDS AT THE HOPE SLIDE STATION (Lat. 49°17'N, Long. 121°14'W. Elev.674 m) YEAR JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC TOTAL 1979 (R)1 12.4 98.4 47.8 35.8 31.0 56.0 14.2 27.6 58.4 76.1 16.0 333.0 806.7 (S)2 62.4 59.8 22.8 11.5 0.0 0.0 0.0 0.0 0.0 T* 16.0 46.0 218.5 1980 (R) 17.2 60.0 34.2 47.9 94.7 67.4 26.1 53.4 111.4 15.2 232.4 249.8 1009.7 (S) 46.0 9.0 69.0 T 0.0 0.0 0.0 0.0 0.0 0.0 21.0 31.8 176.8 1981 (R) 11.8 179.8 34.0 80.6 71.3 137.4 66.6 29.0 56.4 M** 104.4 M 771.3 (S) 4.8 5.0 T 9.2 0.0 0.0 0.0 0.0 0.0 0.0 14.0 M 33.0 1982 (R) 121.4 143.4 6.0 36.7 47.0 43.4 98.9 66.1 M 123.1 51.8 M 737.8 (S) 170.8 15.0 59.0 22.0 T 0.0 0.0 0.0 0.0 T 29.4 M 296.2 1983 (R) 232.6 71.4 61.7 45.8 33.4 105.2 129.6 40.0 53.6 73.6 168.7 16.6 1032.2 (S) 61.0 8.0 1.0 3.0 0.0 0.0 0.0 0.0 0.0 0.0 8.0 64.0 145.0 1984 (R) 255.4 M 68.2 54.8 129.4 83.0 29.3 68.5 79.0 115.2 144.6 28.2 1055.6 (S) 4.8 M 5.0 4.2 1.8 0.0 0.0 0.0 0.0 39.6 81.0 M 136.4 1985 (R) 23.8 42.5 16.4 113.6 M 43.2 6.8 30.2 97.1 247.8 81.2 3.5 706.1 (S) 13.0 71.9 53.6 10.2 3.0 0.0 0.0 0.0 0.0 20.4 40.8 32.1 245.0 APPENDIX C-l (continued) MONTHLY PRECIPITATION AVERAGES BASED ON 1979-1988 WEATHER RECORDS AT THE HOPE SLIDE STATION (Lat. 49°17'N, Long. 121° 14*W. Elev.674 m) YEAR JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC TOTAL 1986 (R) 135.1 123.8 104.8 60.9 73.1 38.2 90.2 5.0 78.4 M 235.6 M 945.1 (S) 26.5 114.0 3.0 16.4 3.0 0.0 0.0 0.0 0.0 0.0 53.8 M 216.7 1987 (R) 49.2 12.7 92.7 101.4 74.8 28.2 105.4 10.8 38.5 20.4 49.4 77.4 660.9 (S) 105.0 32.0 15.3 19.0 0.0 0.0 0.0 0.0 0.0 0.0 24.6 95.0 290.9 1988 (R) 28.2 41.8 85.2 54.8 64.8 47.4 38.6 16.7 65.0 115.6 169.7 92.2 820.0 (S) 65.0 39.0 89.0 62.0 0.4 0.0 0.0 0.0 0.0 0.0 45.9 39.2 340.5 TOTAL 1446.41127.5 868.7 789.8 627.7 649.4 605.7 347.3 637.8 847.0 1588.31108.810644.4 MEAN 144.64125.3 86.9 79.0 69.7 64.9 60.6 34.7 70.9 105.9 158.8 158.4 MEAN (R) 88.7 86.0 55.1 63.2 68.8 64.9 60.6 34.7 70.9 98.4 125.4 114.4 MEAN (S) 55.9 39.3 35.3 17.5 0.9 0.0 0.0 0.0 0.0 7.5 33.4 51.4 1 R = rainfall in mm 2 S = snowfall in cm * T = trace M = missing record 10 year total precipitation (rainfall-I-snowfall) = 10644.4 mm Mean annual precipitation (ranifall+snowfall) = 1172.3 mm Mean annual rainfall = 931.1 mm Mean annual snowfall = 241.2 cm APPENDIX C-2 MONTHLY TEMPERATURE AVERAGES BASED ON 1979-1988 WEATHER RECORDS AT THE HOPE SLIDE STATION (Lat. 49°17'N, Long. 121°14'W. EIev.674 m)  YEAR JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC 1979 MAX* -4.5 1.2 7.2 9.4 13.8 16.3 21.4 21.0 18.4 12.7 5.4 2.8 MIN -15.9 -8.3 -4.3 -2.1 2.2 3.8 7.0 7.0 4.8 0.4 -4.7 -2.4 MEAN-10.2 -3.6 1.5 3.7 8 10.1 14.2 14.1 11.6 6.6 0.4 0.2 1980 MAX -2.4 4.2 4.2 12.6 12.9 14.6 19.5 18.6 15.5 13.5 5.7 1.4 MIN -10.7 3.5 -2.7 1.2 4.6 6.2 8.4 7.4 5.6 1.3 -0.3 -3.9 MEAN-6.6 0.4 0.8 6.9 8.8 10.4 14.0 13.0 10.6 7.4 2.7 -1.3 1981 MAX 5.0 4.9 9.5 9.0 12.9 13.3 19.2 24.4 18.6 11.2 7.3 0.5 MIN -1.4 -3.3 -1.2 0.7 4.8 5.3 9.0 10.4 6.0 0.9 0.4 -4.9 MEAN 1.8 0.8 4.0 4.9 8.9 9.3 14.1 17.4 12.3 6.1 3.9 -2.2 1982 MAX -1.1 1.8 6.0 9.5 14.0 21.2 18.9 20.4 18.4 12.3 3.1 1.9 MIN -7.3 -5.5 -3.7 -1.6 2.2 8.8 9.2 8.4 6.8 1.8 -4.8 -3.9 MEAN-4.2 -1.9 1.2 4.0 8.1 15.0 14.1 14.4 12.6 7.1 -0.9 -1.0 1983 MAX 3.8 6.1 9.7 13.6 17.2 16.3 18.1 21.6 14.5 12.0 5.3 -1.3 MIN -1.8 -2.1 -0.6 0.4 4.9 7.4 8.2 9.3 4.3 0.8 0.5 -9.2 MEAN 1.0 2.0 4.6 7.0 11.1 11.9 13.2 15.5 9.4 6.4 2.9 -5.3 APPENDIX C-2 (continued) MONTHLY TEMPERATURE AVERAGES BASED ON 1979-1988 WEATHER RECORDS AT THE HOPE SLIDE STATION (Lat. 49°17'N, Long. 121°14'W. Elev.674 m)  YEAR JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC 1984 MAX 2.8 5.4 10.1 9.6 10.7 16.2 22.1 20.1 14.7 7.7 3.4 -3.3 MIN -2.8 -1.7 0.4 0.4 2.1 7.2 8.0 8.7 4.0 -1.2 -2.4 -10.1 MEANO.O 1.9 5.3 5.0 6.4 11.7 15.1 14.4 9.4 3.3 0.5 -6.7 1985 MAX 1.5 1.6 6.6 9.6 16.1 18.4 25.7 21.3 15.1 8.9 -3.8 0.8 MIN -9.0 -6.2 -3.6 0.5 3.7 6.1 9.5 8.0 4.1 1.6 -10.9 -6.9 MEAN:3.8 2.3 1.5 5.1 9.9 12.3 17.6 14.7 9.6 5.3 -7.4 -3.1 1986 MAX 4.1 2.4 9.1 8.5 15.2 19.9 17.5 24.9 15.1 15.8 3.8 2.0 MIN -1.9 -6.3 -0.4 0.7 4.3 7.6 8.1 9.8 5.3 1.6 -2.5 -5.7 MEAN 1.1 -2.0 4.4 4.6 9.8 13.8 12.8 17.4 10.2 8.7 0.7 -1.9 1987 MAX 1.6 4.5 8.1 13.0 16.2 20.1 19.5 22.2 19.6 14.6 7.0 0.6 MIN -5.0 -2.4 -0.7 2.1 4.0 7.5 9.4 8.6 6.5 0.9 0.4 -6.0 MEAN-1.7 1.1 3.7 7.6 10.1 13.8 14.5 15.4 13.4 7.8 3.7 -2.7 1988 MAX -0.5 3.7 5.8 10.7 15.3 17.4 20.9 21.1 18.2 14.1 4.5 1.8 MIN -8.3 -4.2 -1.3 0.8 4.1 7.1 . 8.2 8.8 5.1 3.5 0.0 -3.4 MEAN-4.4 -0.3 2.3 5.8 9.7 12.3 14.6 15.0 11.7 8.8 2.3 -0.8 APPENDIX C-2 (continued) MONTHLY TEMPERATURE AVERAGES BASED ON 1979-1988 WEATHER RECORDS AT THE HOPE SLIDE STATION (Lat. 49°17'N, Long. 121° 14'W. Elev.674 m)  MEAN JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC MAX 1.0 3.6 7.6 10.6 14.4 17.4 20.3 21.6 16.8 12.4 4.2 0.7 MIN -6.4 -3.7 -1.8 0.3 3.7 6.7 8.5 8.7 5.3 1.2 -2.4 -5.6 MONTHLY -2.7 0.1 2.9 5.5 9.1 12.1 14.4 15.1 11.1 6.8 0.9 -2.3 * All temperatures are given in degree Celcius (°C) APPENDIX C-3 MONTHLY AVERAGES BASED ON 1979-1988 WEATHER RECORDS FOR THE HOPE SLIDE STATION (HS) AND 30 YEAR NORMAL VALUES (1951-1980) FOR THE HOPE AIRPORT (HA) AND ALLISON PASS (AP) STATIONS TOTAL PRECIPITATION (mm) JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC TOTAL HOPE AIRPORT 256.7 195.6 147.3 104.8 71.6 64.7 37.0 50.3 102.9 171.7 223.8 289.4 1715.8 HOPE SLIDE 144.6 125.3 86.9 79.0 69.7 64.9 60.6 34.8 70.9 105.9 158.9 158.4 1159.9 ALLISON PASS 262.9 174.7 143.6 85.5 73.7 74.1 30.2 45.9 64.3 103.1 188.7 277.8 1524.5 MEAN RAINFALL (mm) JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC TOTAL HOPE AIRPORT 184.1 165.2 131.2 103.3 71.6 64.7 37.0 50.3 102.9 171.6 208.5 249.1 1539.5 HOPE SLIDE 88.4 86.0 55.1 63.2 68.8 64.9 60.6 34.7 70.9 98.4 125.4 114.4 931.1 ALLISON PASS 28.1 23.5 9.9 16.9 44.9 73.6 30.1 45.9 61.3 61.9 44.0 32.9 473.0 MEAN SNOWFALL (cm) JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC TOTAL HOPE AIRPORT 81.7 31.3 15.6 1.4 T 0.0 0.0 0.0 0.0 0.1 16.7 45.8 192.6 HOPE SLIDE 55.9 39.3 35.3 17.5 0.9 0.0 0.0 0.0 0.0 7.5 33.5 51.4 241.3 ALLISON PASS 227.2 200.7 98.0 215.1 24.3 0.3 0.2 0.0 3.3 358.3 121.2 182.9 1431.5 APPENDIX C-3 (continued) MONTHLY AVERAGES BASED ON 1979-1988 WEATHER RECORDS FOR THE HOPE SLIDE STATION (HS) AND 30 YEAR NORMAL VALUES (1951-1980) FOR THE HOPE AIRPORT (HA) AND ALLISON PASS (AP) STATIONS MEAN DAILY TEMPERATURE (°C) JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC HOPE AIRPORT -0.4 3.4 5.6 9.3 13.0 15.8 18.5 18.4 15.5 10.47 4.7 1.6 HOPE SLIDE -2.7 0.1 2.9 5.5 9.1 12.1 14.4 15.1 11.1 6.8 0.9 -2.5 ALLISON PASS -7.9 -5.4 -3.5 0.8 4.5 8.3 12.1 11.7 8.3 3.0 -3.4 -6.5 MEAN DAILY MAXIMUM TEMPERATURE (°C) JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC HOPE AIRPORT 2.3 6.9 10.1 14.6 18.6 21.0 24.4 24.1 21.1 14.8 7.6 4.1 HOPE SLIDE 1.0 3.6 7.6 10.6 14.4 17.4 20.3 21.6 16.8 23.4 4.2 0.7 ALLISON PASS -4.1 -0.9 1.9 6.4 10.5 14.8 19.8 19.2 15.0 8.2 0.3 -2.9 MEAN DAILY MINIMUM TEMPERATURE (°C) JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC HOPE AIRPORT -3.1 -0.2 1.0 3.9 7.4 10.6 12.5 12.6 9.9 6.0 1.7 -0.9 HOPE SLIDE -6.4 -3.7 -1.8 0.3 3.7 6.7 8.5 8.7 5.3 1.2 -2.4 -5.6 ALLISON PASS -11.7 -9.8 -8.8 -4.7 -1.5 1.6 4.3 4.2 1.5 -2.2 -7.0 -10.0 APPENDIX C-4 DAILY TEMPERATURE AND PRECIPITATION RECORD FROM DEC.16, 1964 TO JAN.9, 1965 INCLUSIVE HOPE SLIDE VALUES BASED ON INTERPRETATION AND INTERPOLATIONS ALLISON PASS (AP) HOPE AIRPORT (HA) HOPE SLIDE (HS) TEMPERATURE1 TEMPERATURE TEMPERATURE PRECIPITATION (mm) 1964 MAX MIN MAX MIN MAX MIN AP HA HS Dec. 16 -5.0 -37.2 -15.0 -20.6 -10.0 -28.9 0.0 0.0 0.0 17 -22.8 -32.8 -11.7 -17.8 -17.3 -25.3 0.3 0.5 0.2 18 -16.1 -21.7 -7.2 -12.8 -11.7 -17.3 0.0 2.3 0.6 19 -11.7 -18.9 -3.3 -7.2 -7.5 -13.1 2.3 2.3 1.3 20 -10.6 -17.8 -6.1 -7.8 -8.4 -12.8 10.4 2.5 3.5 21 -9.4 -11.1 -7.2 -9.4 -8.3 -10.3 11.4 8.6 5.5 22 -7.2 -13.9 -6.7 -9.4 -7.0 -11.7 15.0 30.5 12.5 23 -13.3 -16.1 -5.0 -8.9 -9.2 -12.5 3.8 0.3 1.1 24 -10.6 -13.9 -7.8 -10.0 -9.2 -12.0 5.6 0.5 1.7 25 -12.8 -24.4 -8.3 -15.6 -10.6 -20.0 4.8 0.0 1.3 26 -3.3 -12.8 -6.1 -10.0 -4.7 -11.4 12.2 8.9 5.8 27 -5.6 -13.9 -2.2 -6.7 -3.9 -10.3 14.7 7.1 6.0 28 -11.1 -20.0 -1.1 -8.3 -6.1 -14.2 0.0 0.0 0.0 APPENDIX C-4 (continued) DAILY TEMPERATURE AND PRECIPITATION RECORD FROM DEC.16, 1964 TO JAN.9, 1965 INCLUSIVE HOPE SLIDE VALUES BASED ON INTERPRETATION AND INTERPOLATIONS ALLISON PASS (AP) HOPE AIRPORT (HA) HOPE SLIDE (HS) TEMPERATURE1 TEMPERATURE TEMPERATURE PRECIPITATION (mm) 1964 MAX MIN MAX MIN MAX MIN AP HA HS Dec.29 -12.2 -20.6 -3.3 -9.4 -7.8 -15.0 0.3 0.0 0.1 30 -11.1 -22.2 -2.8 -10.0 -7.0 -16.1 0.0 2.3 0.6 31 -8.9 -21.7 -4.4 -8.9 -6.7 -15.3 7.9 5.1 3.6 1965 Jan. 1 -4.4 -5.6 3.3 -6.1 -0.6 -5.9 56.6 26.7 22.9 2 -3.9 -7.2 2.8 -0.6 -0.6 -3.9 20.3 19.3 10.9 3 -6.7 -18.9 1.7 -11.7 -2.5 -15.3 5.6 5.6 3.1 4 -7.2 -17.2 -11.1 -12.8 -9.2 -15.0 1.0 0.5 0.4 5 -2.8 -14.4 -10.1 -12.8 -6.4 -13.6 14.7 1.3 4.4 6 -4.4 -13.3 -8.3 -12.2 -6.4 -12.8 10.2 3.3 3.7 7 -6.7 -9.4 -8.9 -10.6 -7.8 -10.0 11.2 9.4 5.7 8 -2.8 -7.8 2.8 -9.4 0.0 -8.6 4.1 14.0 4.2 9 -2.2 -11.7 2.8 0.0 0.3 -5.9 0.0 1.5 0.4 All temperatures are given in degree Celcius (°C) APPENDIX D SUMMARY OF STABILITY ANALYSES APPENDIX D-l: SCENARIO 5 APPENDLX D-2: SCENARIO 6 APPENDIX D-3: SCENARIO 7 APPENDIX D-4: SCENARIO 8 189 APPENDIX D-l SG-Slope Analyses results from Scenario 5 Profile BB', with ^ = 2 2 ° and cc=0 kPa, gouge on slice bases 2,4,6 and side 7 Run no. R (iPa) F Kc a (cm/s2) M L Rh(km) C3-1 0 34 0 0.969 -0.017 C3-1-1 0 32 0 0.906 -0.051 C3-1-2 0 36 0 1.035 0.019 18.62 3.8 18 4.5 35 0 37 0 1.069 0.038 0 40 0 1.177 0.101 C3-1-3 0.25 34 0 0.937 -0.034 C3-1-4 0.25 32 0 0.876 -0.067 C3-1-5 0.25 36 0 1.001 0.001 0.98 1.5 18 2.3 35 0.25 37 0 1.035 0.019 0.25 40 0 1.140 0.08 C3-1-6 0.5 34 0 0.904 -0.051 C3-1-7 0.5 32 0 0.845 -0.083 C3-1-8 0.5 36 0 0.966 -0.018 0.5 37 0 0.999 -0.001 0.5 40 0 1.101 0.056 C3-2 0 34 100 1.063 0.034 C3-2-1 0 32 100 1.000 0.00 0 C3-2-2 0 36 100 1.128 0.07 0 37 100 1.162 0.09 0 40 100 1.128 0.156 C3-2-3 0.25 34 100 1.032 0.017 190 APPENDIX D-l (continued) Run no. R (fPa) F a (cm/s2) M L Rh(km) C3-2-4 0.25 32 100 0.971 -0.016 C3-2-5 0.25 36 100 1.096 0.052 0.25 37 100 1.129 0.071 0.25 40 100 1.234 0.134 C3-2-6 0.5 34 100 0.999 0.00 C3-2-7 0.5 32 100 0.940 -0.032 C3-2-8 0.5 36 100 1.061 0.033 32.34 4.2 18 5.0 35 0.5 37 100 1.094 0.051 0.5 40 100 1.196 0.111 191 APPENDIX D-2 SG-Slope Analyses results from Scenario 6 profile BB', with <f>c=22° and cc=0 kPa, gouge on fault outcrop only Run no. R h (ipa) F Kc a (cm/s2) M L Rh(km) C3-3 0 32 0 0.971 -0.015 C3-3-1 0 34 0 1.049 0.025 24.5 4 18 4.8 35 C3-3-2 0 36 0 1.13 0.066 C3-3-3 0.25 32 0 0.936 -0.033 C3-3-4 0.25 34 0 1.011 0.005 4.9 2.8 18 3.6 35 C3-3-5 0.25 36 0 1.089 0.045 C3-3-6 0.5 32 0 0.900 -0.052 C3-3-7 0.5 34 0 0.971 -0.015 C3-3-8 0.5 36 0 1.046 0.023 C3-3-9 0 32 100 1.106 0.055 C3-3-10 0 34 100 1.183 0.094 G3-3-11 0 36 100 1.264 0.134 C3-3-12 0.25 32 100 1.071 0.037 C3-3-13 0.25 34 100 1.146 0.074 3-3-14 0.25 36 100 1.224 0.113 C3-3-15 0.5 32 100 1.035 0.018 17.64 3.8 18 4.5 35 C3-3-16 0.5 34 100 1.107 0.054 C3-3-17 0.5 34 100 1.182 0.091 192 APPENDIX D-3 SG-Slope Analyses results from Scenario 7, Profile BB' (lower six slices only of Scenario 5) with <f>c=22° and cc=0 kPa, gouge on slice bases 2,4,6 and side 7 Run no. R fr (iPa) F Kf. a (cm/s2) M L R h (km) C2LOWER10 34 0 1.059 0.032 C2LOWER40 32 0 1.005 0.003 C2LOWER3 0 36 0 1.116 0.066 C2LOWER20 35 0 1.087 0.048 0 37 0 1.146 0.085 C2LOWER50.25 34 0 0.961 -0.022 C2LOWER80.25 32 0 0.909 -0.050 C2LOWER70.25 36 0 1.017 0.010 C2LOWER60.25 35 0 0.988 -0.007 0.25 37 0 1.046 0.028 C2LOWER90.5 34 0 0.863 -0.082 C2LOW12 0.5 32 0 0.812 -0.036 C2LOW11 0.5 36 0 0.918 -0.053 C2LOW10 0.5 35 0 0.890 -0.068 0.5 37 0 0.946 -0.036 C2LOW13 0 34 100 1.172 0.094 C2LOW16 0 32 100 1.119 0.063 C2LOW15 0 36 100 1.228 0.131 C2LOW14 0 35 100 1.199 0.112 0 37 100 1.257 0.152 C2LOW17 0.25 34 100 1.071 0.040 C2LOW20 0.25 32 100 1.020 0.011 APPENDIX D-3 (continued) Run no. R (?Pa) F K j . a (cm/s2) M L R h (km) C2LOW19 0.25 36 100 1.125 0.075 C2LOW18 0.25 35 100 1.097 0.057 0.25 37 100 1.153 0.094 C2LOW21 0.5 34 100 0.967 -0.020 C2LOW24 0.5 32 100 0.918 -0.047 C2LOW23 0.5 36 100 1.019 0.012 C2LOW22 0.5 35 100 0.993 -0.005 0.5 37 100 1.047 0.031 194 APPENDIX D-4 SG-Slope Analyses results from Scenario 8 Profile BB' (top three slices only of Scenario 5) with <f>c=22° and cc=0 kPa, gouge on fault outcrop Run no. R ciPa) F Kc a (cm/s2) M L R h (km) C2TOP1 0 34 0 0.9050-0.049 C2TOP4 0 32 0 0.838 -0.085 C2TOP3 0 36 0 0.975 -0.013 C2TOP2 0 35 0 0.940 -0.031 0 37 0 1.011 0.005 C2TOP5 0.25 34 0 0.831 -0.086 C2TOP8 0.25 32 0 0.77 -0.120 C2TOP7 0.25 36 0 0.895 -0.052 C2TOP6 0.25 35 0 0.863 -0.069 0.25 37 0 0.929 -0.035 C2TOP9 0.5 34 0 0.731 -0.137 C2TOP12 0.5 32 0 0.676 -0.168 C2TOP11 0.5 36 0 0.786 -0.106 C2TOP10 0.5 35 0 0.757 -0.121 0.5 37 0 0.676 -0.909 C2TOP13 0 34 100 0.990 -0.005 C2TOP16 0 32 100 0.923 -0.04 C2TOP15 0 36 100 1.060 0.03 C2TOP14 0 35 100 1.024 0.012 0 37 100 1.096 0.048 C2TOP17 0.25 34 100 0.917 -0.042 C2TOP20 0.25 32 100 0.856 -0.075 APPENDIX D-4 (continued) Run no. R fr (fPa) F K c a (cm/s2) M L R h (km) C2TOP19 0,25 36 100 0.981 -0.009 C2TOP18 0.25 35 100 0.949 -0.026 0.25 37 100 1.014 0.007 C2TOP21 0.5 34 100 0.816 -0.093 C2TOP24 0.5 32 100 0.763 -0.124 C2TOP23 0.5 36 100 0.873 -0.063 C2TOP22 0.5 35 100 0.844 -0.078 0.5 37 100 0.902 -0.0478 1 /'.I*?/* *\* * *'?$s£0> u buff felsite i , f c ^ white felsite < P-^^:X;^j:^r~ . ,« ' * shear zones .._ *.$4 *«r"*' ' yk'XT^-yt^ ""' - «*»-»—*•—. (defined,approximated,assumed) \ 11800}'^J&&r*x"^*^sL*-** * ^ • ^ ' ^ ^ (defined, approximated, assumed) LEGEND + j greenstone (Hozameen Complex) granodiorite (Mount Outram Pluton) scree cones buff felsite it  f l it  quartz vein seepage location —••• lithological contact wavy greenstone approximate axis of colluvium outcrop label by number APPENDIX A-3: Index map of the Hope Slide detachment surface and ridge area, showing s i t e geology, s t r u c t u r a l and morphological features, and the approximate p r e h i s t o r i c s l i d e l i m i t . 

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.831.1-0052603/manifest

Comment

Related Items