UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Seismic response of the George Massey Tunnel Puar, Surinder S. 1996

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

Item Metadata

Download

Media
831-ubc_1996-0270.pdf [ 36.94MB ]
Metadata
JSON: 831-1.0050338.json
JSON-LD: 831-1.0050338-ld.json
RDF/XML (Pretty): 831-1.0050338-rdf.xml
RDF/JSON: 831-1.0050338-rdf.json
Turtle: 831-1.0050338-turtle.txt
N-Triples: 831-1.0050338-rdf-ntriples.txt
Original Record: 831-1.0050338-source.json
Full Text
831-1.0050338-fulltext.txt
Citation
831-1.0050338.ris

Full Text

SEISMIC RESPONSE OF THE GEORGE MASSEY TUNNEL by SURINDERS.PUAR B.A.Sc , The University of British Columbia, 1992 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF: MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Civil Engineering ~ Geotechnical Engineering Programme) We accept this thesis as conforming to the^^quired standard THE UNIVERSITY OF BRITISH COLUMBIA April 1996 © Surinder Singh Puar, 1996 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 of C/Vi Bv|Weri?A' The University of British Columbia Vancouver, Canada Date Ar i^i- 2H, iw DE-6 (2/88) ABSTRACT The George Massey Tunnel , i n Richmond, Br i t i sh Co lumbia , is a 630-meter long submerged concrete tunnel, w i th 550-meter and 335-meter long approaches on the north and south ends, respectively. The tunnel crosses the Fraser R i v e r and is founded on a deep deposit o f unconsolidated sediments consisting main ly o f sands and silts that are susceptible to l iquefaction during earthquake loading. This thesis represents a comprehensive analytical investigation to evaluate the l iquefaction potential o f the foundation soils and the performance o f the tunnel during a major earthquake. The evaluation procedures and post-liquefaction stability and deformation results are presented. Liquefact ion potential analyses based on the total stress approach were conducted. Liquefact ion was predicted by comparing the earthquake-induced stresses to so i l resistance. Dynamic ground response analyses were performed to assess the magnitude o f the cyc l i c stresses; the cyc l i c resistance o f the soi l was computed using various methods, depending on the so i l type. Estimated acceleration levels could potentially trigger l iquefaction i n substantial zones o f the tunnel's foundation. The residual (peak post-liquefaction) shear strength o f l iquefied soils was estimated to be adequate to maintain post-earthquake stability o f the tunnel at a l l o f the locations analyzed. The main problem to be addressed, therefore, was ii Abstract the displacements due to triggering o f liquefaction i n directions transverse to- and parallel to the tunnel alignment, as a result o f the 475-year seismic event. Post-liquefaction deformations o f the tunnel were computed using both empir ical and numerical methods. The numerical methods incorporate post-liquefaction stress-strain relationships and account for the effects o f both gravity and inertia forces. Analyses suggested that liquefaction w o u l d occur at four o f the five locations. Liquefact ion was not predicted at the south shore. The l iquefaction resistance at the south shore location was on the borderline i n terms o f the triggering criteria. The south shore locat ion stratum is very similar to that o f the north shore (where significant l iquefaction is predicted). The displacement analyses at the two locations were compared and contrasted, revealing what movements could be expected at either end i f l iquefaction were to occur or not (i.e., depending on assessment o f different earthquake magnitudes). Since the computed liquefaction-induced displacements were, often, beyond tolerances, potential remedial options were analyzed at the offshore locat ion detennined to be the most susceptible to liquefaction. Those analyses showed that the use o f certain remedial schemes w i l l decrease the displacements significantly. Because it is very difficult to access the stratum directly beneath the tunnel, densification o f zones adjacent to the tunnel is the most effective and economical ly feasible solution to l imi t displacements. iii Table of Contents Abstract i i Table of Contents i v List of Tables x List of Figures x v Acknowledgements x x 1 Introduction 1 1.1 Purpose 1 1.2 Scope & Organization o f Thesis 2 2 SITE DETAILS 4 2.1 Site Loca t ion 4 2.2 Tunnel Layout , Construction, and Cross-section V i e w 4 2.3 Locations o f Analyses 9 2.4 Ava i l ab le Information & Literature R e v i e w 9 3 SEISMICITY & SEISMIC LOADING 13 3.1 L o c a l Seismici ty 13 3.2 Seismic Load ing for Ground Response Analyses 14 3.2.1 Modi f i ca t ion o f Accelera t ion Time-Histor ies 15 4 ANALYSIS PROCEDURES & ENGINEERING PARAMETERS 18 4.1 Introduction 18 4.2 Liquefact ion Assessment 21 4.2.1 Triggering Resistance ( C R R ) o f Sands 22 4.2.1.1 Overburden Pressure Correct ion ( K 0 ) 25 4.2.1.2 Magnitude Correct ion ( K ^ 27 4.2.1.3 Static Shear Correct ion (K„) 27 4.2.1.4 Est imation o f Fines Content 28 iv Table of Contents 4.2.2 Triggering Resistance ( C R R ) o f Fine-grained Soils 30 4.2.2.1 Non-Plas t ic Silts 30 4.2.2.2 Plastic Silts 32 4.2.3 C y c l i c Load ing ( C S R ) -- S H A K E Analyses 34 4.2.3.1 M a x i m u m Shear Modu lus ( G m a x ) 35 4.2.3.2 Spectral Response 37 4.3 Post-liquefaction Stability 38 4.3.1 F l o w Slide 38 4.3.2 Est imation o f Residual Strengths (S r) 39 4.3.3 Est imation o f L i m i t i n g Strains ( Y u m ) 43 4.4 Liquefact ion-induced Displacements 46 4.4.1 Introduction 46 4.4.2 E m p i r i c a l Case-His tory Based Methods 47 4.4.2.1 Introduction 47 4.4.2.2 Bartlett & Y o u d . . 47 4.4.2.3 Hamada 49 4.4.2.4 Tokimatsu & Seed -- Liquefaction-Induced Settlements 50 4.4.3 N u m e r i c a l Methods 52 4.4.3.1 Introduction 52 4.4.3.2 S O I L S T R E S S : Finite Element Analyses 53 4.4.3.2.1 Est imation o f Pre-Liquefact ion Shear M o d u l u s Constant 53 4.4.3.2.2 Est imation o f Post-Liquefaction Shear M o d u l u s Constant 53 4.4.3.3 L I Q D I S P : Single-Degree-of-FreedomAnalyses 58 5 L O C A L G E O L O G Y & R E V I E W O F S O I L D A T A 60 5.1 Introduction 60 5.2 General Surf ic ia l Geology ; 60 5.3 Ava i l ab le S o i l Data - Site Specific Surf ic ia l Geo logy 62 5.4 Interpretation o f Ava i l ab le So i l Data 66 5.4.1 Introduction 66 5.4.2 Loca t ion #2 68 5.4.3 Loca t ion #3 69 v Table of Contents 5.4.4 Loca t ion #4 70 5.4.5 Loca t ion #7.. 71 5.4.6 Loca t ion #8 72 5.4.7 Development o f Longi tudinal -Direct ion S o i l Profi le 73 6 RESULTS: Liquefaction Triggering and Post-Earthquake Stability 74 6.1 Introduction 74 6.2 Liquefact ion Assessment and Associated Parameters 74 6.2.1 Zones o f Liquefact ion 74 6.2.1.1 Loca t ion #2 78 6.2.1.2 Loca t ion #3 81 6.2.1.3 Loca t ion #4 83 6.2.1.4 Loca t ion #7 85 6.2.1.5 Loca t ion #8 88 6.3 Post-Liquefact ion Stabili ty & Strength and Stiffness Parameters 91 6.3.1 F lows l ide Potential 91 6.3.2 Residual Strength (S r) & L i m i t Strain (yhim) Summaries 95 7 DISCUSSION of Chapter 6 Results 100 7.1 Introduction 100 7.2 Liquefact ion Assessment 100 7.2.1 Zones o f Liquefact ion 100 7.2.2 S H A K E : C y c l i c Load ing and Spectral Response 102 7.3 Post-Liquefact ion Stability 106 7.4 Future W o r k 108 8 RESULTS: Post-Earthquake Displacements & Associated Parameters 109 8.1 Introduction 109 8.2 Empi r i c a l Methods - Displacement Predictions 109 vi Table of Contents 8.3 Numer i ca l Methods ~ Displacement Predictions & Engineer ing Parameter Summaries 112 8.3.1 Introduction 112 8.3.2 Finite Element ( S O I L S T R E S S ) Displacements 112 8.3.2.1 Transverse-direction 113 8.3.2.1.1 Displacements at N o r t h and South Shores 115 8.3.2.1.2 Displacements at Offshore Locat ions 125 8.3.2.2 Longi tudinal-Direct ion Displacements 139 8.3.3 L I Q D I S P » Single-degree-of-freedom Displacements 148 9 D I S C U S S I O N of C h a p t e r 8 Resul ts 150 9.1 Introduction.. 150 9.2 Compar ison o f Displacement Predictions 150 9.2.1 Compar ison o f Empi r i c a l and Numer i ca l M e t h o d Predictions 150 9.2.2 Underground Structure Case-Histories 154 9.3 E m p i r i c a l M e t h o d Displacements 155 9.4 Numer i ca l M e t h o d Displacements 159 9.4.1 Finite Element ( S O I L S T R E S S ) Displacements 159 9.4.1.1 Transverse-direction Analyses 163 9.4.1.2 Longitudinal-direct ion Analyses 166 9.4.2 L I Q D I S P ~ Single-degree-of-freedom Analyses 168 10 R E M E D I A L M E A S U R E S 171 10.1 Introduction 171 10.2 Presentation and Discuss ion o f Results 172 10.3 Prel iminary Remediat ion Recommendation 189 11 C O N C L U S I O N S 192 vii Table of Contents R E F E R E N C E S 194 A P P E N D I C E S 201 Appendix A Numerical Method Displacement Predictions 201 A. 1 Additional SOILSTRESS Analyses 201 A. 1.1 Displacements at Offshore Locations ~ Case 2: No Sediment Loading 201 A. 1.2 Case 3: Displacements With Increased Sediment Loads 205 A. 1.3 Case 4: Displacements With Increased Ground Velocity 208 A. 1.4 Full-Section Longitudinal Analyses 210 A. 2 Extended Newmark Model 215 A.3 Extended Newmark Model - Integration with SOILSTRESS Code 221 Appendix B SHAKE Analyses 225 B. 1 SHAKE Analysis Method 225 B.2 Comparison of CSR's Using Varyious Bedrock Velocities 226 B.3 Surface Spectral Response Results 227 B.4 Ground Motion Amplification Summary ..232 B.5 Liquefaction Analysis Summaries 233 B.5.1 Location #2 233 B.5.2 Location #3 234 B.5.3 Location #4 235 B.5.4 Location #7 236 B.5.5 Location #8 237 B. 6 Idriss (1990) Ground Motion Attenuation 238 Appendix C Post-Earthquake Stability 239 C. 1 Flowslide Analyses: Limit Equilibrium Stability at Offshore Locations..240 C.2 Flowslide Analyses with Higher Residual Strengths 243 viii C.3 Residual Strength (Sr) in Sands -- Stark & Mesri (1992) 246 C. 4 Residual Strength -- Seed & Harder (1990). 247 Appendix D Empirical Method Displacement Predictions 249 D. l Bartlett/Youd: Analysis Details 249 D. 1.1 Displacement Predictions with Epicentral Distance of 30km and Varying Slopes 249 D. 1.2 Displacement Predictions with Epicentral Distance of 60km and Varying Slopes 250 D.2 Hamada: Analysis Details 252 D.2.1 Model Details 252 D.2.2 Analysis Parameters and Predictions 253 D.3 Tokimatsu/Seed: Analysis Details 255 A p p e n d i x E Addi t iona l Figures and Charts U s e d i n Analyses 256 A p p e n d i x F Remediat ion Calculations for Each Loca t ion 259 A p p e n d i x G Ava i l ab le S o i l Data 260 G. 1 CPT and Borehole Soil Profiles and Summaries 261 G.2 Fine-grained Soils Data 275 ix List of Tables Table 3.1 - Seismic Des ign Parameters for George Massey Tunnel Site 13 Table 3.2 - G r o u n d mot ion parameters o f chosen earthquake motions 15 Table 4 . 1 - Corrections to Measured ( N ^ for Fines Content 28 Table 6.1 - Depth o f Liquefact ion at E a c h Loca t ion 76 Table 6.2 - Estimated S o i l Parameters for Liquefact ion Assessment at #2 Loca t ion 79 Table 6.3 - Estimated S o i l Parameters for Liquefact ion Assessment at #3 Loca t ion 81 Table 6.4 - Est imated S o i l Parameters for Liquefact ion Assessment at #4 Loca t ion 84 Table 6.5 - Estimated S o i l Parameters for Liquefact ion Assessment at #7 Loca t ion 86 Table 6.6 - Est imated S o i l Parameters for Liquefact ion Assessment at #8 Loca t ion 89 Table 6.7 - Res idual Strength (S r) and L i m i t i n g Strain ( Y L i m ) Magnitudes for Transverse Analyses 95 Table 6.8 - Res idua l Strength (S r ) and L i m i t i n g Strain (yLim) Magnitudes for Longi tudinal Analyses 96 Table 8.1 - E m p i r i c a l M e t h o d Displacement Estimates 110 Table 8.2 - Tokimatsu/Seed M e t h o d Post-liquefaction Settlement Estimates I l l Table 8.3 - S O I L S T R E S S Pre- and Post-earthquake Inputs for Transverse Analyses at A l l Locations 114 Table 8.5 - S O I L S T R E S S Pre- and Post-earthquake Inputs -- Loca t ion #7 117 x List of Tables Table 8.6 - S O I L S T R E S S Pre- and Post-earthquake Inputs - Loca t ion #8 117 Table 8.4 - S O I L S T R E S S Inputs for Liquef ied Materials at N o r t h Shore 115 Table 8.7 - Displacements at N o r t h and South Shore Locations 119 Table 8.8 - S O I L S T R E S S Pre- and Post-earthquake Inputs -- Loca t ion #2 125 Table 8.9 - S O I L S T R E S S Pre- and Post-earthquake Inputs -- Loca t ion #3 128 Table 8.10 - S O I L S T R E S S Pre- and Post-earthquake Inputs - Loca t ion #4 128 Table 8.11 - S O I L S T R E S S S o i l Input Parameters for Liquef ied Materials at Offshore Locations 130 Table 8.12 - Displacements at Offshore Locations 131 Table 8.13 - S O I L S T R E S S Pre- and Post-Earthquake Inputs for Mater ia ls C o m m o n to B o t h Halves 139 Table 8.14 - S O I L S T R E S S Pre- and Post-Earthquake Inputs for N o r t h E n d ~ Longi tudinal Ana lys i s 140 Table 8 . 1 5 - S O I L S T R E S S Pre- and Post-Earthquake Inputs for South E n d » Longi tudinal Ana lys i s 141 Table 8.16 - Northern H a l f — Longitudinal-direct ion Displacement Predictions 142 Table 8.17 - Southern H a l f - Longitudinal-direct ion Displacement Predictions 143 Table 8.18 - L I Q D I S P Input Parameters and Displacement Predictions at Offshore Locat ions 148 Table 8.19 - L I Q D I S P Input Parameters and Displacement Predictions at Onshore Locat ions 149 Table 9 . 1 - Results f rom E m p i r i c a l and Numer i ca l Displacement Methods 151 Table 9.2 - Compar ison o f S O I L S T R E S S Results - V a r y i n g Sediment Loads 164 xi Table 10.1 - Descriptions o f Densif icat ion Schemes 172 Table 10.2 - M o d e l l i n g o f Remediat ion Schemes ~ Displacements at Loca t ion #2 173 Table 10.3 - Required Densif icat ion to Prevent Liquefact ion at Loca t ion #2 189 Table 10.4 - Required Densif icat ion to Prevent Liquefact ion at Loca t ion #3 190 Table 10.5 - Required Densif icat ion to Prevent Liquefact ion at Loca t ion #4 190 Table 10.6 - Required Densif icat ion to Prevent Liquefact ion at Loca t ion #7 190 Table 10.7 - Required Densif icat ion to Prevent Liquefact ion at Loca t ion #8 191 Table A . 1.1 - Displacements at Offshore Locat ions -- N o Sediment L o a d i n g 201 Table A . 1.2 - Displacements at Loca t ion #2 — Increased Sediment Load ing 205 Table A . 1.3 - Displacements at Loca t ion #2 — Increased Ground V e l o c i t y 208 Table A . 1.4 - S O I L S T R E S S Pre- and Post-liquefaction Inputs Fo r Longi tudinal Section Analyses 210 Table A . 1.5 - Longitudinal-direct ion Ana lys i s Displacement Predictions 211 Table B.4 .1 - Ground M o t i o n Ampl i f i ca t ion at E a c h Loca t ion 232 Table B.5 .1 - Triggering Summary for Loca t ion #2 233 Table B .5 .2 - Triggering Summary for Loca t ion #2 -- Cont inued 233 Table B.5 .3 - Triggering Summary for Loca t ion #3 234 Table B .5 .4 - Triggering Summary for Loca t ion #3 -- Cont inued 234 Table B.5 .5 - Triggering Summary for Locat ion #4 235 Table B .5 .6 - Triggering Summary for Loca t ion #4 -- Cont inued 235 Table B .5 .7 - Triggering Summary for Loca t ion #7 236 xii List of Tables Table B.5 .8 - Triggering Summary for Loca t ion #7 -- Cont inued 236 Table B .5 .9 - Triggering Summary for Locat ion #8 237 Table B .5 .10 - Triggering Summary for Loca t ion #8 — Cont inued 237 Table B .6 .1 - Est imat ion o f Target Spectrum for Mod i f i ca t i on o f Time-His tor ies 238 Table C.4.1 - Res idua l Strength Estimates - Seed & Harder (1990) 248 Table D . 1.1 - Bar t le t t /Youd Parameters and Displacement Predictions at Loca t ion #2 -- R=30km 249 Table D . 1.2 - Bar t le t t /Youd Parameters and Displacement Predictions at Loca t ion #3 -- R=30km 249 Table D.1.3 - Bar t le t t /Youd Parameters and Displacement Predictions at Loca t ion #4 -- R=30km 249 Table D.1 .4 - Bar t le t t /Youd Parameters and Displacement Predictions at Loca t ion #8 -- R=30km 250 Table D . 1.5 - Bar t le t t /Youd Parameters and Displacement Predictions at Loca t ion #2 -- R=60km 250 Table D . 1.6 - Bar t le t t /Youd Parameters and Displacement Predictions at Loca t ion #3 -- R=60km 250 Table D . 1.7 - Bar t le t t /Youd Parameters and Displacement Predictions at Loca t ion #4 - R=60km 251 Table D . 1.8 - Bar t le t t /Youd Parameters and Displacement Predictions at Loca t ion #8 -- R=60km 251 Table D.2.1 - Hamada Parameters and Predictions at Loca t ion #2 253 Table D .2 .2 -- Hamada Parameters and Predictions at Loca t ion #3 253 Table D.2.3 -- Hamada Parameters and Predictions at Loca t ion #4 253 xiii List of Tables Table D.2 .4 — Hamada Parameters and Predictions at Loca t ion #8 254 Table D.3 .1 - Tokimatsu/Seed Parameters and Predictions at Loca t ion #2 255 Table D .3 .2 - Tokimatsu/Seed Parameters and Predictions at Loca t ion #3 255 Table D.3.3 - Tokimatsu/Seed Parameters and Predictions at Loca t ion #4 255 Table D.3 .4 - Tokimatsu/Seed Parameters and Predictions at Loca t ion #8 255 Table E . l - Correct ion Factors for Earthquake Magnitude 258 Table F . 1 - Blowcounts Required to Prevent Liquefact ion at Loca t ion #2 259 Table F .2 - Blowcounts Required to Prevent Liquefact ion at Loca t ion #3 259 Table F.3 - Blowcounts Required to Prevent Liquefact ion at Loca t ion #4 259 Table F .4 - Blowcounts Required to Prevent Liquefact ion at Loca t ion #7 260 Table F.5 - Blowcounts Required to Prevent Liquefact ion at Loca t ion #8 260 Table G.2.1 - Test Information and Summary o f Si l t Test Results ( B C Hydro , 1991) 291 xiv List of Figures Figure 2.1 - M a p o f Fraser De l t a 5 Figure 2.2 - Layout o f Construct ion Site 6 Figure 2.3 - George Massey Tunnel Cross-section 8 Figure 2.4 - Overhead (plan) V i e w o f Cone Penetration Test and Preconstruction Borehole Locations 10 Figure 2.5 - Locat ions o f Test Holes — Profile V i e w 11 Figure 3 . 1 - Target Spectrum Corresponding to the M e a n + l a L e v e l o f Attenuation 16 Figure 4.1 - Seismic Assessment Procedure Flowchart 20 Figure 4.2 - Relationship Between Stress Ratios Causing Liquefact ion and ( N J ^ for M=7.5 Earthquakes 24 Figure 4.3 - K Q vs. Effective Overburden Pressure ~ Compar ison o f Duncan D a m Estimations w i th Other K 0 Curves 25 Figure 4.4 - Correlat ion Between Norma l i zed Shear W a v e V e l o c i t y and C y c l i c Stress Rat io ( C S R ) to Cause Liquefact ion 31 Figure 4.5 - Idealized Post-cycl ic Stress-strain Response 40 Figure 4.6 - Shear Strains as a Funct ion o f Factor o f Safety to Liquefac t ion and (N,.)« 43 Figure 4.7 - Shear Strains and L i k e l y Damage as a Func t ion o f (N i ) 6 0 i f Liquefact ion is Triggered 44 Figure 4.8 - Determination o f Volumet r ic Strains i n Saturated Sands 51 Figure 5 . 1 - Compar ison o f S o i l Data at Loca t ion #2 68 Figure 5.2 - Compar ison o f S o i l Data at Loca t ion #3 69 Figure 5.3 - Compar ison o f S o i l Data at Loca t ion #4 70 Figure 5.4 - Compar ison o f S o i l Data at Loca t ion #7 71 Figure 5.5 - Compar ison o f S o i l Data at Loca t ion #8 72 xv List of Figures Figure 6.1 - Zones o f Liquefact ion i n Profi le V i e w 75 Figure 6.2 - Compar ison o f C y c l i c Stress and C y c l i c Resistance Ratios at Loca t ion #2 78 Figure 6.3 - Factor o f Safety Against Liquefact ion ( F S L ) at Loca t ion #2 80 Figure 6.4 - Factor o f Safety Against Liquefact ion ( F S L ) at Loca t ion #3 82 Figure 6.5 - Factor o f Safety Against Liquefact ion (FS L ) at Loca t ion #4 83 Figure 6.6 - Compar ison o f C y c l i c Stress and C y c l i c Resistance Ratios at Loca t ion #7 85 Figure 6.7 - Factor o f Safety Agains t Liquefact ion ( F S L ) at Loca t ion #7 87 Figure 6.8 - Compar ison o f C y c l i c Stress and C y c l i c Resistance Ratios at Loca t ion #8 88 Figure 6.9 - Factor o f Safety Against Liquefact ion ( F S L ) at Loca t ion #8 90 Figure 6.10 - Post-liquefaction L i m i t Equ i l i b r ium Stabili ty Ana lys i s at Loca t ion #7 (South Shore Dyke) 93 Figure 6 . 1 1 - Post-liquefaction L i m i t Equ i l i b r ium Stabili ty Ana lys i s at Loca t ion #8 (South Shore Dyke ) 94 Figure 6.12 - Mater ia l Number ing Scheme for Transverse Sections at Onshore Locat ions 97 Figure 6.13 - Mate r i a l Number ing Scheme for Transverse Sections at Offshore Locations 98 Figure 6.14 - Mate r ia l Number ing Scheme for Longi tudinal Section 99 Figure 8 . 1 - F i n a l Stratigraphy at Loca t ion #7 116 Figure 8.2 - F i n a l Stratigraphy at Loca t ion #8 118 Figure 8.3 - Displacement Pattern — Locat ion #7 120 Figure 8.4 - Displacement Vectors - Loca t ion #7 121 Figure 8.5 - Displacement Pattern — Loca t ion #8 123 Figure 8.6 - Displacement Vectors -- Loca t ion #8 124 xvi List of Figures Figure 8.7 - F i n a l Stratigraphy at Loca t ion #2 126 Figure 8.8 - F i n a l Stratigraphy at Loca t ion #3 127 Figure 8.9 - F i n a l Stratigraphy at Loca t ion #4 129 Figure 8.10 - Displacement Pattern for Case #1 -- Loca t ion #2 132 Figure 8 . 1 1 - Displacement Vectors for Case #1 -- Locat ion #2 133 Figure 8.12 - Displacement Pattern for Case #1 — Loca t ion #3 134 Figure 8 . 1 3 - Displacement Vectors for Case #1 — Loca t ion #3 135 Figure 8.14 - Displacement Pattern for Case #1 — Loca t ion #4 136 Figure 8.15 - Displacement Vectors for Case #1 -- Loca t ion #4 137 Figure 8 .16 - Displacement Pattern for Northern H a l f o f Tunnel 144 Figure 8.17 - Displacement Vectors for Northern H a l f o f Tunnel . 145 Figure 8 . 1 8 - Displacement Pattern for Southern H a l f o f Tunnel 146 Figure 8.19 - Displacement Vectors for Southern H a l f o f Tunnel 147 Figure 10.1 - Displacement Pattern — Remediat ion Case #D1 175 Figure 10.2 - Displacement Pattern ~ Remediat ion Case #D2 176 Figure 10.3 - Displacement Pattern — Remediat ion Case #D3 178 Figure 10.4 - Displacement Vectors — Remediat ion Case #D3 179 Figure 10.5 - Displacement Pattern -- Remediat ion Case #D4 181 Figure 10.6 - Displacement Vectors -- Remediat ion Case #D4 182 Figure 10.7 - Displacement Pattern -- Remediat ion Case #D5 184 xvii List of Figures Figure 10.8 - Displacement Vectors -- Remediat ion Case #D5 185 Figure 10.9 - Displacement Pattern — Remediat ion Case #D6 187 Figure 10.10 - Displacement Vectors -- Remediat ion Case #D6 188 Figure A . 1.1 - Displacement Pattern for Case #2 -- Loca t ion #2 202 Figure A . 1.2 - Displacement Pattern for Case #2 — Loca t ion #3 203 Figure A . 1.3 - Displacement Pattern for Case #2 — Loca t ion #4 204 Figure A . 1.4 - Displacement Pattern for Case #3a (4 meters o f Sediment Loading) 206 Figure A . 1.5 - Displacement Pattern for Case #3b (6 meters o f Sediment Loading) 207 Figure A . 1.6 - Displacement Pattern for Case #4 (Increased Ground Ve loc i t y ) 209 Figure A . 1 . 7 - Displacement Pattern — Paral lel to the A x i s o f the Tunnel 213 Figure A . 1.8 - Displacement Vectors — Paral lel to the A x i s o f the Tunnel 214 Figure A . 2 . 1 - B l o c k on an Inclined Plane Subjected to a V e l o c i t y Pulse ~ Newmark M o d e l 215 Figure A . 2 . 2 - Work-Energy Pr inciple -- Extended Newmark 217 Figure A . 2 . 3 - Work-Energy (Rigid-Plast ic Behaviour) — Newmark M o d e l 219 Figure A . 2 . 4 - S O I L S T R E S S ~ Idealized Nonl inear Shear M o d u l u s Es t imat ion 223 Figure A . 2 . 5 - S O I L S T R E S S -- Idealized Nonl inear B u l k M o d u l u s Est imat ion 224 Figure B . l . l - Est imation o f Shear Modu lus and Damping — S H A K E 225 Figure B.2 .1 - Compar ison o f C . S . R . ' s U s i n g V a r y i n g Bedrock Veloc i t ies at Loca t ion #2 226 Figure B.3 .1 - Ground Surface Spectral Response at Loca t ion #2 227 Figure B .3 .2 - Ground Surface Spectral Response at Loca t ion #3 228 xviii List of Figures Figure B.3.3 - Ground Surface Spectral Response at Loca t ion #4 229 Figure B . 3 . 4 - G r o u n d Surface Spectral Response at Loca t ion #7 230 Figure B.3 .5 - Ground Surface Spectral Response at Loca t ion #8 231 Figure C . l . l - Post-liquefaction L i m i t Equ i l i b r ium Stabili ty Ana lys i s at Loca t ion #2 240 Figure C.1 .2 - Post-liquefaction L i m i t Equ i l i b r ium Stabili ty Ana lys i s at Loca t ion #3 241 Figure C.1.3 - Post-liquefaction L i m i t Equ i l i b r ium Stabili ty Ana lys i s at Loca t ion #4 242 Figure C.2.1 - Post-liquefaction Stabili ty Ana lys i s at Loca t ion #7 w i th Higher Residual Strength 244 Figure C.2.2 - Post-liquefaction Stabili ty Ana lys i s at Loca t ion #8 w i th Higher Residual Strength 245 Figure C.3.1 - Compar i son o f Undra ined Cr i t i ca l and Y i e l d Strength Rat ios B a c k --calculated from F i e l d Case-histories 246 Figure C.4.1 - Relationship Between Res idual Strength and ( N ^ 247 Figure D.2.1 - Displacement Vectors and S P T Boreholes for Part o f Ni iga ta , Japan Ana lys i s by Hamada 252 Figure E . l - Ranges i n K a Factors 256 Figure E .2 - Relationship Between Volumetr ic Strain Rat io and Number o f Cyc le s 257 Figure G.2.1 - Pos t -Cyc l i c Load ing o f Si l t -- Test #HSS1 ( B C Hydro , 1991) 292 Figure G.2 .2 - Pos t -Cyc l i c Load ing o f Si l t -- Test #HSS2 ( B C Hydro , 1991) 293 xix A CKNO WLED GEMENTS This research project was carried out under the supervision o f D r . P . M . Byrne , whose expert advice, guidance, and confidence are gratefully acknowledged. The project was done under contract wi th the Br i t i sh C o l u m b i a M i n i s t r y o f Transportation & Highways ( M . O . T . H ) . I am thankful for the f inancial support that facilitated the study. I w o u l d l ike to very sincerely thank Dr . Turgut Ersoy for his support and advice throughout the course o f this study. Special thanks are also extended to D r . Hendra Jitno and D r . D o n Gi l lespie for providing their insight i n the early stages o f this project; it was a pleasure work ing w i t h a l l o f you . A n d f inal ly , thanks are offered to the support staffs at the minis t ry ' s Headquarters (Vic tor ia) , and Operations (Burnaby) branches; thank y o u for a l l o f the little favours y o u provided. Surinder Puar A p r i l , 1996 Vancouver , Br i t i sh C o l u m b i a xx CHAPTER 1 INTRODUCTION 1.1 Purpose This report presents the results and recommendations o f a geotechnical seismic response assessment o f the George Massey Tunnel . The tunnel is i n Br i t i sh C o l u m b i a ' s Fraser Del ta region ~ an area where liquefaction is a major concern. The purpose o f this report is to assess the seismic response o f the submerged port ion o f the tunnel, and the connecting sections at the north and south banks o f the Fraser River . The tunnel is owned and operated by the Br i t i sh C o l u m b i a M i n i s t r y o f Transportation & Highways ( M . o . T . H . ) . It represents a vi ta l l ink across the lower Fraser River , between the municipali t ies o f R i c h m o n d and Delta . The site is underlain by loose saturated soils that could l iquefy during the analyzed 1:475 year earthquake event. Seismic performance o f the tunnel depends on the extent o f l iquefaction, post-earthquake l imi t equi l ibr ium stability, and the total deformations — the ma in concern being l iquefaction-induced displacements. The study is developed i n four stages. First, available data describing the extent o f the so i l deposits and their engineering properties were collected and analyzed to produce so i l profiles describing the dynamic soi l properties o f the site. The second stage invo lved dynamic analysis to predict the ground response to design earthquake motions. The thi rd stage consisted o f l imi t equi l ibr ium analyses using post-liquefaction strength and strain 1 CHAPTER 1 Introduction parameters to detenriine post-liquefaction stability o f the tunnel. The f inal stage entailed using empir ical and numerical solution methods to estimate earthquake-induced displacements. These displacements are the result o f inertia forces, and gravity forces acting on the softened l iquefied zones. Displacements were computed using a finite element method, a closed form solution, and two empir ical methods. The settlements that occur w i t h time (due to dissipation o f excess pore pressure) were also computed. 1.2 Scope and Organization of Thesis General information about the site and structure are provided i n Chapter 2. Input seismic motions and the design earthquake ground motions are described i n chapter 3. The ma in purpose o f this study is to assess post-liquefaction deformations. B o t h numerical and empir ical methods were applied to assess the potential displacements. Fini te element (numerical) deformation analyses are the focal point o f this study. Descript ions o f the liquefaction assessment procedures and the different methods used to assess l iquefaction-induced displacements are found i n chapter 4. Information about the general and loca l geology o f the area is presented i n chapter 5 . The methods used to interpret available soi l data and the procedures used to estimate so i l input parameters for the various analyses are also found i n chapter 5 . Summaries o f the so i l stratigraphy input parameters used i n the analyses are found i n chapter 6 and 8. Chapter 6 contains summaries o f the results o f the l iquefaction and post-earthquake 2 CHAPTER 1 Introduction stability analyses, and chapter 7 contains discussions about those results. Chapter 8 summarizes the results o f the displacement analyses, and chapter 9 discusses those results. The analyses are focussed on foundation response and do not deal w i t h the structural integrity o f the tunnel. In the finite element analyses there is some recognit ion o f soil-structure interaction, but the tunnel structure is not analyzed. A l though investigation o f structural tolerances is beyond the scope o f this investigation, remediation options to reduce earthquake-induced displacements have been analyzed. Chapter 10 is concerned wi th the effect that remediation schemes w o u l d have on the tunnel's response to earthquake loading, and also contains a prel iminary retrofit recommendation. Chapter 11 contains the report's conclusions. 3 CHAPTER 2 SITE DETAILS 2.1 Site Location The George Massey Tunnel crosses the south arm o f the Fraser R i v e r between L u l u Island and Deas Island, connecting the municipali t ies o f R i c h m o n d and Del ta v i a Highway#99 (refer to figure 2.1). The boundary separating the two municipal i t ies is approximately at the midpoint o f the tunnel. 2.2 Tunnel Layout, Construction, and Cross-section Geometry Because this study focusses on the geotechnical aspects o f the tunnel's seismic response, the description o f the tunnel's structural details has been kept to a rninimum. (Refer to H a l l et al . (1957) for a comprehensive review o f the structural features o f the tunnel). The Br i t i sh C o l u m b i a T o l l Highways and Bridge Author i ty commissioned the development o f the tunnel i n 1955. It was designed by the Dan i sh f i rm o f Chr is t iani & Nie l sen , and constructed by the Foundation o f Canada Engineer ing Corporat ion from Mont rea l , Quebec. The project was completed i n 1958. A s shown i n figure 2.2, the tunnel consists o f three general sections: the 550-meter L u l u Island (northern) approach, the subaqueous section consisting o f six precast concrete elements sparming 630 meters, and the 335-meter Deas Island (southern) approach. The 4 CHAPTER 2 Site Details CHAPTER 2 Site Details Figure 2.2: Layout of Construction Site CHAPTER 2 Site Details subaqueous port ion o f the tunnel consists o f six precast concrete elements o f rectangular cross-section. E a c h element is approximately 105 meters long, 24 meters wide , and 7.2 meters high. The s ix sections were constructed i n a drydock, and were then floated out and sunk into posi t ion i n a pre-dredged trench. E a c h element was placed on a series o f foundation blocks, and hydraulic jacks were then used to posi t ion it vert ical ly. Sand was jetted into the space between the bottom o f the element and the excavation. U p o n complet ion o f the sand jetting, the remainder o f the excavation was f i l l ed w i t h gravel and rock f i l l materials (refer to figure 2.3). These materials prevent erosion and buoyant uplift o f the tunnel. The trench side slopes were constructed to be shal low (1:3) to prevent ref i l l ing o f the trench during excavation. Pr imary differential settlements were accounted for by applying a 3-month delay before constructing the structural joints that connect the subaqueous tunnel section to the ventilation buildings. The tunnel ( in the cross-sectional plane) consists o f two 2-lane roadways separated by a partition w a l l (refer to figure 2.3). Vent i l a t ion ducts run along the outer sides o f the fu l l length o f the roadway. M . o . T . H provided copies o f preconstruction and construction (as-built) drawings o f the tunnel. These drawings were the main source for summarizing the tunnel geometry and layout. Tunnel dimensions were obtained from drawings #14-E-1615 and #14-E-1613, and the details o f the protection plan (rip rap) from #14-J-1710. Based on drawing #14-E-1618, the 15001b rock varies i n thickness from 2.5 feet to 6.9 feet, so an average thickness o f 7 CHAPTER 2 Site Details 8 C H A P T E R 2 Site Details 4.7 feet was applied to construct the finite element meshes for the displacement analyses. 2.3 Locations of Analyses The analyses correspond to the locations where five o f s ix M . o . T . H Cone Penetration tests ( C P T ) were done. ( The M . o . T . H C P T ' s are discussed i n section 2.4). The five locations were chosen because both C P T and S C P T data were available there. S h o w n i n figure 2.4 is an overhead (plan) v i ew o f the locations o f the 1991 C P T ' s and the 1956 boreholes. The figure shows me offsets o f the C P T ' s relative to me tunnel centerline. Figure 2.5 shows a profile v i ew o f the tunnel and the locations o f the test holes. The locations have been numbered according to the or iginal M . o . T . H 1991 C P T ' s (i.e., locat ion #2 corresponds to C P T #91-2, etc.). 2.4 Available Information & Literature Review In 1991, M . o . T . H undertook a geotechnical site investigation using current state-of-the-practice in-si tu testing methods. Cone Penetration Tests ( C P T ) , Seismic Cone Penetration tests ( S C P T ) , and Standard Penetration Tests (SPT) were performed at the L u l u Island (north shore) and Deas Island (south shore) on-ramps, and at offshore locations. The offshore tests were performed at off-center locations that were outside the riprap boundaries on either side o f the tunnel. The data was summarized and analyzed 9 CHAPTER 2 Site Details CHAPTER 2 Site Details C H A P T E R 2 Site Details for l iquefaction potential i n a report by Dr . D o n Gi l lesp ie o f M . o . T . H . Dr .G i l l e sp i e indicated that the foundation soils comprised main ly deltaic sands and silts that may be susceptible to liquefaction. S o i l test data is l imited. Before M . o . T . H ' s 1991 C P T ' s , so i l data at the tunnel site was l imi ted to the original Shelby tube and D y n a m i c Cone Penetration test ( D C P T ) data acquired during the preconstruction geotechnical site investigation i n 1956. The f i rm o f R i p l e y & Associates were responsible for acquiring and summarizing the data. Because no correlation exists between the D C P T and S P T , the penetration test data was o f little value; however, the Shelby tube sample data played a significant role i n this study. H a l l et al . (1957) described the site conditions, design requirements, and construction procedures that were to be applied i n the development o f the yet incomplete tunnel project. Traffic considerations and historical background are also discussed i n the report. In 1989, M . o . T . H . commissioned a prel iminary seismic analysis o f the tunnel. K e r Priestman & Associates ( K P A ) carried out a structural evaluation o f the seismic response o f the tunnel. Hardy B B T L t d . (now known as A g r a Earth & Environmental Ltd . ) assessed available geotechnical data to provide parameters necessary for K P A ' s dynamic analysis o f the runners earthquake response. The geotechnical analysis was constrained. by the lack o f data relevant to current seismic analysis procedures. 12 CHAPTER 3 SEISMICITY & SEISMIC LOADING 3.1 Local Seismicity In accordance w i t h M . O . T . H Seismic Des ign and Rehabil i ta t ion P o l i c y (1994) and the Fraser Del ta Task Force Report (Anderson & Byrne , 1991) recommendations, a Richter magnitude M 7 M a x i m u m Des ign Earthquake ( M D E ) was selected i n this study. The Task Force report predicts f i rm ground horizontal accelerations ranging between 0.16g and 0.27g, depending upon a site's locat ion wi th in the Fraser Del ta . Results o f a 1989 Paci f ic Geoscience Centre ( P G C ) seismic risk assessment at the north end o f the tunnel are summarized i n table 3.1. The table shows the peak ground accelerations and velocit ies w i t h their respective return periods and probabilities o f exceedance. The design earthquake corresponds to a return period o f 475 years or, i n other terms, a 10% probabi l i ty o f exceedance i n a 50-year period (refer to table 3.1). A design peak horizontal ground acceleration o f 0.24g was applied i n this study. Probabi l i ty o f exceedance i n 50 years 4 0 % 2 2 % 1 0 % 5 % Return Per iod (years) 100 200 475 1000 Peak Hor izon ta l Ground Accelera t ion (g) 0.097 0.147 0.238 0.344 Peak Hor izonta l Ground V e l o c i t y (m/s) 0.078 0.124 0.219 0.334 Table 3.1 - Seismic Design Parameters for George Massey Tunnel Site 13 C H A P T E R 3 Seismicity & Seismic Loading 3.2 Seismic Loading for Ground Response Analyses Seismic loading depends on the seismicity o f the region and the level o f r isk the designer wishes to take. W h e n choosing earthquake records (acceleration time-histories), ideal ly they should represent both loca l tectonic and soi l conditions. Important factors to consider when assessing w h i c h input acceleration time-histories should be used are: earthquake magnitude, type (eg. strike-slip), duration, surface on w h i c h motions were recorded (i.e., rock or soil) , frequency content, distance from epicenter to recording station, and focal depth. Accelera t ion time-histories recorded on bedrock are preferred because the frequency content, signal amplitude, and signal duration become altered after leaving the base bedrock and propagating up through the soi l (Telford, W . et al . , 1976). The chosen design earthquake input mot ion is applied b y the total stress dynamic ground response program S H A K E (described i n section 4.2.3) as a base mot ion at an identified f i rm base wi th in the soi l profile. The S H A K E program scales the input accelerations to a site-specific f i rm ground max imum magnitude defined b y the user (i.e., amax = 0.24g at the tunnel site). In this study, the recorded motions from the magnitude M=6 .4 1971 San Fernando, Ca l i fo rn ia earthquake were applied to the soi l columns. Acce le ra t ion time-histories were modi f ied to match a target response spectrum before being applied to the so i l columns. (The next section describes the modif icat ion procedure). The chosen motions were those 14 CHAPTER 3 Seismicity & Seismic Loading recorded at three sites located less than 70 kilometers from the epicenter: C a l Tech Seismological Laboratory, Lake Hughes, and Grif f i th Park Observatory. G r o u n d mot ion parameters recorded at the three sites are summarized i n table 3.2. Loca t ion B A S E Type A V A / V (g) (m/s) (ratio) C a l Tech Granite 0.19 0.12 1.58 Gr i f f i th Granite 0.18 0.21 0.86 Park Lake Weathered 0.17 0.06 2.83 Hughes #4 Granite Table 3.2 - Ground Motion Parameters From Chosen Earthquake Records The P G C predictions o f peak ground acceleration and veloci ty at the tunnel site for the 1:475 year earthquake are approximately 0.24g and 0:22m/s. Consequently, acceleration time-histories w i t h A / V (acceleration/velocity) ratios approximately equal to 1 were preferred; although the Lake Hughes #4 record was an exception. 3.2.1 Modification of Acceleration Time-Histories The 3 acceleration time-histories used i n the ground response analyses were modi f ied to match a target response spectrum before being applied i n the S H A K E analyses. The target response spectrum (shown i n figure 3.1) corresponds to the mean-plus-one standard-deviation level o f attenuation. T o determine the target spectrum, Tdriss ' attenuation relation (1990) was applied. 15 CHAPTER 3 Seismicity & Seismic Loading ( A n attenuation relationship describes the decrease i n a given earthquake ground mot ion parameter (Eg. peak acceleration) as a function, pr imari ly , o f both earthquake magnitude 0.70 —I 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 Period (sec) Figure 3.1- Target Spectrum Corresponding to the Mean+la Level of Attenuation and the distance to the epicenter). A l l current attenuation relationships use a log-normal distribution to represent the ground mot ion parameters. The fo l lowing equation estimates the attenuation i n ground motion: Ln(Y) = [aQ+exp(al + a2lvl)]+[J30-Qxp(j31+ (Equat ion 3.1) 16 CHAPTER 3 Seismicity & Seismic Loading where: Y = the ground mot ion parameter, M = earthquake magnitude, R = closest distance to the source ( in km), F = factor representing type o f fault, e = the standard error term, and values o f an and P„ (summarized i n appendix B .6 ) are for peak horizontal accelerations corresponding to a series o f periods at a spectral damping ratio o f 5%. A s mentioned before, a suitable design earthquake for dynamic analysis i n the Fraser De l t a is one o f magnitude 7.0, producing a bedrock acceleration o f about 0.24g. A n epicentral distance o f 31 kilometers was applied i n the attenuation estimate to anchor the target spectrum to a f i rm ground (i.e., period, T=0) acceleration o f 0.24g. The program S Y N T H (Naumoski , 1985) was used to generate synthetic acceleration time-histories for each o f the 3 records., The program generates an acceleration time-history whose response spectrum is a reasonable fit to a selected target spectrum. T o match the computed artificial history's spectrum w i t h the target spectrum, the computed spectrum is raised (or suppressed) iteratively as the Fourier coefficients are modif ied. 17 CHAPTER 4 ANALYSIS PROCEDURES & ENGINEERING PARAMETERS 4.1 Introduction T o analyze seismic response, two general approaches can be used: - dynamic effective stress analysis, or - total stress analyses. A total stress equivalent-elastic dynamic analysis procedure was used i n this study to evaluate the smal l strain onset o f triggering. U n l i k e the total stress approach, the effective stress method attempts to capture the complete stress-strain—pore-pressure response o f a so i l element under cyc l i c loading. This approach accounts for both the small strain response and pore-pressure rise prior to triggering, as w e l l as the large strain response after triggering. These results are then incorporated i n a finite element analysis to predict the development o f zones o f l iquefaction and the ensuing deformations. Refer to F i n n et al . (1986) for an example o f a proposed dynamic effective stress procedure. A l though the effective stress approach is the most realistic, due to the nature o f the data at this site, the determination o f the dynamic so i l properties w o u l d have been l imi ted; consequently, the total stress analysis procedure was used i n this study. The ma in concerns when assessing liquefaction are: 1. ) W h a t level o f cyc l i c stresses w i l l trigger liquefaction(?), and i f triggered 2. ) Wha t is the residual strength, and is it sufficient to prevent a f l o w slide(?), and 18 CHAPTER 4 Analysis Procedures & Engineering Parameters 3.) W h a t deformations w i l l occur? T o evaluate these concerns, the procedure was uncoupled as fo l lows: 1. ) Assess w h i c h zones w i l l be triggered to l iquefy using a conventional total stress dynamic analysis procedure (explained i n section 4.2), 2. ) Use l imi t equi l ibr ium stability analyses ~ applying post-liquefaction residual strengths i n the l iquefied zones ~ to assess f l ow failure (explained i n section 4.3), 3. ) Estimate liquefaction-induced displacements (section 4.4) using: a. ) empir ical methods, and b. ) numerical methods, 4. ) Estimate the further settlements that occur w i th time due to dissipation o f excess pore-pressures. A f l o w chart out l ining the general framework used to do the seismic response analysis is shown i n figure 4.1. Sections 4.2 to 4.4 present the procedures used to perform steps 1 to 4, as w e l l as the methods used to estimate the engineering parameters required for each analysis. 19 CHAPTER 4 Analysis Procedures & Engineering Parameters <D (0 C " O C C 3 13 H O L L CO UJ I-cc LU Q_ O LY. CL _ J o CO o H o < L L LU g _ i LU LY. D_ a: a: o c o u T3 4) CD £ 3 T J O as CO CO <D a: O X >» c O CD c £ CO Q T3 C CO cu E o cu O O D) TO (Do CO—I 0) CO a> o < OJ LL. C LU CD Z> g « _ l CO r— 43 CO -3 O a) CL CL Ui <D o o •J3 -l-> I a> a> o.E T3 C E * w ^ — F c — r: o (0 o ® ~ o -LZ o "55 'o. £ a. -£ § .<2 to HI Z Q LU 0£ O to to .>» CO w c cog is g o >% H Q C/3 O CD CO o (0 LU C o "•I—< o CD CO tz o t>- LU < CO >-< LU O o o a CD Q -3 (0 < LU >- o 1-o LU LU a ID < o LU LU a 3 20 CHAPTER 4 Analysis Procedures & Engineering Parameters 4.2 Liquefaction Assessment The characteristic behaviour o f saturated granular soils under cyc l i ng loading is complex. Loose to medium dense sands and silts may be triggered to l iquefy b y the osci l la t ing shear stresses induced by an earthquake. The shear strains required to trigger l iquefaction are small ~ generally between 0 .1% and 1% (Byrne, 1991). U p o n earthquake loading, granular soils undergo a pore-pressure rise during the sudden undrained loading condit ion; i f the pore-pressures increase unt i l they are equal to the overburden (total) stress, the effective stress i n the so i l w i l l be nu l l i f i ed and, therefore, resistance to shear forces w i l l be negated. S o i l i n this state is considered l iquefied, and its behaviour resembles that o f a dense f lu id . Structures on (or in) the f lu id so i l w i l l change elevation unt i l their buoyant force is equal to the weight o f the displaced f lu id . The l iquefaction assessment procedure involves the analysis o f each so i l layer, and is segregated into 3 parts: i . ) quantification o f so i l resistance (assessment o f the cyc l i c resistance ratio, C R R ) (sections 4.2.1 and 4.2.2), and i i . ) quantification o f dynamic earthquake effects (assessment o f the cyc l i c stress ratio, C S R ) (section 4.2.3), i i i . ) application o f the liquefaction criterion. The l iquefaction criteria consists o f a factor o f safety against l iquefaction ( F S L ) . The F S L is s imply the ratio o f the C R R to the C S R ( F S L = C R R / C S R ) . I f this ratio is 21 CHAPTER 4 Analysis Procedures & Engineering Parameters less than the prescribed value (for the site i n question), then l iquefaction is predicted to occur for that particular point wi th in the soi l . Depending on the type o f structure, generally, a factor o f safety between 1.1 and 1.4 is considered acceptable (Anderson, Byrne et a l . , R i c h m o n d Task Force ~ 1991). In this study, when a F S L less than 1.1 was estimated, that layer was predicted to liquefy. It should be noted that the actual factor o f safety is generally higher than that g iven b y the F S L ratio because the C R R is generally based on a lower bound o f the observed data rather than an average value (Anderson, Byrne et a l . , 1991). Results o f the l iquefaction assessments for each locat ion are summarized i n section 6.2. 4.2.1 Triggering Resistance (CRR) of Sands S o i l has an abil i ty to resist cyc l i c loading defined i n terms o f its c y c l i c resistance ratio, C R R = x / o 0 ' , where T is the cyc l i c shear stress required to trigger liquefaction, and a Q ' is the in i t ia l effective normal stress. Factors that can affect the c y c l i c resistance o f saturated sandy soils are: density, effective number o f cycles (o f loading), fines content, and existing static shear. The C R R is pr imari ly dependent on so i l density. Triggering resistance o f sands was determined using a procedure developed b y Seed et al . (1984). The Seed correlation is an indirect method based on f ie ld penetration data obtained at sites where liquefaction has occurred. It should be noted that 22 CHAPTER 4 Analysis Procedures & Engineering Parameters l iquefaction resistance can also be estimated from direct testing (o f undisturbed so i l samples). Refer to P i l l a i & Stewart (1994) for an example o f the applicat ion and comparison o f the two different approaches. Seed's l iquefaction assessment chart (figure 4.2) correlates the C S R ^ t (cri t ical cyc l i c stress ratio to cause liquefaction) w i t h normal ized penetration resistance. The penetration resistance (denoted as (N^) is based on standard penetration values normal ized to a confining stress o f 1 T / F t 2 and corrected to an energy level o f 60%. The applicable cyc l i c resistance is calculated by correction o f the C S R ^ as fo l lows: CRR = (CSR)CRIT-Kc-KM-Ka (equation 4.1) where K 0 is an overburden pressure correction, K M is an earthquake magnitude correction, and K a accounts for existing static shear. The fo l lowing sections describe the correction factors. 23 CHAPTER 4 Analysis Procedures & Engineering Parameters CHAPTER 4 Analysis Procedures & Engineering Parameters 4.2.1.1 Overburden Pressure Correction (K )^ Liquefaction is dependent upon the effective overburden stress ~ the greater the effective stress, the greater the potential for liquefaction. This relationship is depicted in figure 4.3 by the decrease in K„ with increasing effective confrnrng pressure. K a values 1.2 1.0 0.8 0.6 *6 0.4 0.2 0.0 Seed & Harder (1990) Duncan Dam -+- -4-500 1000. Effective Confining Stress (kPa) UBC Data UBC Data Dr=80°/o Dr=85% Dr=S0% Dr=S5% — Dr=70% Dr=20°/< » Dr=40% Dr=S0% 1500 Figure 4.3 - K„ vs. Effective Overburden Pressure ~ Comparison of Duncan Dam Estimations with Other K 0 Curves (after Pillai & Byrne, 1994) 25 CHAPTER 4 Analysis Procedures & Engineering Parameters were estimated using a correlation (figure 4.3) published by P i l l a i & Byrne (1994). The ' D u n c a n D a m ' curve was developed by comparing test results o f C R R w i t h estimations based on Seed's method. Tests were conducted on frozen samples taken during the extensive f ie ld tests for the Duncan D a m study. Figure 4.3 shows the D u n c a n D a m curve and the commonly used one developed by Seed and Harder (1990). The Seed & Harder curve is based on penetration resistance, and was developed from laboratory tests. The Seed & Harder f ie ld experience data base is l imited to confining stresses o f approximately 100 kPa . Addi t iona l ly , figure 4.3 shows the results o f laboratory test data carried out at the Univers i ty o f Br i t i sh C o l u m b i a ( U B C ) on both natural and tailings sands ( V a i d & Thomas, 1994). A s the figure shows, the U B C tests reveal that sand type has little effect, whereas the sand's density (i.e., relative density ~ D r ) does have a noticeable effect on K a . The U B C estimates o f K 0 are much larger than those o f Seed & Harder (1990). The choice o f w h i c h curve to use is cr i t ical i n estimating the cyc l i c shear resistance o f the soi l . Fo r instance, use o f the Seed & Harder curve may lead to lower C R R ' s , whereas use o f the U B C results may lead to an overestimation o f the C R R . A s figure 4.3 shows, the Duncan D a m curve lies between the U B C and Seed & Harder curves. Since the liquefaction testing o f the Duncan D a m foundation soils was very comprehensive, the use o f those results is more appropriate than the use o f the Seed & Harder results. Furthermore, since the U B C test results indicate that sand type does 26 C H A P T E R 4 Analysis Procedures & Engineering Parameters not have a significant effect on K 0 , the Duncan D a m curve should be applicable to the sands o f the Fraser Delta. 4.2.1.2 Magnitude Correction (KM) Seed's chart (figure 4.2) is for an earthquake magnitude o f 7.5 corresponding to 15 cycles; therefore an earthquake magnitude correction K M =1.1 was applied, as suggested by Seed (refer to table E . 1 i n appendix E ) , to account for the design magnitude 7, w h i c h is expected to produce approximately 12 effective cycles o f dynamic loading. 4.2.1.3 Static Shear Correction (Ka) A static bias (static dr iving shear stresses) can significantly decrease the liquefaction resistance o f loose soils, ' a ' is defined as the ratio o f shear stress on the horizontal plane to the effective normal stress. The factor ' a ' is used to estimate K a ; K a is the correction factor ( in equation 4.1) that takes into account the effects o f in i t i a l static bias on the horizontal plane. Rigorous finite element programs can be used to estimate a (i.e., the exist ing stress state i n non-level ground conditions). A correlation developed by Seed & Harder (1990) is commonly used to estimate K a . (The Seed & Harder correlation is shown i n figure E . 1 ( in appendix E)) . K K is the ratio o f the C R R at the in i t ia l static shear stress to the C R R at zero static shear stress. 27 CHAPTER 4 Analysis Procedures & Engineering Parameters Because the Massey tunnel is founded on a level excavation, static shear is anticipated to be rninimal; therefore, a value o f K „ = l was assumed for this study. Refer to P i l l a i & Stewart (1994) for a correlation between K a and a that was determined b y direct testing o f undisturbed sand samples. 4.2.1.4 Estimation of Fines Contents The greater the fines content, the greater the resistance to l iquefaction. Seed (1987) recommended that ( N J ^ -values be corrected for fines content by adding to the b l o w count as shown i n table 4.1. Fines Content (%) (Nx) 6 0 10 1 25 2 50 4 75 5 Table 4.1 - Corrections to Measured ( N j ) 6 0 for Fines Content In this study, fines content estimates for sands were derived from two sources. The off-center C P T data was compared wi th the original Shelby tube samples that were obtained along the tunnel centerline. First, a CPT-based estimate was derived: fr ic t ion (F r ) ratio and pore pressure ( B q ) ratio soi l classification zone values were used to estimate the fines content o f the sands. This procedure is explained i n detail i n the Robertson & Campanel la publication, "Guidel ines for use, interpretation and applicat ion o f the C P T 28 C H A P T E R 4 Analysis Procedures & Engineering Parameters and C P T U " (1986). S o i l classification types 8 and 9 were presumed to contain approximately 5% fines, type 7 was taken to contain 15%, and type 6 was assumed to contain 3 5 % fines. The estimates derived from that procedure were compared w i t h the preconstruction borehole data obtained along the centerline o f the tunnel. The comparison o f the data at each location is discussed i n more detail i n section 5.3. 29 • CHAPTER 4 Analysis Procedures & Engineering Parameters 4.2.2 Triggering Resistance (CRR) of Fine-Grained Soils 4.2.2.1 Non-plastic Silts T o assess the liquefaction resistance o f non-plastic silts, both the Seed (1984) approach (refer to section 4.2.1) and a shear wave velocity-based criteria were applied. The Seed (1984) data base represents an upper bound fines content o f 35%. Fines content corrections were applied as discussed i n section 4.2.1.4. A shear wave veloci ty criteria published by Robertson (1990) was also applied to assess the l iquefaction resistance o f non-plastic silts. Accurate shear wave ve loc i ty profiles can be determined using current seismic downhole methods. Shear wave ve loc i ty is influenced by many variables that affect l iquefaction resistance, such as so i l density, confinement, stress history, and geologic age; therefore, V s is a promis ing f ie ld index for evaluating l iquefaction susceptibility. The advantage o f using shear wave ve loc i ty as an index o f l iquefaction resistance is that it can be measured i n soils that are hard to sample, such as cohesionless silts and sands. S h o w n i n figure 4.4 is the proposed correlation between normal ized shear wave veloc i ty (Vs , ) and the cyc l i c stress ratio necessary to cause liquefaction. Since shear wave veloci ty varies w i th v o i d ratio and effective confining stress, the V s o f a sand o f constant density w i l l increase w i th increasing depth; consequently, the measured V s magnitudes were normal ized to the effective overburden stress: 30 CHAPTER 4 Analysis Procedures & Engineering Parameters p V = V (—V25 (equation 4.2) o\.„ where Wsl is the normalized shear wave velocity, and P a is atmospheric pressure expressed in the same units as the effective overburden pressure (o v o). o.o l 1 1 1 1— 50 100 150 200 250 Normalized Shear Wave Velocity, Vs, (mis) Figure 4.4 - Correlation Between Normalized Shear Wave Velocity and Cyclic Stress Ratio (CSR) to Cause Liquefaction (after Robertson, et al., 1990) 31 CHAPTER 4 Analysis Procedures & Engineering Parameters 4.2.2.2 Liquefaction Assessment Procedure for Plastic Silts In this study, when dealing w i th the liquefaction assessment o f plastic silts, a different approach was used ~ the Chinese criteria (Wang, 1979). The upper and lower l imits o f the range o f water content over w h i c h a soi l exhibits plastic behaviour are k n o w n as the l i qu id l imi t ( L . L ) and the plastic l imi t (P .L) , respectively. The water content range i tself is defined as the plasticity index (P.I) (i.e:, P.I . = L . L - P . L ) . The major variables that influence the liquefaction resistance o f soils containing significant fines are the plasticity o f the fines and the amount o f c lay size particles. I f uncertainties i n the measured soi l variables can be accounted for, then the fo l lowing criteria should be applied: - percent finer than 0.005mm (#200 sieve) < 2 0 % - l i q u i d l imi t ( L . L . ) £ 35% - natural water content (w n ) > 0.9-L.L. - l iquid i ty index (I w) > 0.75 A N D / O R - S P T b lows per foot ( N l ) 6 0 < 4 Soils that satisfy the five criteria are considered vulnerable to l iquefaction or significant strength loss. I f uncertainties i n the measured variables cannot be accounted for, then the l iquid i ty index should be ignored and the fo l lowing criteria should be applied: - percent finer than 0.005mm (#200 sieve) < 15% - l i q u i d l imi t ( L . L . ) < 3 6 % - natural water content ( w j > 0.92-L.L. 32 CHAPTER 4 Analysis Procedures & Engineering Parameters The Chinese criteria are only a prel iminary means o f assessment, and should not be rel ied upon solely i f the liquefaction resistance o f a soi l is questionable. I f a so i l fails the Chinese criteria, then its l ike l ihood for l iquefaction should be assessed b y cyc l i c shear tests. 33 C H A P T E R 4 Analysis Procedures & Engineering Parameters 4.2.3 Estimation of Cyclic Loading (CSR) T o estimate the cyc l i c stress ratio ( C S R ) , the max imum shear stress ( ( x c ) m a x ) at the midpoint o f each so i l layer is normal ized by div id ing by the effective overburden stress (a ' v o ) . Equat ion 4.3 estimates this uniform cyc l ic shear stress ratio ( C S R ) : CSR = 0.65-^^ Q 1 (equation 4.3) The factor 0.65 is applied to convert from a random loading to an equivalent uni form cyc l i c loading. The one-dimensional wave propagation analysis program S H A K E was used to estimate C S R ? s . The computer code S H A K E ~ a current state-of-the-practice total stress procedure ~ was applied to estimate the small strain onset o f triggering. The program conducts a total stress equivalent-elastic dynamic analysis; so, the effects o f pore pressure development and dissipation, and soi l hardening are not accounted for. A modif ied vers ion o f the S H A K E program ( S H A K E [4]) was used; it a l lows nine different modulus reduction and damping curves to be used. S H A K E uses the wave-equation method o f solution, w h i c h is based on the theory o f one-dimensional wave propagation i n a continuous medium. The method assumes that the earthquake can be represented by a shear wave propagating vert ical ly through soi l layers that extend infini tely i n the horizontal direction. Shear waves are input as 34 CHAPTER 4 Analysis Procedures & Engineering Parameters accelerations at equally spaced intervals. The wave equation methodology is discussed i n the S H A K E instruction manual (Schnabel, Lysmer , and Seed, 1972). The S H A K E solution method is discussed i n more detail i n appendix B . 1. The selection and modif ica t ion o f input ground motions (acceleration time-histories) for the S H A K E analyses is discussed i n section 3.1. The dynamic so i l properties required for the analysis are: the m a x i m u m shear modulus ( G m a x ) at l o w strains; the reduction o f G m a x w i th increasing shear strain; equivalent visco-elastic damping reduction curves, fraction o f cr i t ical damping, total unit weights (Ysat), and shear modulus coefficient ( K 2 m a x ) . The methods used to estimate these parameters are discussed i n upcoming sections (4.2.3.1 and 4.2.3.2). T o account for so i l damping and modulus reduction, the data publ ished by Sy et al . (1991 ) for the Fraser Del ta were applied i n the S H A K E analyses. The data is based on resonant co lumn tests supplemented by other published data (Sy et a l . , 1991). A t a l l locations the estimated so i l columns were taken to f i rm ground (Pleistocene) at a depth o f 200 meters. (The f i rm ground estimate is discussed i n detail i n section 5.3). Input motions were applied at the top o f the Pleistocene. 4.2.3.1 Maximum Shear Modulus (Gmax) The shear modulus at l o w strain amplitude ( G m a x ) was estimated f rom seismic downhole test shear wave veloci ty estimates. Estimates o f G m a x were required for both 35 C H A P T E R 4 Analysis Procedures & Engineering Parameters the ground response and deformation analyses. The assessment o f shear modulus was based on shear wave veloci ty ( V s ) measurements: 2 (Equat ion 4.4) where p is the so i l density, and V s is the shear wave veloci ty normal ized w i t h respect to effective overburden pressure (as described i n section 4.2.2.1). F o r comparison, low-strain shear modu l i ( G m a x ) were also estimated using a relation based on penetration resistance proposed by Seed & Idriss (1970): G -i\i.(K\ -P .(a'm\^ (Equat ion 4.5) atm where, (^2)max=1°-((^i)6o)06-F (Equation 4.6) The relation for estimating the max imum shear modulus coefficient ( K 2 m a x ) was proposed by Harder & Byrne (1992). Generally, K 2 m a x i n sands ranges from 30 (loose) to 90 (dense). ' F ' is a correction for material type: F = 0.6 i n Silt , and F = 1.0 i n Sand (Byrne , 1993). P a t m is atmospheric pressure, and o ' m is the mean normal effective stress. 36 CHAPTER 4 Analysis Procedures & Engineering Parameters 4.2.3.2 Surface Spectral Response S H A K E was used to estimate the surface mean-plus-one standard deviat ion pseudo-acceleration response spectra for a damping ratio o f 5%. The spectral response represents the response o f a smgle-degree-of-freedom system to the estimated surface accelerations. Effect ively, the single-degree-of-freedom system is subject to the acceleration time-history o f the ground surface. Response spectra output at each o f the five locations are summarized i n appendix B.3. 37 C H A P T E R 4 Analysis Procedures & Engineering Parameters 4.3 Post-liquefaction Stability Post-liquefaction performance o f the tunnel was detennined i n terms o f l imi t equi l ibr ium stability. Stabili ty (and the extent o f deformation) depends on the shear stiffness and unchained residual shear strength. In this section, the procedure for assessing the potential for a f l o w slide is in i t i a l ly discussed (section 4.3.1). A description o f the methods used to estimate post-liquefaction strength and stiffness parameters fol lows (sections 4.3.2 and 4.3.3). 4.3.1 Flow Slide Liquefact ion o f soi l layers can lead to the development o f a f l o w slide (unlimited strain). A f low slide occurs when dr iving stresses exceed the peak post-liquefaction ( ' residual ') strength o f the soi l . (Residual strength (S r) is discussed i n the next section). E v e n i f a f l ow slide does not result, lateral spreading may occur due to so i l softening upon liquefaction. T o assess f low slide potential, l imi t equi l ibr ium analyses were performed using B i s h o p ' s (Bishop, 1955) method o f irregular surfaces. The analyses required defining the tunnel structure and foundation geometry. Reduced post-liquefaction strengths were applied to l iquefied layers. A factor o f safety against f l ow slide ( (F .S . ) F L ) was set at 1.3, since this value is commonly used to represent the post-earthquake short-term undrained 38 CHAPTER 4 Analysis Procedures & Engineering Parameters failure condit ion (P i l l a i & Salgado, 1994). Analyses at the three offshore locations (described i n section 2.3) were performed for tunnel cross-sections (i.e., i n the plane transverse to the tunnel roadway). The north and south river banks (locations #8 and #7, respectively) were analyzed i n the longi tudinal plane. Results o f the analyses are summarized i n section 6.3. 4.3.2 Residual Strengths (Sr) A s discussed earlier, a soi l is considered l iquefied when pore-pressure rises to equal the total stress. After l iquefying, sand shows a stram-harderring response due to di la t ion and the consequent pore-pressure decrease, w h i c h leads to increased effective stress (as strains increase). The so i l w i l l recover some strength at large deformations; the m a x i m u m strength it regains is termed the residual strength (S r ). The relative density (D r ) o f the so i l w i l l dictate h o w much di la t ion w i l l occur after l iquefaction; therefore, the D r o f the soi l w i l l control h o w much residual strength the so i l can develop. Figure 4.5 graphically shows the applicable shear stress-strain relations as an ideal ized bil inear curve w i th ini t ia l slope equal to the shear modulus ( G L ) o f the l iquefied soi l , leading to the horizontal line at w h i c h the residual strength (S r) is then i n effect. 39 CHAPTER 4 Analysis Procedures & Engineering Parameters The proposed bil inear post-liquefaction stress-strain curve models have been CO CO (U (D -C CO Idealized Post-Liquefaction (Strain-Hardening) s Actual Response Shear Strain (yj Figure 4.5 - Idealized Post-liquefaction Stress-strain Response validated on samples from B C H y d r o ' s Duncan D a m (Byrne et a l . , 1994). The prel iminary results from testing carried out on undisturbed frozen core samples show that the post-liquefaction stress-strain relations can be adequately model led as a bi l inear curve. The test results conf i rm that, lacking site-specific laboratory tests, a reasonable 40 CHAPTER 4 Analysis Procedures & Engineering Parameters estimate o f the bil inear post-liquefaction stress-strain curve can be obtained by using the methods described be low to estimate the residual strength (S r) and the l imi t ing strain (YLim)- The Extended Newmark (Byrne, 1990) displacement estimation method recognizes this stress-strain relation. (Liquefaction-induced displacements are discussed i n upcoming sections). W i t h i n this study, the post-liquefaction stress-strain parameter estimates for sands are based on penetration resistance ((N^go) magnitudes. Parameter estimates for silt are based on laboratory monotonic and cyc l i c load test results on samples f rom another site. Stark & M e s r i (1992) showed that the post-liquefaction (residual) shear strength o f sands varies w i t h effective vertical stress (o ' v o ) , as opposed to common assumptions w h i c h disregard the l ink between residual strength and effective stress. Add i t i ona l ly , analyses o f undisturbed so i l samples from the Duncan D a m suggested that undrained shear strength is proportional to the in i t ia l consolidation effective vert ical stress (P i l l a i & Salgado, 1994). The Stark & M e s r i relation was developed through comparison w i t h the ( N j ^ - b a s e d correlation by Seed & Harder (1990). For reference and comparison, the Seed & Harder (1990) correlation is discussed i n more detail i n appendix C .4 . Stark & M e s r i (1992) applied a sampling and laboratory testing program i n w h i c h a relationship between the cri t ical strength ratio (a) and the equivalent clean sand S P T b l o w count was determined. Values o f y i e l d strength, at 15 equivalent cycles (earthquake magnitude, M=7.5) , and cri t ical strength ratio were measured using c y c l i c 41 CHAPTER 4 Analysis Procedures & Engineering Parameters t r iaxial , cyc l i c simple shear, and cyc l i c torsional shear tests. In this study, a general relation based on the findings o f Stark & M e s r i (1992) was used for estimating the unchained residual shear strength o f the sands and silts: Su (equation 4.7) — = a o' where, i . ) i n sands a = 0.07, 0.2-0.6, 0.6 for ( N ^ = 0-4, 10-14, and (NJeo * 15, respectively (Byrne, 1994a), and i i . ) i n silts cc = 0.4 was applied. This value was derived from test data from B C Hydro 's Pr ior i ty Transmission Tower Study (1990). (The B C H y d r o test results are summarized i n appendix G.2) . Sample #HSS1 showed the most s imilari ty (i.e., i n terms o f Atterberg l imits) to the non-plastic silts at the tunnel site, so that data was considered most suitable for this study. Furthermore, to confirm the residual strength estimates i n the silts, a relation suggested by Senneset (1981) was applied: Su=qc'/Nc' where q c - q c - U t and U t is the total measured dynamic pressure, and N c ' varies between 9 and 20. This relat ion has been developed to assess the undrained shear strength o f clays, but because the c lay content o f the silts is moderately h igh i n some cases, the relation was assumed to be crudely applicable for comparison purposes. Res idua l strength estimations are summarized i n section 6.3. 42 CHAPTER 4 Analysis Procedures & Engineering Parameters 4.3.3 Limiting Strains (YLim) The l imi t ing shear strain (Yiim ) is that strain at w h i c h di la t ion o f the so i l skeleton ceases, and the effective stress (and therefore, stiffness and strength) ceases to increase. The residual strength (S r) o f the so i l is then considered i n effect. (Refer to figure 4.5). Limiting Strain Potential (%) Figure 4.6 - Shear Strains as a Function of Factor of Safety to Liquefaction and (Ni)60 (After Seed et al., 1986) T o estimate l imi t strains i n sands, one o f two methods was employed. First , i f it 43 CHAPTER 4 Analysis Procedures & Engineering Parameters was possible to estimate the factor o f safety against l iquefaction ( (FS) L ) , then the Seed et al . (1986) graphic correlation (figure 4.6) was applied. The correlation requires b l o w counts ( (N x ) 6 0 ) and the ( F S ) L o f the sand. (The procedures used to estimate the ( F S ) L are 40 -f 30 4-20-4-10+ LIMTTING S H E A R STRAIN EF L I Q U E F A C T I O N O C C U R S Estimated Range (1984) 10 Proposed by Seed (1979) Based on Test Data by Tokimatsu & Yoshimi (1984) 50 High Damage Potential 60 Intermediate No Significant Damage Figure 4.7 - Shear Strain and Likely Damage as a Function of (N^t) if Liquefaction is Triggered (After Seed et al., 1984) 44 CHAPTER 4 Analysis Procedures & Engineering Parameters described i n section 4.2). W h e n the ( F S ) L o f the sand could not be estimated, then the correlation developed by Seed et al.(1984) was used: _ [2.2-o.o5(w)60] (equation 4.8) Him 1 U Equation.4.8 represents an average approximation based on the correlation shown i n figure 4.7 (Seed et a l . , 1984). The Seed et al . (1984) data is based on laboratory tests inc lud ing tests on undisturbed samples o f frozen cored samples. L i m i t strain estimates for l iquefied silts were based on the B C Hydro Transmission Tower (1990) laboratory monotonic and cyc l i c load test data. W h e n the ( F S ) L o f a sand was l o w (i.e., less than about 0.7), the predicted l imi t strains were high. The effect o f h igh l imi t strain predictions was discussed i n a review o f o f the seismic retrofit o f the Second Narrows Bridge (Byrne, 1994c). In accordance w i th the recommendations o f that review, an upper bound o f ( Y l i m ) m a x = 3 0 % was applied to the predictions for the sands i n this study. L i m i t strain estimates are summarized i n section 6.3. 45 CHAPTER 4 Analysis Procedures & Engineering Parameters 4.4 Liquefaction-Induced Displacements 4.4.1 Introduction Liquefaction-induced permanent ground displacements corresponding to the 475-year seismic event have been evaluated using two general types of methods: - empirical, and - numerical Empirical methods are based on data from field observations, and numerical methods are physics-based approaches. Two numerical procedures based on the Extended Newmark method (Byrne, 1991) were applied in this study, and two empirical methods were applied for verification and comparison purposes. Liquefaction-induced settlements are discussed, also. Initially, the procedures used in the empirical method analyses are outlined (in section 4.4.2). The empirical (Hamada, Bartlett & Youd, and Tokimatsu & Seed) methodologies are followed by descriptions of the numerical methods ( S O I L S T R E S S and L I Q D I S P ) . 46 C H A P T E R 4 Analysis Procedures & Engineering Parameters 4.4.2 Empirical Case History-based Methods 4.4.2.1 Introduction For comparison purposes, the Hamada (Hamada et al . , 1987) and Bar t l e t t -Youd (Bartlett & Y o u d , 1992) methods were applied to each transverse section profile to assess lateral displacements. B o t h methods use simplif ied equations to model observed trends o f l iquefaction induced displacements at l iquefied sites. The data presented b y Hamada et al . (1987) and by Bartlett & Y o u d (1992) are for slopes that retained sufficient residual strength to prevent a f low slide from occurring. Since these methods are based on observations o f actual l iquefaction induced displacements, they can serve the purpose o f provid ing a range o f displacements wi th in w h i c h the numerical estimations should fa l l . 4.4.2.2 Bartlett &Youd The Bar t le t t -Youd (1992) data base consists o f lateral displacements compi led from 8 major earthquakes. It is an empirical model developed using mult iple l inear regression to determine w h i c h parameters most affect the horizontal ground displacement. The data base consists o f displacements that occurred during the fo l lowing earthquakes: 1971 San Fernando, Cal i forn ia ; 1964 Ni igata , Japan; 1906 San Francisco, Ca l i forn ia ; 1964 A l a s k a ; 1979 Imperial V a l l e y , Cal i fornia ; 1983 B o r a h Peak, Idaho; 1983 N i h o n k a i -Chubu , Japan; and 1986 Superstition H i l l s , Cal i fornia . The regression analyses delineated two different types o f lateral spreads: 47 C H A P T E R 4 Analysis Procedures & Engineering Parameters i . ) lateral spread down gentle ground slopes ("Ground slope failure model") , and i i . ) lateral spread toward a free face ("Free face failure model") . The ground slope model is applied to s loping terrain. The free face model is applied i n cases where there is a lack o f lateral resistance, such as those that w o u l d occur near the shore o f a water body. Fo r example, free-field displacements at the north and south banks o f the Fraser R i v e r w o u l d be computed using the free face model (equation 4.9): LOG(DH+0.0\) = - 16.366 + 1.178-M - 0.921*LOG R- 0.013»i? + + 0.657»Z<9G W + 0.348»ZOG Tl5 + (eq'n. 4.9) + 4.527»Z(9G(100 - F 1 5 ) - 0.922»(£» 5 0 ) 1 5 Displacements at the offshore locations w o u l d be computed using the ground slope model LOG(DH+0.01) = - 15.787 + 1.178-M - 0.921*LOG R- 0.013-R + + 0.429-ZOG S + 0 .348- IGG Tl5 + (eq'n. 4.10) + 4 .527»IOG(100 - F 1 5 ) - 0.922»(£> 5 0 ) 1 5 where M = earthquake moment magnitude, R = horizontal distance ( in kilometers) f rom the seismic energy source, S = ground slope (in percent), W ( in percent) = 100* (Height o f free face/distance from free face), T 1 5 = cumulative thickness ( in meters) o f saturated granular layers w i th ( N l ) 6 0 < 15, F 1 5 = average fines content ( in percent) o f saturated granular layers included i n T 1 5 , and ( D 5 0 ) 1 5 = the average mean grain size ( in mil l imeters) i n layers included i n T 1 5 . The fo l lowing parameter l imits apply to the use o f these equations: 48 CHAPTER 4 Analysis Procedures & Engineering Parameters 6 . 0 < M < 8 . 0 0 . 1 < S < 6 . 0 5 % < W < 2 0 % 0.3 < T 1 5 < 15 0 . 0 < F 1 5 < 5 0 0 . 1 < ( D 5 0 ) 1 5 < 1 . 0 In cases where ' W is between 1 and 5 percent, it is suggested that both equations be applied, and the larger estimate be acknowledged (Bartlett & Y o u d , 1992). 4.4.2.3 Hamada The Hamada empir ical model (Hamada et a l . , 1987) was developed through analysis o f l iquefaction induced lateral spreads i n Japan. The case histories consist o f the l iquefaction induced displacements caused by the 1964 Ni iga ta (magnitude = 7.5) and 1983 Nosh i ro ( M = 7.7) earthquakes. The Hamada model was developed us ing pre- and post-earthquake aerial photographs. F r o m these, vector maps o f l iquefact ion induced ground displacements were developed based on ground deformation patterns w i t M n areas o f s imilar surface topography. ( A displacement vector map for part o f the Ni iga t a site is shown i n appendix D.2) . U p o n doing regression analyses, the fo l lowing s impl i f ied equation was developed to model the observed displacements: D=0 .75» i / ° 5 0 » 6 ° 3 3 (equation 4.11) where ' H ' is the l iquefied layer thickness ( in meters) and ' 0 ' is the m a x i m u m ground 49 CHAPTER 4 Analysis Procedures & Engineering Parameters slope or slope at the base o f the l iquefied layer ( in percent). 4.4.2.4 Tokimatsu & Seed --' Liquefaction-Induced Settlements Post-liquefaction settlements occur due to dissipation o f excess pore water pressures. F r o m a structural standpoint, vertical differential movements are the ma in focus when assessing post-earthquake consolidation. The post-liquefaction volumetric stiffness upon dissipation o f excess pore pressures can be determined from laboratory tests, but i n their absence, a correlation based on f ie ld and laboratory test data (Tokimatsu & Seed, 1987) can be applied. (N{)60 values o f each l iquefied layer were correlated w i t h the c y c l i c stress ratio at the midpoint o f each o f those layers using the volumetric strain chart (figure 4.8). The volumetric strain was used to estimate vertical displacements ( A v ) due to l iquefaction: where ' z ' is the thickness o f each l iquefied layer, ' H ' is the total layer thickness, and ' e v ' is the volumetr ic (i.e., vertical) strain induced by liquefaction. It should be noted that figure 4.8 is applicable to Richter magnitude M=7 .5 earthquakes, therefore figure E.2 ( in appendix E) was used to adjust the volumetr ic strain according to the design event (M=7.0) applied i n this study. (equation 4.12) 50 CHAPTER 4 Analysis Procedures & Engineering Parameters Cyclic Stress Ratio T 0.6 0.5 4 fVolumetric Strain, % 10 5 4 3 2 1 0.5 (N,) 60 40 50 Figure 4.8 - Determination of Volumetric Strains in Saturated Sands (After Tokimatsu & Seed, 1987) 51 CHAPTER 4 Analysis Procedures & Engineering Parameters 4.4.3 Numerical Models 4.4.3.1 Introduction A s mentioned earlier, numerical models are physics-based methodologies. Current procedures for estimating liquefaction-induced displacements range f rom simple to complex. N e w m a r k (1965) developed a renowned simple procedure, whereas dynamic effective stress analyses represent the complex end o f the spectrum o f methods. (Refer to section 4.1 for a br ie f discussion o f the dynamic effective stress method). M o s t current procedures are hindered by too many s impl i fy ing assumptions. A relatively simple and realistic method o f predicting seismic deformations was developed b y Byrne (1990): Byrne ' s procedure is an extension o f the simple N e w m a r k method from a single-degree-of-freedom rigid-plastic to a multi-degree-of-freedom flexible system using post-liquefaction stress-strain relations and energy concepts. The method is commonly cal led the 'Extended Newmark ' method. The Extended N e w m a r k methodology is described i n detail i n appendix A . 2 B o t h multi-degree-of-freedom finite element ( S O I L S T R E S S ) and single-degree-of-freedom ( L I Q D I S P ) computerized approaches were applied to estimate earthquake-induced displacements w i th in this study. B o t h computer program codes are based on the Extended N e w m a r k method. The application o f the Extended N e w m a r k method w i t h i n a multi-degree-of-freedom framework is discussed i n the next section, and a description o f 52 CHAPTER 4 Analysis Procedures & Engineering Parameters the single-degree-of-freedom closed form ( L I Q D I S P ) analyses fo l lows ( in section 4.4.3.3). 4.4.3.2 SOILSTRESS: Pseudo-Dynamic Finite Element Analyses M a n y simple procedures are currently available for estimating permanent l iquefaction induced displacements; but most do not a l low two-dimensional and rotational effects to be modelled. Often, the pattern o f displacements, and effects o f variations i n geometry and stratigraphy cannot be accounted for. In this study, the two-dimensional finite element program S O I L S T R E S S was used to assess i n detail seismically-induced displacements o f the tunnel. The analyses have been carried out using a modif ied version o f the code developed by B y r n e and Janzen (1989). Detai ls o f the computer code and the methods applied are g iven i n Byrne and Janzen (1989). The program models the behaviour o f soi l using a finite element formulation and equivalent-linear strain dependent modul i to model changes i n so i l properties. Liquefact ion effects are model led using the Extended Newmark method (Byrne, 1990). The S O I L S T R E S S multi-degree-of-freedom approach sums the displacements due to gravity loads (acting on the l iquefied soil) and the earthquake-induced inertia forces to arrive at an estimate o f the displacement at each node wi th in the discretized domain. The incorporation o f the Extended Newmark model wi th in the S O I L S T R E S S finite element 53 CHAPTER 4 Analysis Procedures & Engineering Parameters procedure is described i n detail i n appendix A . 3 . The S O I L S T R E S S code discriminates post-earthquake displacements b y applying the fo l lowing procedure: 1. ) analyze using pre-cycl ic stress-strain properties, 2. ) re-analyze acknowledging post-cyclic stress-strain properties and gravi ty loads acting on the softened.soil; apply a pseudo seismic coefficient (k) to represent the additional force needed to achieve the energy balance, 3. ) determine earthquake induced displacements by subtracting the displacements obtained i n step #1 from those i n step #2, 4. ) estimate post-liquefaction settlements due to excess pore-pressure dissipation, and add those to the displacements from step #3. Hyperbo l i c stress-strain models are used to represent the shear modulus (G s ) and bu lk modulus (B s ) i n the S O I L S T R E S S program code. Excess pore-pressure dissipation is simulated by applying an accordingly reduced bulk modulus i n the l iquef ied zones. The program requires a volumetric strain estimate w h i c h is used to estimate the reduction i n the bulk modulus. Volumet r ic strains were estimated using the Tokimatsu & Seed empir ical method. (The Tokimatsu & Seed method is described i n section 4.4.2.4). The key parameters that control the post-liquefaction stress-strain response are the residual shear strength (S r) and the l imi t ing shear strain (Yi i m ) - These parameters are described i n section 4.3. A l l S O I L S T R E S S inputs for pre- and post-earthquake 54 CHAPTER 4 Analysis Procedures & Engineering Parameters conditions are summarized i n section 8.3. The seismic coefficient (k) is described i n detail i n appendix A . 3. Displacements were analyzed i n both the longitudinal plane and i n the transverse (cross-sectional) plane. The longitudinal section was analyzed i n two separate analyses — northern and southern halves. The pre-consolidation vertical displacements from the longitudinal section analyses were added to the f ina l vertical displacements (i.e., after settlement) from the corresponding cross-section analyses to arrive at the estimates o f the total vert ical displacements. The reverse procedure was applied to estimate the total vert ical displacements i n the longitudinal plane. These vertical displacements are referred to as '3-dimensional ' i n the result summaries (section 8.3.2). In both the longitudinal and cross-section analyses, boundary pressures were acknowledged. Because a f ixed condit ion w o u l d not be representative o f actual conditions i n the field, horizontal soi l pressures corresponding to a l iquefied so i l (i.e., K Q = 1) were applied to each element at the left and right boundaries o f the finite element mesh i n the transverse analyses, and at the outer boundary i n each o f the two longitudinal analyses. The central boundary i n each o f the longitudinal meshes was f ixed because o f the presumed horizontal ly counter-acting effect o f the opposing northern and southern slopes. This , effectively, implies that there w o u l d be no lateral displacement at the midpoint o f the length o f the tunnel. Add i t iona l ly , at the offshore locations, vertical pressures corresponding to the 55 C H A P T E R 4 Analysis Procedures & Engineering Parameters aqueous load were applied on each surface element o f the mesh. K e y parameters i n the S O I L S T R E S S method are described i n the fo l l owing sections. 4.4.3.2.1 Pre-earthquake Shear Modulus Constant (k^ ) One o f the key input parameters i n the S O I L S T R E S S analysis is the shear modulus constant (k g ) . Experimental data suggest that G m a x can be approximated by: G m a x = kgpatm(-p—) (Equation 4.13) atm v 1 ' where o m ' is the mean normal effective stress, ' n ' is the shear modulus exponent that lies i n the range 0 to 0.5 and is determined from experimental data (Duncan et a l . , 1980), and P a t m is atmospheric pressure, w h i c h is applied to make k g and ' n ' dimensionless. A s mentioned earlier, the max imum shear modulus ( G m a x ) was estimated from shear wave veloci ty data; that estimate was subsequently applied i n equation 4.13 to solve for k g i . Since an in i t ia l shear modulus estimate for the tunnel excavation's coarse gravel/rockfiH protection was not available, the gravel k g-estimates o f Duncan , et al . (1980) were referenced. S O I L S T R E S S inputs for each particular analysis are found i n section 8.3.2 i n tabular summaries preceding each displacement summary. 56 CHAPTER 4 Analysis Procedures & Engineering Parameters 4.4.3.2.2 Post-liquefaction Shear Modulus Constant [(kg)Uq] The post-liquefaction shear modulus constant (k g ) l i q represents the large degradation o f stiffness o f l iquefied soils after pore-pressure rise. A s described i n the preceding section, k g can be solved for using equation 4.13. Since the shear modulus is the ratio o f shear strength to shear strain, the post-liquefaction shear modulus ( G l i q ) can be estimated as: • I ™ (Equation 4.14) where Sr is the residual shear strength, and Yum is the l imi t ing shear strain. The procedures used to estimate Sr and Yum ®cQ described i n section 4.3. F o r a l iquefied material, the shear modulus exponent (n) is approximately zero, so equation [4.14] reduces to: (Equation 4.15) Gliq = (kg)ljq'Patm C o m b i n i n g equations [4.14] and [4.15], (k g ) H q can be solved for: b q Yiim'-^ 'atm (Equation 4.16) 57 C H A P T E R 4 Analysis Procedures & Engineering Parameters 4.4.3.2.2 LIQDISP: Single-Degree-of-Freedom Analyses The Extended Newmark method (Byrne, 1990) was incorporated i n a single-degree-of-freedom analysis by Byrne (1990). The efficacy o f the computer code L I Q D I S P has been confirmed through comparison w i th observed f ie ld and laboratory shaking table values. The L I Q D I S P procedure requires the fo l lowing inputs: non-l iquefied ( 'crust ') and l iquefied layer thicknesses, residual strengths (S r) and l imi t strains (Yi i m ) w i t h i n l iquefied layers, and m a x i m u m ground velocity. The method considers the crust (i.e., non-l iquefied layer) and l iquefied soi l to be a single-degree-of-freedom elastic-plastic system. The concept is extended to a two-dimensional analysis i n a manner s imilar to that outl ined b y N e w m a r k (1965). T o estimate the static dr iving stress (xst) that cou ld be expected i n a two-dimensional analysis, a factor o f safety obtained from l imi t equi l ibr ium analysis is used to simulate the effect: st p (equation 4.17) where S r is the residual strength o f the l iquefied layer. A s i n the post-earthquake static stability analyses (section 4.3), l imi t equi l ibr ium analyses were performed us ing B i s h o p ' s method o f irregular surfaces (Bishop, 1955). The factor o f safety estimation is based on a circular failure surface that intersects the crust and l iquefied zones. W h e n applying the L I Q D I S P method, linear (elastic-plastic) stress-strain relations can be assumed for simplif icat ion; alternatively, nonlinear stress-strain response 58 CHAPTER 4 Analysis Procedures & Engineering Parameters can be applied. Displacements based on the nonlinear stress-strain response tend to be larger and more accurate than those using the linear approximation; however, the linear approximation method has been corroborated by cyc l ic shear tests that have shown that linear stress-strain response can be used adequately to model (simulate) those results (Byrne, 1990). The L I Q D I S P analysis results at each o f the analyzed locations are summarized i n section 8.3.3). 59 CHAPTER 5 LOCAL GEOLOGY & REVIEW OF SOIL DATA 5.1 Introduction This chapter provides an overall description o f the so i l conditions at the site. Sect ion 5.2 contains a broad description o f the loca l geology. Section 5.3 provides information about the available so i l data, and section 5.4 describes the interpretation procedure used to compare the data at each location. 5.2 General Surficial Geology The geological history o f the Western Fraser Lowlands was reviewed and information concerning the nature o f the thick so i l deposits that form the Fraser Del ta was examined. ( A map o f the Fraser Del ta is shown i n Figure 2.1). The Fraser De l t a has formed from the heavy load o f sand, silt, and clay transported by the Fraser River . M o s t o f the sediment is deposited i n the Strait o f Georgia, adding to the growing delta. E igh ty percent o f the sediment load is transported during M a y , June, and July. D u r i n g these months, the bed o f the r iver channel is heavi ly altered by rapid deposit ion and scour. The Fraser R i v e r sediments are o f Holocene age, and they over ly g lac ia l deposits o f the Late W i s c o n s i n age. The methods o f transportation and the environments o f deposition were the major factors that determined the phys ica l makeup (i.e., variations i n sand, silt, and clay content) o f the Quaternary deposits (i.e., sedimentary units). These 60 CHAPTER 5 Review of Soil Data units consist ma in ly o f unconsolidated fine-grained glacio-marine sediments (silts and clays). The sediments are overlain by a thick unit o f sandy foreset beds gently d ipping to the south-southwest into Boundary Bay . The sandy unit is overlain by a thin (approximately 2-meter thick) sequence o f fine silt and sand deposited i n overbank environments (Clague et al . , 1991). Throughout the Fraser Del ta , the sediment deposits vary i n thickness f rom 10 to 300 meters. Bedrock outcrops are found on the southern slopes o f the Coast Mounta ins and western slopes o f the Cascade Mountains . Outcrops are also found on the south shore o f Bur ra rd Inlet, on isolated hi l l s i n the Fraser L o w l a n d , and along several L o w l a n d creeks. Bedrock is w i th in 10 meters o f the surface i n less than 5 % o f the Fraser L o w l a n d . D u r i n g the Quaternary the lowland was subjected to repeated glaciations separated by nonglacial intervals. The ice, o f thickness up to 1800 meters or more, overrode a l l the Fraser L o w l a n d and much o f the adjoining mountainous areas. Deposits o f w i d e l y diversif ied or ig in were l a id down and molded as a series o f erosion (during deglaciation phases) and sedimentation sequences formed the present landscape. The Quaternary deposits (i.e., over lying bedrock) o f the L o w l a n d are underlain b y plant-bearing, freshwater, sedimentary rocks (interbedded sandstones, siltstones, mudstones, shales, and conglomerates) o f Upper Cretaceous (comprising 10% or less) and Tertiary (90% or more) ages. In downtown Vancouver , the Tertiary sedimentary rocks are at or w i t h i n a few meters o f the surface. Site-specific data is discussed i n the next section. 61 C H A P T E R 5 Review of Soil Data 5.3 Available Soil Data — Site Specific Surficial Geology S o i l properties can be obtained using direct or indirect methods. Di rec t methods such as testing o f undisturbed soi l samples are often inappropriate due to the h igh cost o f obtaining the samples. Indirect methods such as the cone penetration test ( C P T ) , seismic cone penetration test ( S C P T ) , Standard penetration test (SPT) , Becker penetration test ( B P T ) , and the pressuremeter are relatively inexpensive methods that are commonly used to obtain data required to estimate pre- and post-liquefaction so i l properties. The surficial geology at the tunnel site was assessed using Cone Penetration Test ( C P T ) data, preconstruction borehole logs, and Geolog ica l Survey o f Canada ( G S C ) shear wave veloci ty survey data. M . o . T . H carried out six Cone Penetration tests ( C P T ) , five Seismic Cone Penetration tests ( S C P T ) , and two Standard Penetration tests (SPT) i n 1991 to identify the basic engineering properties o f the foundation soi l units, and to estimate the general stratigraphy along the length o f the tunnel. The locations o f the G e o c o n boreholes and the M . o . T . H . C P T ' s are shown on the plan and profile v iews o f the tunnel site i n figures 2.4 and 2.5. C P T and borehole data and interpretations are summarized i n appendix G . In addition to the M . o . T . H data, soi l test data was available f rom the or ig inal preconstruction foundation investigation along the centerline o f the proposed tunnel excavation. The testing was done by Geocon L td . , and the data was summarized b y R i p l e y & Associates ( in 1956). The original so i l data is based on dynamic cone 62 CHAPTER 5 Review of Soil Data penetration tests ( D C P T ) and Shelby tube samples. A 3-inch (outside) diameter thin-wa l l ed Shelby tube and a 2-inch, 60-degree cone were advanced adjacent to each other. A rope and cathead system that dropped a 140-lb hammer a distance o f 1 foot was used to advance both the Shelby tube and the 2-inch cone. The number o f b lows required to advance the tube and the cone a distance o f 1 foot were recorded. The D C P T results cannot be correlated w i th current methods (i.e., S P T , C P T ) used to estimate liquefaction potential, therefore, the penetration data was o f no use i n this investigation. The Shelby tube soi l samples, on the other hand, were used to estimate the stratigraphy (soi l types and layer thicknesses) directly beneath the tunnel. The fo l lowing soi l classification tests were done: min imum and max imum v o i d ratio tests, moisture content estimations, and grain size analyses. Addi t iona l ly , t r iaxial and consol idat ion tests o f fine grained samples were performed. The basic engineering properties and strata were also estimated using the 1991 C P T data, but those data correspond to off-center test locations; therefore, the C P T -derived so i l engineering properties were transferred to the corresponding centerline Shelby tube sample-based strata estimates. The comparisons o f the C P T data w i t h the borehole sample data at each location are discussed i n section 5.4. The more recent data used i n this study was obtained using the C P T and S C P T . The four offshore C P T ' s were done using a d r i l l r ig and spud barge. Fo r this study, the C P T data (cone resistance, fr ict ion ratio, pore-pressures) was interpreted us ing the 63 . C H A P T E R 5 Review of Soil Data program C P T I N T , w h i c h interprets data according to the methods outl ined b y Robertson & Campanel la (1986). In figure 2.5, each C P T is plotted at an elevation relative to the geodetic survey control. The 1956 boreholes were done along the tunnel centerline at elevations corresponding to the 'or ig inal ground surface' shown i n figure 2.5. Offshore SPT's were also done, but due to equipment difficulties, the S P T data was considered unreliable; therefore, it was not used i n this study. The C P T ' s , on the other hand, provided reliable continuous stratigraphy data. The seismic cone penetration test ( S C P T ) is s imply an extension o f the C P T apparatus. The technique applies the downhole test procedure. A veloc i ty seismometer is hor izontal ly incorporated i n the cone to measure the horizontal component o f shear wave arrivals. Fo r M . o . T . H ' s offshore investigation (locations #2, #3, #4), explosives were used to generate shear waves, and at the onshore locations (#7, #8), the conventional hammer-plank shear beam source was used. Refer to Robertson & Campanel la (1986) for detailed descriptions o f the C P T and S C P T equipment. The testing conditions and techniques i n the M . o . T . H investigation are described i n a report prepared by M . o . T . H engineer D r . D o n Gi l lespie . A t a l l locations, the C P T logs end approximately 40 meters be low the geodetic datum survey control point (shown i n figure 2.5). Where the C P T information ends, a silt layer is applied over the remainder o f the stratigraphy to a depth o f 200 meters. A l t hough no shear wave data was available at locat ion #1 (between locations #8 and #4), the so i l 64 CHAPTER 5 Review of Soil Data type (silt) was discernable from the C P T bearing resistance, fr ict ion ratio, and pore pressure ratio data. The silt encountered at the end o f the C P T #91-1 l og was extrapolated to a f i rm ground depth o f 200 meters (for completing the S H A K E so i l columns). The Geolog ica l Survey o f Canada ( G S C ) performed shear wave refraction surveys at 70 sites wi th in the Fraser Del ta ~ site #16 being near the tunnel. Because o f the smooth travel-time-distance plot throughout the surveyed depth, it was presumed acceptable to extrapolate the silt layer to the f i rm ground depth. Recent seismic reflection surveys taken along the Fraser R i v e r show that f i rm ground is encountered at an approximate depth o f 200 meters (Hamil ton, 1994). The survey shows that it is the first firm surface encountered, and it has been assumed that this f i rm ground is the Pleistocene. B l u n d e n (1975) estimated the bedrock surface to occur at depths o f 250 meters or deeper under most o f the delta. A recent estimate by the G S C suggests that the depth to bedrock i n the v ic in i ty southeast o f the tunnel may be 700 meters (Hunter, 1994). The G S C ' s estimates are based on application o f a velocity-depth function derived specif ical ly for the Fraser Del ta . The function was derived f rom unpublished 2-w a y travel time data obtained by the Dynamic O i l Company o f Vancouver . The fo l lowing section provides comparisons o f the C P T and borehole data at each locat ion. 65 s CHAPTER 5 Review of Soil Data 5.4 Interpretation of Available Soil Data 5.4.1 Introduction This section compares the data from the 1956 boreholes w i t h the 1991 C P T ' s . S o i l layer types and their approximate thicknesses are identified for each location. Because the 1956 boreholes were done before excavation o f the trench took place, and the or iginal preconstruction ground surface elevation along the tunnel centerline was known, it was possible to determine h o w much was eventually excavated i n the v ic in i ty o f each borehole. F r o m this, it was possible to summarize the borehole-based estimates o f the post-construction stratigraphies. The engineering soi l properties o f those under lying so i l layers were then derived from the off-center C P T data. The offshore C P T ' s were taken downstream o f the tunnel centerline, at off-center distances o f between 45 and 79 meters. (Figure 2.4 shows the off-center C P T locations). Since the C P T cannot measure some soil properties very precisely, properties such as fines content and D 5 0 (mean grain size) were estimated by analyzing the corresponding layer i n the 1956 borehole sample data. W h e n comparing the borehole samples w i th the C P T data, patterns emerged at each location, such that so i l layer types identified b y the C P T were readi ly corroborated b y an adjacent borehole log. It was determined w h i c h so i l layers corresponded wi th each other i n the CPT-based strata estimates and the borehole-based strata estimates. Consequently, transposing the C P T data to the centerline was possible, 66 CHAPTER 5 Review of Soil Data thereby taking advantage o f both the detailed C P T data and the informative Shelby tube samples. The S O I L S T R E S S displacement analyses consisted o f detailed discretization o f the foundation and tunnel structure; therefore, it was deemed important to estimate as accurately as possible the characteristics o f the soils that actually underlay the tunnel at each o f the analyzed locations. The fu l l C P T profile was used i n the ground response ( S H A K E ) analyses. Since structural effects o f the tunnel could not be accounted for i n the S H A K E so i l co lumn estimate, i t was appropriate s imply to do the analyses using the fu l l C P T profiles to represent the free-field conditions wi th in w h i c h the tunnel is constructed. 67 CHAPTER 5 Review of Soil Data 5.4.2 Location #2 A t locat ion #2, the data from C P T #91-2 was compared w i t h the sampling data from boreholes #7 and #8. A s shown i n figure 2.4, C P T #91-2 is 52 meters downstream o f the tunnel centerline. Borehole #7 is 53 meters north o f C P T #91-2, and borehole #8 is 68 meters south o f the C P T . Figure 5.1 shows the point where the excavation ends ( ' tunnel invert ' ) relative to the beginning o f each data source. Borehole #8 Figure 5.1- Comparison of Soil Data at Location #2 The top 10 meters consist o f interbedded layers o f clean sand and sil ty sand; these layers are underlain by an 8-meter thick clayey silt unit. The clayey silt is underlain b y a 68 CHAPTER 5 Review of Soil Data 3-meter thick dense sand lens. 5.4.3 Location #3 For locat ion #3, the data from C P T #91-3 was compared to the sampling data from boreholes #5 and #6 (refer to figure 5.2). Borehole #5 is 75 meters north o f C P T #91-3, and borehole #6 is 44 meters south o f the C P T location. C P T #91-3 is 43 meters downstream o f the tunnel centerline. CPT #91-3 Data Borehole #5 Borehole #6 (No data for first 3.4 meters) 3 m 3 m 9 m _ / -Tunnel Invert Non-plastic Silt Plastic Silt 13 m (End of Data) Plastic Silt (A Tunnel Invert Non-plastic Silt 3 m - t 11 m 3 m Figure 5.2 - Comparison of Soil Data at Location #3 Figure 5.2 shows the point where the excavation ends relative to the beginning o f each test hole. C l e a n loose sands are interbedded wi th loose s i l ty sands to a depth o f approximately 6 meters; these interbedded layers are underlain b y 3 meter thick 69 CHAPTER 5 Review of Soil Data nonplastic sandy silt and 9 meter thick plastic clayey silt units. Borehole #6 confirms that the material encountered at the end o f C P T #91-3 is a sand seam, and is approximately 4 meters thick. 5.4.4 L o c a t i o n #4 A t locat ion #4, the data from C P T #91-4 was compared w i t h the Shelby tube samples from boreholes #4 and #5 (figure 5.3). Borehole #4 is 50 meters north o f C P T #91-4, and borehole #5 is 60 meters south o f the C P T location. Though C P T #91-4 is 79 meters downstream o f the tunnel centerline, the soi l profile corresponds w e l l w i t h the two boreholes. CPT #91-4 Data 3 m 4 m 14 m (No Data for first 3.5 meters) Sandy Silt Tunnel Invert Silt Borehole #4 Borehole #5 10 m oy/.-v.-v. Plastic Silt Silty Sand Tunnej Invert Plastic Silt 6 m 12 m (End of Data) Figure 5.3 - Comparison of Soil Data at Location #4 70 CHAPTER 5 Review of Soil Data A t locat ion #4, s ix meters o f loose to dense clean sand is underlain by medium-dense sandy silts to a depth o f 24 meters. 5.4.5 Location #7 For locat ion #7, the data from C P T #91-7 was compared to the sampling data from borehole #9 (refer to figure 5.4). Borehole #9 is 15 meters south o f the C P T #91-7 log , w h i c h is found 47 meters downstream o f the tunnel centerline. CPT #91-7 Data Water Table (@ 3.3m) Borehole #9 (No Data for first 2.7 meters) Tunnel "Invert Silty Sand 35 m (End of Data) Silt Silty Sand Silt Figure 5.4 - Comparison of Soil Data at Location #7 Loca t ion #7 is comprised main ly loose to medium-dense sil ty sands. The 71 CHAPTER 5 Review of Soil Data relat ively clean sands (i.e., fines content = 5 % to 15 %) extend throughout the C P T data log to an approximate depth o f 38 meters. Borehole #9 compares w e l l w i t h C P T #91-7. 5.4.6 Location #8 For locat ion #8, C P T #91-8 was compared wi th the data from borehole #1. A s shown i n figure 2.4, borehole #1 is 30 meters south o f the C P T #91-8 location, and C P T #91-8 is 48 meters upstream o f the tunnel centerline. CPT #91-8 Borehole #1 Data ..., Water Table (©4.2m) (No Data for first 5 meters) ^— ^ 5m t -26 m I (End of Data) Clay Fill 1 HI p i -•oar /////'/,', i Silt Figure 5.5 - Comparison of Soil Data at Location #8 Figure 5.5 shows the point where the excavation ends relative to the begirining o f 72 CHAPTER 5 Review of Soil Data each test hole. Loca t ion #8 is comprised main ly loose silty sands. The relat ively clean sands extend throughout the C P T log to an approximate depth o f 27 meters. 5.4.7 Development of Longitudinal-Direction Soil Profile The longitudinal stratigraphy was developed by comparing data f rom the C P T ' s and the boreholes. S o i l profiles vary along the length o f the tunnel. (Figure 6.1 shows the longitudinal soi l profile). Deposi t ion processes along the river bed have resulted i n two general contrasting profiles: those o f the offshore locations (i.e., #2, #3, and #4) versus locations #7 and #8 at the south and north shores o f the river, respectively. Loose sand is found to greater depths at the river banks, whereas along the course o f the river, silts extend to greater depths. Stratigraphies at the five analysis locations were developed as described i n sections 5.4.2. to 5.4.6. E a c h o f the cross-section stratigraphies were plotted on a profile drawing and the so i l layers were extrapolated. The stratigraphies at the north and south boundaries were estimated from the 1956 borehole data. Though data f rom boreholes #3 and #10 were not used i n the development o f the cross-section stratigraphies, they aided i n detenriining the so i l types at the north and south ends o f the tunnel. (Refer to figure 2.5 for a profile v i e w o f the test hole locations). 73 CHAPTER 6 RESULTS: LIQUEFACTION TRIGGERING & POST-EARTHQUAKE STABILITY 6.1 Introduction Results corresponding to the analyses described i n sections 4.2 and 4.3 are presented i n this chapter. Init ial ly, results o f the liquefaction assessments at each o f the five locations are presented. Post-liquefaction stability is then assessed, out l in ing the f l ow slide analyses. Input parameters cri t ical to the analyses are summarized, also. Results are discussed i n chapter 7. E a c h o f the upcoming sections points out the corresponding section i n chapter 7 that should be referred to wh i l e rev iewing the ind iv idua l result summaries i n this chapter. 6.2 Liquefaction Assessment and Associated Parameters 6.2.1 Zones of Liquefaction A t each o f the five locations (refer to figure 6.1), soi l columns were analyzed for l iquefaction resistance. Points o f discussion relating to the l iquefaction assessment are presented i n section 7.2. Refer to section 4.2 for the l iquefaction assessment procedure. A p p e n d i x B . 5 contains detailed information ( C S R , C R R , etc.) f rom the l iquefaction assessment at each location. Fo r reference, ground mot ion amplif icat ion factors are summarized i n appendix B . 4 . 74 75 C H A P T E R 6 Results: Liquefaction Assessment Figure 6.1 shows the zones o f liquefaction on a so i l profile paral lel to the axis o f the tunnel. Table 6.1 summarizes the depths to liquefaction both before and after construction o f the tunnel. A s the table shows, the largest zone o f l iquefact ion extends to a depth o f about 11 meters be low the tunnel at location #8. Amongs t the offshore locations, locat ion #2 is o f greatest concern, where the depth to l iquefact ion is approximately 8 meters. The zones o f liquefaction extend no more than about 3 meters at the remaining locations. A t the south shore (location #7), the depth to l iquefaction coincides w i th the point at w h i c h the tunnel excavation ends. Location Depth to Tunnel Invert (meters) Total Thickness of Liquefied Zone (meters) Thickness of Liquefied Zone Below Tunnel (meters) #2 4 11.7 7.7 #3 6.4 9.1 2.7 #4 8 10.5 2.5 . #7 20 20 0 #8 16 27.2 11.2 Table 6.1 - Depth of Liquefaction at Each Location Figures 6.2 to 6.9 graphically summarize the C S R ' s , C R R ' s , and Factors o f Safety at each location. The liquefaction assessment results at the three offshore locations (#2, #3, #4) are summarized first, fo l lowed by the summaries for the south (#7) and north (#8) shores. 76 CHAPTER 6 Results: Liquefaction Assessment In the tables i n sections 6.2.1.1 to 6.2.1.5, V S 1 is the normal ized shear wave veloci ty, and G m a x is the max imum (initial) shear modulus. (Ni)60 is the corrected b l o w count as estimated using the program C P T I N T , w h i c h uses a Q c / N ratio that varies according to so i l type, as described by Robertson & Campanel la (1986). 'F ines content' is the percentage o f fines passing a no.200 U . S . standard sieve, and K 0 i s the overburden correction factor for the liquefaction analyses. 'Cr i t e r i a ' is the l iquefact ion assessment cri teria used to assess the triggering potential o f the so i l layer, and 'status' refers to the outcome o f the l iquefaction analysis for that particular layer. In the tables, where TNT. A ' is found it means that the data for that entry was not available. It should be noted that for so i l layers where 'Seed & Robertson ' is shown as the l iquefaction criteria, both the Seed (1984) and Robertson (1990) criteria were indiv idual ly applied, and then the results were compared. K e y liquefaction parameters are summarized i n table 6.2 to 6.6. Refer to appendix B . 5 for detailed summaries o f ground motion parameters at each location. 77 6.2.1.1 Location #2 CHAPTER 6 Results: Liquefaction Assessment Cyclic Stress Ratio 0.10 0.20 0.30 0.40 0.50 Figure 6.2 - Comparison of Cyclic Stress and Cyclic Resistance Ratios at Location #2 78 Layer Soil Type Depth (m) (m/s) (N,)M Fines Content (%) K 0 Criteria Status No data from first 3 meters of CPT log 1 Sand 3-4.5 185 N.A. 5 1 Seed & Robertson Liquefied * Tunnel Invert Location (excavation ends at 4 meters below the original ground surface) 2 Sand 4.5-6 185 N.A. 5 1 Seed & Robertson Liquefied 3 Sand 6-7.4 160 13 5 1 Seed Liquefied 4 Sand 7.4-8.8 160 13 5 1 Seed Liquefied 5 Sandy Silt 8.8-9.2 160 7 45 1 Seed Liquefied 6 Silty Sand 9.2-9.7 160 18 45 1 Seed Liquefied 7 Clayey Silt 9.7-13.7 175 5 45 1 Seed & Robertson ** top 2 meters is Liquefied 8 Clayey Silt 13.7-17.5 185 5 95 N.A. Chinese Not Liquefied 9 Sand 17.5-18.9 240 17 5 0.96 Seed Not Liquefied 10 Sand 18.9-21 240 17 5 0.92 Seed Not Liquefied 11 Silt 21 -50 150 N.A. N.A. N.A. N.A. Not Liquefied 12 Silt 50-75 150 N.A. N.A. N.A. N.A. Not Liquefied 13 Silt 75-100 150 N.A. N.A. N.A. N.A. Not Liquefied 14 Silt 100-125 150 N.A. N.A. N.A. N.A. Not Liquefied 15 Silt 125-150 150 N.A. N.A. N.A. N.A. Not Liquefied 16 Silt 150-175 150 N.A. N.A. N.A. N.A. Not Liquefied 17 Silt 175-200 150 N.A. N.A. N.A. N.A. Not Liquefied Table 6.2 - Soil Parameters for Liquefaction Assessment at Location #2 79 CHAPTER 6 Results: Liquefaction Assessment Factor of Safety Against Liquefaction 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 1.10 1.20 Figure 6.3 - Factor of Safety Against Liquefaction (FSL) ~ Location #2 80 CHAPTER 6 Results: Liquefaction Assessment 6.2.1.2 Location #3 The comparison o f C R R ' s and C S R ' s is not shown for locat ion #3. C R R ' s cou ld not be estimated i n the plastic silts (which are predominant at this location). Layer Soil Type Depth (m) (Vs)i (m/s) (N,) w Fines Content (%) Criteria Status No CPT data over the first 3.4 meters 1 Silty S a n d 3 . 4 - 5 142 11 15 1 Seed Liquefied 2 S a n d 5 - 6 . 4 142 14 5 1 Seed Liquefied * Tunnel Invert Location (excavation ends at 6.4 meters below the original ground surface) 3 Silt 6.4-9.1 143 8 80 1 Seed Liquefied 4 Silt 9.1 -12 155 5 85 N.A. C h i n e s e Not Liquefied 5 Silt 12-17.9 150 5 85 N.A. C h i n e s e Not Liquefied 6 Sand 17.9-22 240 17 N.A. 0.93 S e e d Not Liquefied 7 Silt 2 2 - 4 0 150 N.A. N.A. N.A. N.A. Not Liquefied 8 Silt 4 0 - 6 0 150 N.A. N.A. N.A. N.A. Not Liquefied 9 Silt 6 0 - 8 0 150 N.A. N.A. N.A. N.A. Not Liquefied 10 Silt 80 -100 150 N.A. N.A. N.A. N.A. Not Liquefied 11 Silt 100-125 150 N.A. N.A. N.A. N.A. Not Liquefied 12 Silt 125-160 150 N.A. ' N.A. N.A. N.A. Not Liquefied 13 Silt 160-200 150 N.A. N.A. N.A. N.A. Not Liquefied Table 6.3 - Soil Parameters for Liquefaction Assessment at Location #3 81 CHAPTER 6 Results: Liquefaction Assessment Factor of Safety Against Liquefaction 0.20 0.40 0.60 0.80 1.00 1.20 1.40 Figure 6.4 - Factor of Safety Against Liquefaction -- Location #3 82 CHAPTER 6 Results: Liquefaction Assessment 6.2.1.3 Location #4 The comparison o f C R R ' s and C S R ' s is not shown for locat ion #4 because plastic silts were predominant at that location; therefore, the C R R ' s cou ld not be estimated. Factor of Safety Against Liquefaction I 1 I 1 I 1 I 1 I 1 | ' — | — i — | — i — | 0.40 0.50 0.60 0.70 0.80 0.90 1.00 1.10 1.20 15.0-Figure 6.5 - Factor of Safety Against Liquefaction (FSL) -- Location #4 83 CHAPTER 6 Results: Liquefaction Assessment Layer Soil Type Depth (m) (m/s) (NJso Fines Content (%) K 0 Criteria Status No CPT data over the first 3.5 meters depth 1 Sand 3 .5 -4 184 23 5 1 Seed Liquefied 2 Sand 4 - 6.25 160 22 5 1 Seed Liquefied 3 Sandy Silt 6.25 - 8 150 12 3 5 - 5 0 1 Seed Liquefied * Tunnel Invert Location (excavation ends at 8 meters below the original ground surface) 4 Sandy Silt 8 - 9 . 9 150 12 3 5 - 5 0 1 Seed Liquefied 5 Silty Sand 9 .9-10.5 160 20 15 1 Seed Liquefied 6 Sandy Silt 10 .5 -14 182 6 85 N.A. Chinese Not Liquefied 7 Sandy Silt 14 - 1 7 151 5 85 N.A. Chinese Not Liquefied 8 Sandy Silt 1 7 - 2 2 202 4 85 N.A. Chinese Not Liquefied 9 Sandy Silt 22-24.1 140 5 85 N.A. Chinese Not Liquefied 11 Silt 24 .1 -50 150 N.A. N.A. N.A. N.A. Not Liquefied 12 Silt 5 0 - 7 5 150 N.A. N.A. N.A. N.A. Not Liquefied 13 Silt 7 5 - 1 0 0 150 N.A. N.A. N.A. N.A. Not Liquefied 14 Silt 1 0 0 - 1 2 5 150 N.A. N.A. N.A. N.A. Not Liquefied 15 Silt 125 -150 150 N.A. N.A. N.A. N.A. Not Liquefied 16 Silt 1 5 0 - 1 7 5 150 N.A. N.A. N.A. N.A. Not Liquefied 17 Silt 175 -200 150 N.A. N.A. N.A. N.A. Not Liquefied Table 6.4 - Soi l Parameters for Liquefact ion A s s e s s m e n t at Locat ion #4 84 6.2.1.4 Location #7 CHAPTER 6 Results: Liquefaction Assessment Cyclic Stress Ratio 0.10 0.15 0.20 0.25 0.30 0.3 0.40 0.45 A' Figure 6.6 - Comparison of Cyclic Stress and Cyclic Resistance Ratios at Location #7 85 C H A P T E R 6 Results: Liquefaction Assessment Layer Soil Type Depth (m) (m/s) Fines Content (%) Criteria Status No CPT data over first 2.7 meters 1 Silty Sand 2.7-3.3 110 12 16 1 Seed Liquefied * Watertable located at 3.3 meters depth 2 Silty Sand 3.3-7.25 110 12 16 1 Seed Liquefied 3 Sand 7.25-11 110 9 6 1 Seed Liquefied 4 Sand 11 - 14 160 9 5 1 Seed Liquefied 5 Sand 14-17 160 9 10 0.98 Seed Liquefied 6 Silty Sand 17-20 140 7 10 0.96 Seed Liquefied * Tunnel Invert Location (excavation ends at 20 meters below the original ground surface) 7 Sand 20-24 220 11 9 0.9 Seed Not Liquefied 8 Sand 24-27 190 11 9 0.89 Seed & Robertson Not Liquefied 9 Sand 27-34 230 13 9 0.85 Seed & Robertson Not Liquefied 10 Silty Sand 34 - 38.3 170 10 52 0.8 Seed Not Liquefied 11 Silt 38.3-50 150 N.A. N.A. N A. N.A. Not Liquefied 12 Silt 50-75 150 N.A. N.A. N.A. N.A. Not Liquefied 13 Silt 75 - 100 150 N.A. N.A. N.A. N.A. Not Liquefied 14 Silt 100-125 150 N.A. N.A. N.A. N.A. Not Liquefied 15 Silt 125-150 150 N.A. N.A. N.A. N.A. Not Liquefied 16 Silt 150-175 150 N.A. N.A. N.A. N.A. Not Liquefied 17 Silt 175-200 150 N.A. N.A. N.A. N.A. Not Liquefied Table 6.5 - Soil Parameters for Liquefaction Assessment at Location #7 86 CHAPTER 6 Results: Liquefaction Assessment Factor of Safety Against Liquefaction I 1 I 1 I 1 I 1 I 1 I 1 I 1 I 1 I 1 I ' I 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 1.10 1.20 1.30 35.0 -Figure 6.7 - Factor of Safety Against Liquefaction (FSL) — Location #7 87 6.2.1.5 Location #8 CHAPTER 6 Results: Liquefaction Assessment Cyclic Stress Ratio I — 1 — I — 1 — I — 1 — I — 1 — I 1 I 1 I 1 I 1 I 1 I 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 5.00 — i 10.00 30.00 Liquefied C.S.R. C.R.R. Figure 6.8 - Comparison of Cyclic Stress and Cyclic Resistance Ratios at Location #8 88 CHAPTER 6 Results: Liquefaction Assessment Layer Soil Type Depth (m) (m/s) ( N ^ o Fines Content (%) Criteria Status No CPT data over the first 5 meters * Watertable located at 4.2 meters depth 1 S a n d 5 - 7 170 12 10 1 Seed Liquefied 2 S a n d 7 - 9 150 12 3 1 Seed Liquefied 3 Sand 9 - 1 1 170 12 3 1 Seed Liquefied 4 S a n d 11 - 12.8 170 12 4 1 Seed Liquefied 5 Silty sand 1 2 . 8 - 1 3 . 6 170 16 4 1 Seed Liquefied 6 Sand 1 3 . 6 - 1 6 190 11 4 1 Seed Liquefied * Tunnel Invert Location (excavation ends at 16 meters below the original ground surface) 7 Sand 1 6 - 1 8 190 11 4 0.98 Seed Liquefied 8 Sand 1 8 - 2 2 . 5 160 9 4 0.96 Seed Liquefied 9 Silty sand 2 2 . 5 - 2 3 . 3 170 8 30 0.92 Seed Liquefied 10 S a n d 23.3 - 25.2 170 11 4 0.91 Seed Liquefied 11 Silty sand 25.2 - 26.2 170 7 30 0.9 Seed Liquefied 12 S a n d 2 6 . 2 - 2 7 . 2 170 12 4 0.89 Seed Liquefied 13 Silt 27.2 - 50 150 N.A. N.A. N.A. N.A. Not Liquefied 14 Silt 5 0 - 7 5 150 N.A. N.A. N.A. N.A. Not Liquefied 15 Silt 7 5 - 1 0 0 150 N.A. N.A. N.A. N.A. Not Liquefied 16 Silt 1 0 0 - 1 2 5 150 N.A. N.A. N.A. N.A. Not Liquefied 17 Silt 1 2 5 - 1 5 0 150 N.A. N A N.A. N.A. Not Liquefied 18 Silt 1 5 0 - 1 7 5 150 N.A. N.A. N.A. N.A. Not Liquefied 19 Silt 1 7 5 - 2 0 0 150 N A N.A. N.A. N A Not Liquefied Table 6.6 - Soil Parameters for Liquefaction Assessment at Location #8 89 CHAPTER 6 Results: Liquefaction Assessment Factor of Safety Against Liquefaction i — i — i — i — i — i i i i i i i i i 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 1.10 1.20 30.0^ Figure 6.9 - Factor of Safety Against Liquefaction (FSL) -- Location #8 90 C H A P T E R 6 Results: Post-Liquefaction Stability 6.3 Post-liquefaction Stability — Flow Slides, Residual Strength (Sr), and Limiting Strains (Yum) Refer to section 4.3 for descriptions o f the procedures used i n the post-liquefaction stability analyses. The results i n this section should be reviewed i n conjunction w i t h the material i n section 7.3. Section 6.3.1 presents the results o f the post-liquefaction l imi t equ i l ib r ium analyses, and section 6.3.2 presents summaries o f two key post-liquefaction stress-strain parameters. Post-earthquake strength properties for each o f the locations are l is ted on the graphic output (figures 6.10 and 6.11). Tables 6.7 and 6.8 summarize the residual strength (S r ) and l imi t ing strain ( Y H m ) magnitudes for each l iquefied layer (at each location). 6.3.1 Flowslide Potential T w o separate types o f analyses were done. A t locations #7 and #8 (south and north r iver banks, respectively) the stability o f the over lying dykes was analyzed; the analyses correspond to the plane parallel to the axis o f the tunnel (i.e., longi tudinal-direction). The offshore locations, on the other hand, were analyzed i n the transverse section v iew. In the l imi t equi l ibr ium analyses at the river banks (locations #7 and #8), the failure surface was confined to the overlying dyke material (refer to figures 6.10 and 6.11). A lower bound residual strength was estimated for the dyke sands. A s figure 6.10 91 CHAPTER 6 Results: Post-Liquefaction Stability shows, the estimated residual strength o f the dyke material can be as l o w as 9 k P a at the south shore (location #7) and st i l l remain stable (under the F.S F L >1.3 criteria). A t locat ion #8 (north shore) the dyke material cou ld have a residual strength as l o w as 10 kPa . A p p e n d i x C . 2 contains graphic output using a more realistic residual strength estimate o f 26 kPa . Based on the results o f these analyses, it can be concluded that the north and south shores are stable. The graphic output from the f lowslide analyses at the offshore locations (#2, #3, and #4) are shown i n appendix C . 1. Since the tunnel is founded on a leve l excavation, the tunnel is very stable i n the direction transverse to the roadway (i.e., cross-section v iew) . Refer to appendix C . 1 for a description o f the results o f the analyses at the offshore locations. 92 93 6.3.2 Residual Strength (Sr) and Limit Strain (YuJ Summaries The summaries i n this section should be reviewed when analyzing the S O I L S T R E S S results ( in chapter 8). Res idual strength (S r ) and l im i t strain (ylim ) estimates for each o f the analyzed locations are summarized i n tables 6.7 and 6.8. The Location Material Type Material Number (Ni) f.« Factor of Safety Against Liquefaction ( F S , ) Su/a\„ Ylim (%) sr (kPa) #2 Sand 5 13 0.5 0.2 30 12 Sand 6 13 0.5 0.2 30 12 Si l t 7 Not App. N o t A p p . 0.4 27 40 #3 Sil t 5 Not App. N o t A p p . 0.4 27 30 #4 Si l t 5 Not App. N o t A p p . 0.4 27 30 #8 Sand 5 10 0.7 0.2 25 40 Sand 6 10 0.7 0.2 25 46 Sand 7 10 0.7 0.2 25 56 Sand 8 10 0.7 0.2 25 56 Sand 10 10 0.5 0.2 30 26 #2, #3,& #4 Sand 3 4 N o t Ava i l ab le 0.07 30 7 All Sand 4 9 " 0.5 0.18 30 11 Table 6.7 - Residual Strength (Sr) and Limiting Strain (Yl™) Estimates for Transverse Direction Analyses 95 CHAPTER 6 Results: Post-Liquefaction Stability material numbers corresponding to the l iquefied materials i n table 6.7 are graphical ly shown i n figures 6.12 and 6.13. The material numbers corresponding to the l iquefied materials i n table 6.8 are graphically shown i n figure 6.14. Material Number Material Type (Nl ) 6 0 Factor of Safety Against Liquefaction S u / o \ „ YLim (%) s,. (kPa) 2 Sil t N . A . N o t Ava i l ab le 0.4 27 30 3 Sand 4 N o t Ava i l ab le 0.07 30 7 4 Sand 10 0.7 0.2 25 46 6 Sand 9 0.5 0.18 30 11 10 Sand 10 0.5 0.2 30 26 13 Sand 10 0.7 0.2 25 56 Table 6.8 - Residual Strength (Sr) and Limiting Strain (YLim) magnitudes for Longitudinal Direction Analyses 96 CHAPTER 6 Results: Post-Liquefaction Stability i CD O D CO 0 to CO o O CM CO O O o O CO co a s CD s o 0 1 O o o o a o o o <D O O 97 CHAPTER 6 Results: Post-Liquefaction Stability 98 CHAPTER 6 Results: Post-Liquefaction Stability CHAPTER 7 DISCUSSION OF CHAPTER 6 RESULTS 7.1 Introduction This chapter presents interpretations o f the results that are summarized i n chapter 6. Interpretations focus on h o w the results compare w i th expectations, and why , i n some cases, they differ. The information i n each upcoming section should be reviewed w i t h the corresponding section i n chapter 6. 7.2 Liquefaction Assessment 7.2.1 Zones of Liquefaction The discussion material i n this section corresponds to the results o f section 6.2. Refer to section 4.2 for a description o f the liquefaction assessment procedures. In this section, the points o f discussion are segregated by headings. L) Methods of Liquefaction Assessment: The stratigraphy (soi l layers and thicknesses) vary along the length o f the tunnel. A s table 6.1 shows, depths to l iquefaction vary along the length o f the tunnel. D i r ec t ly beneath the tunnel, the soils comprise sands and silts. A s described i n section 4.2, the liquefaction resistance o f sands was based solely on the available C P T data. In plastic silts, the Chinese criteria (Wang, 1979) was used, whereas both penetration data and shear wave veloci ty data were used to assess liquefaction potential o f non-plastic silts. 100 ; C H A P T E R 7 Discussion of Chapter 6 Results Non-plas t ic silts were analyzed using both the Robertson (1990) and Seed (1984) criteria. The Robertson correlation is based on data from sands w i th fines contents ranging from 10% (Niigata) to 35%. It is assumed that the generally higher fines contents o f the silts at the tunnel site w i l l mean that the C R R estimates using the Robertson correlation are lower. The Robertson correlation is l imi ted because it is derived from a smal l data base. A s mentioned earlier, direct testing o f undisturbed samples is the best method for estimating the C R R o f a soi l . ii.) Soil Profile at Location #7: One additional point o f concern was highlighted by the C P T data at loca t ion #7. Because o f the homogeneous nature o f the sand at location #7, it was thought that it may be post-excavation backf i l l . B u t the Canadian Liquefact ion Exper iment ( C A N L E X ) test data ( in the v ic in i ty o f location #7) aren't as uniform, yet they're o f s imilar cone resistance; therefore, the — s a n d s were assumed to be natural. C A N L E X d id some C P T ' s and sample retrievals at the south end o f the tunnel. Though some o f the C A N L E X data is detailed, it was judged to be i n sufficient agreement w i t h C P T #91-7, so that there was no need to distinguish it i n this study. 101 CHAPTER 7 Discussion of Chapter 6 Results 7.2.2 Cyclic Loading and Spectral Response — SHAKE The discussion material i n this section corresponds to the results o f section 6.2. Refer to section 4.2.3 for a description o f the ground response analyses. i. ) Ground-Motion Amplification Factors: A p p e n d i x B . 4 contains a tabular summary o f the amount o f amplif icat ion o f ground motions from the soi l profile base to the surface. In general, there is a great deal o f amplif icat ion through the surface layers. There is significant decoupl ing o f ground motions at the interfaces between surficial loose sands and under lying silts. Ground-motions were amplif ied by a factor o f 1.5 to 2 from base to surface, i n most cases. These amplif icat ion factors are i n agreement w i th the Fraser Del ta ground response study b y Sy et al . (1991). Fo r instance, location #7 shows an increase i n predicted acceleration from 0.28g to 0.38g wi th in the top three surface layers. This generally occurs at the site, at locations where there is a transition from silt up to loose sand. ii. ) Soil Damping & Modulus Reduction: A s discussed i n section 4.2.3, to account for so i l damping and modulus reduction (versus shear strain), data published by Sy et al . (1991) for Fraser Del ta sands and silts was applied i n the S H A K E analyses. Ground response is strongly dependent on these parameters. F o r comparison purposes, S H A K E analyses were performed using other damping and modulus reduction estimations (Idriss 1990, Vuce t i c 1991, etc.), but they 102 CHAPTER 7 Discussion of Chapter 6 Results provided amplifications w h i c h were inconsistent w i th amplif icat ion factors publ ished by L o et al.(1991) and Byrne & Anderson (Fraser Del ta Task Force, 1991) for the Fraser Del ta . iii. ) Response Spectra: Figures B.3 .1 to B.3 .5 ( in appendix B.3) depict the surface spectral response at the five locations. Spectral shapes depend upon earthquake magnitude and site conditions. The frequency content o f the records used is also i l luminated by the shape o f the response spectra. Softer soils often increase the spectrum response values i n the long per iod range. In general, soils tend to attenuate l o w period motions and amplify h igh per iod ground motions. It has been estimated that the tunnel has a fundamental per iod o f vibrat ion i n the range o f 0.5 and 4.0 seconds (Ke r Priestmann, 1989), though it is most l i k e l y i n the longer per iod range. A s the figures show, amplif icat ion o f ground motions w i t h i n the lower per iod range is l ow. The possibli ty o f a magnified response (approaching resonance) between the tunnel and soi l is l imited. iv. ) Depth to Firm Ground: A s discussed i n section 5.3, firm ground (presumed to be Pleistocene) was encountered at a depth o f 200 meters i n a seismic reflection survey. Though many past ground response analyses i n the Fraser Del ta (i.e., W a l l i s , 1979; S y et al . , 1991, etc.) incorporate both bedrock and the Pleistocene wi th in the so i l co lumn estimate, it is k n o w n 103 CHAPTER 7 Discussion of Chapter 6 Results that S H A K E is not designed for simulating the movement o f shear waves through a dense layer (i.e., Pleistocene). Instead, it can be presumed that strong ground motions travel along the surface o f a dense layer, and then propogate up through the surficial layers (Byrne, 1994b). Consequently, it was considered appropriate not to use the bedrock depth estimate o f 700 meters i n this study. A shear wave veloci ty o f 3500 ft/s was applied i n the base material for a l l o f the S H A K E analyses. A graphic comparison o f results at locat ion #2 using velocities o f 1000 ft/s, 3500 ft/s, and 8000 ft/s is shown i n appendix B . 2 . A s expected, figure B.2 .1 shows that there is a noticeable increase i n peak parameters as the f i rm ground veloci ty estimate is increased. v.) Potential Structural Influence and Multi-dimension Dynamic Analyses: The effect o f the tunnel has not been incorporated i n the ground response analyses. It can be speculated that, since the relatively stiff tunnel cou ld not be inc luded as part o f the so i l columns, the peak parameter estimates are conservative. A n alternative w o u l d have been to use rigorous two- or three-dimensional dynamic analyses, but they are, i n general, complex and require extensive evaluation o f parameters. D u e to the lack o f detailed so i l data at this site, the use o f two-dimension analyses w o u l d have been inappropriate. Addi t iona l ly , since the tunnel is founded wi th in a shal low excavation i n the r iver bed, it was deemed unnecessary to account for tunnel-structure effects w i t h i n the free-field, so mult i -dimension dynamic analyses were considered unnecessary. 104 CHAPTER 7 Discussion of Chapter 6 Results Because S H A K E lacks the abil i ty to account for two- or three-dimensional effects, cyc l i c shear stress magnitudes are l i ke ly under-estimated. S H A K E ' S one dimensional method has been compared w i th a two dimensional approach (2-D F L U S H ~ Jong, 1988). Jong found that the one-dimensional analysis peak cyc l i c shear stress estimations were up to 15% lower than those o f the two-dimensional analysis. O n the other hand, it should be noted that when the tunnel was constructed, the excavation resulted i n a decrease i n the overburden stresses (below the tunnel). This change i n o ' v o cannot be accounted for i n the C P T results, so this contributes a little to the uncertainty. A decrease i n effective stress decreases the liquefaction potential o f a so i l , consequently, the under lying so i l ' s resistance to l iquefaction has been increased due to the placement o f the tunnel, whose average unit weight is approximately one-half that o f the or ig inal overburden soi l . 105 CHAPTER 7 Discussion of Chapter 6 Results 7.3 Post-Earthquake Stability ~ Flow Slides, Residual Strength, and Limiting Strains The discussion material i n this section corresponds to the results o f section 6.3. Refer to section 4.3 for a description o f the post-earthquake stability inputs and analyses. i. ) Estimation of Residual Strengths in Sands: A s described i n section 4.3.2, the pr incipal relation used to estimate residual strengths was the procedure suggested by Stark & M e s r i (1992). It should be noted that Stark & M e s r i (1992) stated that post-l iquefaction stability analyses cannot assume drainage and must be based on a constant volume cr i t ica l strength ~ as opposed to the assumption o f some other published relations, i n w h i c h drainage has obvious ly occurred, causing the peak strength to rise. Consequently, the Stark & M e s r i (1992) estimates tend to be lower than those estimates based on typ ica l case-history back-calculations. ii. ) Estimation of Residual Strength and Limiting Strains in Silts: A s described i n section 4.3.2, data from the B C H y d r o Transmiss ion Tower study (summarized i n appendix G.2) was used to estimate residual strengths i n silts. The data from sample #HSS1 indicates an Su/p ratio o f 0.4 at an effective confining stress o f 80kPa. This ratio was applied to obtain the results shown i n tables 6.7 and 6.8. Add i t iona l ly , stress-strain data (figure G.2.1 i n appendix G.2) was used to estimate the 106 CHAPTER 7 Discussion of Chapter 6 Results l imi t strain (the strain required to mobi l ize the residual strength) i n the silts. A n estimate o f Y i i m = 2 7 % was obtained from the figure. Add i t iona l ly , for comparison, the method proposed by Senneset et al.(1981) was applied to estimate undrained strengths i n the silts. C P T penetration resistances varied between 600 k P a and 900 kPa . So, using an average Nc'=15, the undrained strength estimates ranged between 40 k P a and 60 kPa ; therefore, this method was i n agreement w i t h the B C H y d r o data-based residual strength estimates. Furthermore, a lower-bound undrained strength o f 40 k P a was estimated using data from the Chatf ie ld D a m & B i r c h D a m sites (Duncan & Byrne , 1980). 107 CHAPTER 7 Discussion of Chapter 6 Results 7.4 Future Work i. ) Soil Data: Test data is l imited, and a larger data base w o u l d benefit any future studies. M o r e in-si tu testing on the upstream side o f the tunnel should be done. Preferably cone penetration tests ( C P T ) w o u l d be done at points between the 1991 C P T locations. T o conf i rm the val id i ty o f this study's assumption ~ that so i l layers can be extended hor izontal ly (i.e., to both sides o f the tunnel) ~ an adjacent C P T could be performed on the opposite side o f the tunnel. Addi t iona l ly , since the loose sand extends deeper as the north river bank is approached, another C P T should be done between locat ion #4 and the north river bank (location #8). Seismic shear wave veloci ty data should be collected. ii. ) Ground Response Analyses: The acceleration time-histories used i n this study are f rom the San Fernando strike-slip type earthquake. The anticipated major earthquake i n B r i t i s h C o l u m b i a is a subduction type event. A future assessment could be based solely on records from subduction events. 108 CHAPTER 8 RESULTS: POST-EARTHQUAKE DISPLACEMENTS 8.1 Introduction Results f rom the empir ica l and numerical method displacement analysis methods are summarized graphically and i n tabular form i n this chapter. It should be noted that, the results o f the various analyses are independently summarized i n this chapter, but are compared i n section 9.2 ( in chapter 9). A l l results are discussed i n chapter 9. E a c h o f the upcoming sections points out the corresponding section i n chapter 9 that should be referred to whi le reviewing the ind iv idua l result summaries i n this chapter. 8.2 Empirical Methods — Displacement Predictions Refer to section 4.4.2 for descriptions o f the procedures used i n the empir ica l method displacement analyses. Section 9.3 discusses the results i n this section, and appendix D contains detailed summaries o f the inputs used. Table 8.1 summarizes the lateral displacement predictions using the Bar t le t t -Youd (1992) and Hamada (1987) methods. The total l iquefied layer thickness at locat ion #2 is approximately 10 meters, and at locations #3 and #4, there is approximately 3 meters o f l iquef ied sandy silt. A t location #8, there is approximately 14 meters o f l iquef ied sand. B o t h sets o f analyses correspond to ground slopes o f 1% and 3%. The Bar t l e t t -Youd 109 CHAPTER 8 Displacement Results Location Bartlett - Youd (meters) Hamada (meters) i% 3% 1% 3% #2 0.5 0.8 3.4 4.8 #3 N o t App l i cab le N o t Appl icab le 1.1 1.5 #4 N o t App l i cab le N o t Appl icab le 1.1 1.5 #8 0.6 1.0 4.0 5.7 Table 8.1 - Empirical Method Displacement Estimates predictions are based on a horizontal epicentral distance o f 30 kilometers. These inputs are discussed i n section 4.4.2.2. A s the table shows, the Bar t l e t t -Youd predictions are m u c h lower than the corresponding Hamada estimates. Bar t le t t -Youd analyses were not done at locations #3 and #4 due to the high fines content o f the l iquefied soils. A s discussed i n section 4.4.2.4, the method developed by Tokimatsu and Seed (1987) was used to estimate liquefaction induced settlements. Table 8.2 summarizes the predictions. A s expected, the locations w i th the greatest depth to l iquefaction (i.e., locations #2 and #8) show the largest magnitudes. A s discussed i n section 4.4.3.2, the Tokimatsu/Seed correlation was used to estimate volumetric strains for the S O I L S T R E S S analyses. The estimates i n table 8.2 have been provided only as a reference for the S O I L S T R E S S vert ical displacement estimates, w h i c h include both movement due to undrained distortion, and consol idat ion 110 CHAPTER 8 Displacement Results due to excess pore-pressure dissipation. Location Cumulative thickness of Liquefied Layers (meters) Settlement (centimeters) (feet) #2 10 29 1.0 #3 2 6 0.2 #4 2 5 0.2 #8 14 35 1.1 Table 8.2 - Tokimatsu/Seed Method Post-Liquefaction Settlement Estimates 111 CHAPTER 8 Displacement Results 8.3 Numerical Methods — Displacement Predictions 8.3.1 Introduction Section 8.3 consists o f two general parts: summaries o f the S O I L S T R E S S predictions, fo l lowed by the L I Q D I S P predictions. B o t h sections 8.3.2 and 8.3.3 summarize the input parameters and the displacement estimates for each particular analysis. Section 8.3.2 contains graphic output and tabular summaries o f the lateral and vert ical S O I L S T R E S S displacement predictions for a l l five locations. The discussion material i n section 9.4 should be reviewed i n conjunction w i th the results i n this section. 8.3.2 Finite Element Method (SOILSTRESS) Displacements The procedures used to perform the S O I L S T R E S S analyses are described i n section 4.4.3. Results correspond to a peak ground veloci ty (v m a x ) o f 0.30m/s. The choice o f this value is discussed i n section 9.4.1. First, transverse-direction (cross-section) displacements at the north and south shores are summarized. Those analyses are fo l lowed by summaries for the offshore locations. A t the offshore locations, the sediment loading (on top o f the tunnel) is varied, and the effects o f increasing the peak ground veloci ty are analyzed, also. F ina l l y , displacements i n the plane parallel to the axis o f the tunnel ( ' longitudinal direction') are 112 CHAPTER 8 Displacement Results assessed ( in section 8.3.2.2). Section 8.3.3 summarizes the results o f the single-degree-of-freedom L I Q D I S P analyses. S O I L S T R E S S analyses showing the effects o f remedial measures are summarized and discussed i n chapter 10. 8.3.2.1 Transverse-Direction SOILSTRESS Analyses Figures 6.12 and 6.13 show the material numbering schemes for the tunnel cross-section at the onshore and offshore locations, respectively. Figures 8.3 to 8.15 show the displacement pattern and displacement vector graphics for the transverse section analyses. The displacement vector images have been provided to aid i n interpreting the displacement patterns. A s shown i n figures 6.12 and 6.13, " C I " and " C 3 " correspond to the top left and bottom right corners o f the tunnel's concrete section. The displacements at the two corners are provided to show the magnitude o f tunnel rotation and differential movement between the sides o f roadway. In the tabular displacement summaries, a negative vert ical displacement indicates a downward movement. Fo r each particular S O I L S T R E S S analysis, the pre- and post-earthquake parameters are presented i n tables that precede the corresponding displacement summaries. Table 8.3 presents inputs for materials w h i c h were applied i n a l l o f the transverse section analyses (Eg. concrete roadway). Figures 6.12 and 6.13 show the relative locations o f the materials. 113 CHAPTER 8 Displacement Results Because o f the seasonal sediment load variation, two separate analyses have been performed at each o f the offshore locations. The first (Case #1 — section 8.3.2.1.2) o f the two sets o f analyses corresponds to a case o f two meters o f sediment loading on top o f the tunnel, and the second case (Case #2 ~ appendix A . 1.1) corresponds to the condi t ion o f no sediment loading. All Locations Material Number Soil Type n K„ Rr (deg) A(f> c (kl'a) 1 Concrete 50000 (50000) 0.5 200000 0.25 0.5 50 0 1000 2 Compact Gravel 1200 (600) 0.5 1800 0.25 0.6 47 7 0 3 Liquefied Sand Fill 181 (0.1) 0.5 (0) 2000 0.25 0.9 (0) 32 (0) 0 0 4 Loose Sand 505 (0.4) 0.4 2000 0.25 0.8 (0) 33 (0) 0 0 10* Sand (over-hurHerrt 200 (0.9) 0.5 (0) 2000 0.25 0.8 (0) 33 (0) 0 0 Table 8.3 - SOILSTRESS Pre- and Post-liquefaction Inputs For Transverse (Cross-section) Analyses at All Locations Notes: - values in parentheses are post-earthquake estimates - * Material #10 is the sand overburden at the north and south dykes (i.e., Locations #8 and # 7, respectively) Case #3 (appendix A . 1.2) shows the effects o f increasing the sediment load on top o f the tunnel, and Case #4 (appendix A . 1.3) shows the effects o f increasing the peak ground veloci ty (v m a x ) from 0.30 m/s to 0.38 m/s. 114 CHAPTER 8 Displacement Results 8.3.2.1.1 Displacements at North and South Shores (Dykes) Tables 8.4 to 8.6 summarize input parameters for the S O I L S T R E S S analyses at the onshore locations. (Refer to figure 6.12 for the material numbering scheme). Figures 8.1 and 8.2 show the f ina l stratigraphies used i n the analyses. F o r post-earthquake conditions, a 5 0 % reduction i n stiffness was applied to a l l non-l iquefied materials due to the severity o f shaking (i.e., v m a x = 0.3 m/s). Table 8.4 contains key stiffness and strength parameters. Refer to section 6.3 for more detailed results o f post-liquefaction parameters. location (U>kc) Material Type Material Number Shear Modulus (Gm a iinkPa) (Initial) Yl.im (%) s,. i i i i i i (ksMfinal) #8 Sand 4 51170 505 30 i i 0.3 Sand 5 72170 712 25 40 1.6 Sand 6 51200 505 25 46 1.8 Sand 7 57770 570 25 56 2.2 Sand 8 57770 570 25 56 2.2 Table 8.4 - SOILSTRESS Inputs for Liquefied Materials at North Shore 115 CHAPTER 8 Displacement Results 116 Location #7 Material Number Soil Type K n K b m (dee) A<|> (deg) c (kPa) 5 Part ia l ly Liquefied Sand 955 (100) 0.5 2000 0.25 0.7 37 2 0 6 Part ia l ly Liquefied Sand 712 (100) 0.5 2000 0.25 0.7 37 2 0 7 Part ia l ly Liquefied Sand 1044 (100) 0.5 2000 0.25 0.7 37 2 0 8 Part ia l ly Liquefied Sand 1044 (100) 0.5 2000 0.25 0.7 37 2 0 9 Part ia l ly Liquefied Sand 570 (100) 0.5 2000 0.25 0.7 31 0 0 Table 8.5 - SOILSTRESS Pre- and Post-earthquake Inputs - Location #7 Note: values in parentheses are post-earthquake estimates Location #8 Material Soil Type n K b m 111 Ad) c Number (deg) (deg) (kPa) 5 Liquefied 712 0.5 2000 0.25 0.8 35 0 0 Sand (1.6) (0) (0) (0) 6 Liquefied 505 0.5 2000 0.25 0.8 35 0 0 Sand (1-8) (0) ( ° ) (0) 7 Liquefied 570 0.5 2000 0.25 0.8 35 0 0 Sand (2.2) (0) (0) (0) 8 Liquefied 570 0.5 2000 0.25 0.8 35 0 0 Sand (2.2) (0) (0) (0) 9 Si l t 747 0.5 2000 0.25 0.6 34 0 40 (374) Table 8.6 - SOILSTRESS Pre- and Post-earthquake Inputs - Location #8 117 CHAPTER 8 Displacement Results CHAPTER 8 Displacement Results Table 8.7 summarizes the horizontal and vertical displacement predictions at locations #7 and #8. The 'pre-consolidation vertical displacement' is the movement due to undrained distortion, and the '2-Dimension vertical displacement' is the total predicted vert ical movement i n w h i c h settlement due to dissipation o f excess pore-pressures is also included. The Tokimatsu/Seed method (1987) was applied (refer to section 4.4.2.4) to determine the volumetr ic strain inputs for the settlement calculations i n a l l o f the S O I L S T R E S S analyses! The 'total vertical displacement' ( in the f inal co lumn o f the table) represents the fu l l 3-dimensional effect by incorporating the undrained ver t ical distortion estimates from the longitudinal analyses. (Refer to section 8.3.2.2 for longi tudinal pre-consolidat ion estimates). Location Corner HORIZONTAL Displacement (meters) P r e -Consolidation Displacement (centimeters) 2-Dimension V E R T I C A L Displacement (centimeters) TOTAL (3-Dimension) V E R T I C A L Displacement (centimeters) #7 C I 0.021 -0.8 -5.9 -6.5 C 3 0.021 -0.6 -5.6 -6.2 #8 C I 0.82 7.9 -33.5 -31.3 C 3 0.84 14.5 -27.0 -24.8 T a b l e 8.7 - Disp lacements at N o r t h & South Shore Loca t ions Note: -A negative vertical displacement indicates a downward movement Figures 8.3 to 8.6 show the displacement pattern and displacement vector 119 121 •. CHAPTER 8 Displacement Results graphics for locations #7 and #8. The dashed lines represent the in i t i a l state, and the so l id lines show the post-earthquake posi t ion o f the nodes. The displacements are magnif ied by a factor o f two. A t locat ion #7, no liquefaction was predicted i n the under lying so i l units, but at locat ion #8, so i l units #5 to #8 liquefy. Loca t ion #7 (south river bank) shows very smal l horizontal and vertical movements because no liquefaction was predicted to occur be low the tunnel invert. The over lying loose sand (dyke) deforms noticeably, but it has little influence on the underlying movements. Refer to section 9.4.1.1 for a discussion o f the similarities between locations #7 and #8. Figure 8.5 shows the magnitude and pattern o f displacement at locat ion #8. A s summarized i n table 8.7, the horizontal movements are approximately 0.8 meters, and the total vert ical displacement is approximately 0.3 meters. The displacement pattern shows that the movement o f the concrete section is controlled b y the under ly ing uni t #5 and #6 sands. 122 CHAPTER 8 Displacement Result* CHAPTF.R 8 Displacement Results 8.3.2.1.2 Displacements at Offshore Locations The analyses i n this section are referred to as the 'Case #1 analyses' . Cases #2 and #3 are analyses w i t h different levels o f sediment on top o f the tunnel, and case #4 shows the effect o f increasing the earthquake ground veloci ty. The results o f the case #2 to #4 analyses are summarized i n appendix A . 1. Case #1 represents the most realist ic representation o f the current sediment conditions at the tunnel. Figures 8.7 to 8.9 show cross-section summaries for the location #2, #3, and #4 analyses. Tables 8.8 to 8.11 summarize input parameters for the analyses at the offshore locations. (Refer to figure 6.13 for the material numbering scheme). Table 8.11 contains key stiffness and strength parameters. Refer to section 6.3 for more detailed results o f post-l iquefaction parameters. Location #2 Material Number Soil Type n K b m (deg) A<b (deg) C fkPa) 5 Liquefied Sand 675 (0.4) 0.5 (0) 2000 0.25 0.7 (0) 35 (0) 0 0 6 Liquefied-Sand 505 (0.4) 0.5 (0) 2000 0.25 0.7 (0) 34 (0) 0 0 7 Liquefied Sil t 605 (1.5) 0.5 (0) 2000 0.25 0.7 (0) 34 (0) 0 40 8 Si l t 675 (338) 0.5 2000 0.25 0.7 34 0 40 9 Si l t 712 (356) 0.5 2000 0.25 0.6 34 0 40 T a b l e 8.8 - S O I L S T R E S S P r e - and Pos t -ear thquake Inputs ~ L o c a t i o n #2 125 CHAPTER 8 Displacement Remits 126 CHAPTER 8 Displacement Results 127 CHAPTER 8 Displacement Remits Location #3 Material Number Soil Type n K b m R f (deg) (deg) c (kPa) 5 Liquefied Silt 387 (1.1) 0.5 (0) 2000 0.25 0.7 (0) 34 (0) 0 30 6 Silt 474 (237) 0.5 2000 0.25 0.6 34 0 40 7 Silt 444 (222) 0.5 2000 0.25 0.6 34 0 40 8 Silt 444 (338) 0.5 2000 0.25 0.6 34 0 40 9 Dense Sand 712 (356) 0.5 2000 0.25 0.6 37 7 0 Table 8.9 - SOILSTRESS Pre- and Post-earthquake Inputs ~ Location #3 Note: values in parentheses are post-earthquake estimates Location #4 Material Number Soil Type •s n K„ m R f (deg) Act. (deg) c (kPa) 5 Liquefied Silt 444 (1.1) 0.5 (0) 2000 0.25 0.7 (0) 34 (0) 0 30 6 Silt 647 (323) 0.5 2000 0.25 0.6 34 0 40 7 Silt 450 (225) 0.5 '2000 0.25 0.6 34 0 40 8 . Silt 805 (402) 0.5 2000 0.25 0.6 34 0 40 9 Silt 738 (369) 0.5 2000 0.25 0.6 34 0 40 Table 8.10 - SOILSTRESS Pre- and Post-earthquake Inputs - Location #4 Note: values in parentheses are post-earthquake estimates 128 CHAPTER 8 Displacement Remits 129 CHAPTER 8 Displacement Remits Figures 8.10 to 8.15 show the displacement patterns and displacement vectors for the analyses at locations #2, #3, and #4. In a l l o f the figures, the displacements are magnif ied b y a factor o f two. The estimates for the Case #1 analyses are summarized i n table 8.12. Offshore Location Material T>pe Material Num her Shear Modulus (Gm„inkPa) (Initial) Ylim (%) s r (kPa) (k^ Mfinal) #2 Sand 5 68380 675 30 12 0.8 Sand 6 51170 505 30 12 0.8 Si l t 7 61300 605 27 40 1.5 #3 Si l t 5 39190 387 27 30 1.1 #4 Si l t 5 44990 444 27 30 1.1 All Sand (F i l l ) 3 18300 180 30 7 0.1 Sand 4 51170 505 30 11 0.3 Table 8.11 - SOILSTRESS Soil Input Parameters for Liquefied Materials at Offshore Locations A t locat ion #2, soi l units #5 to #7 are l iquefied, and at locations #3 and #4, soi l unit #5 is l iquefied. A s table 8.12 shows, the horizontal displacement o f the roadway at locat ion #2 is approximately 1 meter. The displacement pattern for locat ion #2 (figure 8.10) is uniform. The displacement vectors (figure 8.11) show the effect o f the compression o f the loose sand directly beneath the tunnel. The movement o f the unit #4 loose sand causes a slight undercutting o f the gravel backf i l l (unit #2), thereby pushing the concrete section sl ightly upward. 130 CHAPTER 8 Displacement Results Location Corner HORIZONTAL Displacement (meters) Pre-Consolidation Vertical Displacement (centimeters) 2-Dimension l i l l l l l l l l Displacement (centimeters) T O T A L (3-Dimension) VERTICAL Displacement (centimeters) #2 C I 0.96 8.9 -26.5 -22.6 C 3 0.96 8.3 -26.4 -22.5 #3 C I 0.24 -1.4 -14.4 -15.6 C 3 0.23 -1.7 -14.0 . -15.2 #4 C I 0.20 -1.3 -11.7 -17.7 C 3 0.19 -2.3 -12.8 -18.8 Table 8.12 - SOILSTRESS Displacements at Offshore Locations Note: -A negative vertical displacement indicates a downward movement Locat ions #3 and #4 show significantly lower displacements than locat ion #2. The only difference between the locations is i n the depth to liquefaction. E v e n though the adjacent l iquefied sand (unit #4) is the same i n a l l three analyses, the depth to l iquefaction apparently controls the amount o f movement o f the concrete section. In most o f the displacement pattern graphics there is a noticeable outward movement o f the elements at the left and right boundaries. Refer to section 9.4.1 for a discussion o f this phenomenon. 131 CHAPTER 8 Displacement Results o o C D CD O cz to Q rn o o c 05 o C u e S u "5. 5 • o o o CM o o LU j UOI|DA9|3 o o oo SJ E> 3 HO 132 133 CHAPTER 8 Displacement Results 135 CHAPTF.R 8 Displacement Results 136 CHAPTF.R 8 Displacement Results 137 r CHAPTER 8 Displacement Results Movements are smaller at locations #3 and #4. In the S O I L S T R E S S analyses, the total energy dissipated at each o f the three locations is the same, but more energy is dissipated i n the stronger non-liquefied units (#6 and #7) at locations #3 and #4. Refer to section 9.4.1 for a discussion o f differences i n kinetic energy dissipation at the offshore locations. The displacement estimates at the two corners show h o w m u c h differential movement occurs between the opposite sides o f the tunnel's concrete section. Differential movements are discussed i n section 9.4.1.1. i 138 CHAPTER 8 Displacement Results 8.3.2.2 Longitudinal-direction Displacements Displacements i n the plane parallel to the axis o f the tunnel ( ' longitudinal-direction') are summarized i n this section. Figure 6.1 shows the estimated so i l profi le along the length o f the tunnel, and the distance between the analyzed locations. Refer to section 9.4 for a discussion o f the S O I L S T R E S S results. Figures 8.16 to 8.19 show the displacement patterns and displacement vectors for the longitudinal-direction analyses. The dashed lines represent the in i t ia l state, and the so l id lines show the post-earthquake posi t ion o f the nodes. The displacements are magnif ied by a factor o f three. Pre- and post-earthquake inputs for the S O I L S T R E S S analyses are Longitudinal Section Analysis Material Number Soil Type K n K b m 4> (dee* Ad) c (kPa> 1 Concrete 50000 (50000) 0.5 200000 0.25 0.5 50 0 1000 2 Liquefied Silt 444 (1.1) 0.5 (0) 2000 0.25 0.7 (0) 34 (0) 0 30 3 Liquefied Sand Fill 181 (0.1) 0.5 (° ) 2000 0.25 0.9 (0) 32 (0) 0 0 5 Silt 444 (222-) 0.5 2000 0.25 0.6 34 0 40 Table 8.13 - SOILSTRESS Pre- and Post-liquefaction Inputs For Materials Common to Both Longitudinal Section Halves Note: - values in parentheses indicate post-liquefaction estimates summarized i n tables 8.13 to 8.15. For post-earthquake conditions, a 5 0 % reduction i n stiffness was applied to a l l non-liquefied materials due to the severity o f shaking (i.e., v, 139 • C H A P T E R 8 Displacement Rmfa = 0.3 m/s). Res idual strength and l imi t strain estimates that were applied i n the transverse-direction analyses were also applied i n the longitudinal analyses. (Section 6.3 summarizes the post-liquefaction parameters). Longitudinal Section Analysis Material Number Soil Type K s n K b m Rr <P (dee* Ad) (dee) C (kPa) 4 Liquefied Sand 505 (1.8) 0.5 (0) 2000 0.25 0.8 (0) 35 (0) 0 0 6 Silt 747 (374) 0.5 2000 0.25 0.6 34 0 40 7 Liquefied Sand 200 (0.5) 0.5 (°) 2000 0.25 0.8 (0) 33 (0) 0 0 8 , Sand (Sediment) 200 0.5 2000 0.25 0.6 33 0 0 9 Dense Sand Lens 1137 (568) 0.5 2000 0.25 0.6 37 7 0 10 Liquefied Sand 570 (2 2^ 0.5 2000 0.25 0.8 (0^ 35 (0) 0 0 Table 8.14 - SOILSTRESS Pre- and Post-liquefaction Inputs For North-End -Longitudinal Section Analysis Notes: - values in parentheses indicate post-liquefaction estimates Tables 8.16 and 8.17 summarize the displacements at the base o f the concrete section i n the v ic in i ty o f the five locations at w h i c h the transverse analyses were done. The 'pre-consolidation vertical displacement' is the movement due to undrained distortion, and the '2-Dimension vertical displacement' is the total vertical movement i n w h i c h settlement due to dissipation o f excess pore-pressures is also included. The 'total vert ical displacement' ( in the final co lumn o f each displacement summary table) represents the f u l l 3-dimensional effect by incorporating the undrained vert ical distortion estimates from 140 CHAPTER 8 Displacement Romh* Longitudinal Section Analysis Material Number Soil Type K g n K„ m Re (dee* A(J> (dee) C fkPa* 6 Liquefied Sand 505 (0.4) 0.5 (0) 2000 0.25 0.8 (° ) 33 (0) 0 0 7 Partially Liquefied Sand 955 (100) 0.5 2000 0.25 0.7 37 2 0 8 Partially Liquefied Sand 570 (100) 0.5 2000 0.25 0.7 31 0 0 9 Silt 747 (374) 0.5 2000 0.25 0.6 34 0 40 10 Liquefied Sand 200 (0.5) 0.5 (0) 2000 0.25 0.8 (° ) 33 (0) 0 0 11 Sediment (Sand) 200 0.5 2000 0.25 0.6 33 0 0 12 Dense Sand T.ens 1137 .(568V 0.5 2000 0.25 0.6 37 7 0 Table 8.15 - SOILSTRESS Pre- and Post-liquefaction Inputs For South-End ~ Longitudinal Section Analysis Notes: - values in parentheses indicate post-liquefaction estimates the onshore and offshore (i.e., Case #1) transverse analyses. (Refer to sections 8.3.2.1.1 and 8.3.2.1.2 for transverse-section pre-consolidation estimates). A s figure 8.16 shows, there is a noticeable outward movement o f the elements at the left boundary o f the northern section. This phenomenon is explained i n section 9.4. The vert ical displacement estimates at the five locations show h o w m u c h differential movement occurs along the length o f the tunnel. Differential movements are discussed i n section 9.4.1.2. 141 C H A P T E R 8 Displacement Remits The displacement vectors (figures 8.17 and 8.19) show the uneven effect o f the transitions from l iquefied to non-liquefied soils. For instance, the unit #7 sands i n the south section were not predicted to liquefy; consequently, the l iquef ied unit #6 and #2 soils experience slightly greater distortion as the horizontal ve loc i ty impulse is applied. A l l locations show small displacement at the roadway. In zones o f l iquefaction, more noticeable deformations occur, but those zones do not have m u c h effect on the movements o f the concrete section. The central boundary is f ixed i n the horizontal direct ion i n both analyses. Th i s f ixed condi t ion is meant to simulate the effect o f the northern and southern slopes counteracting each other horizontally. Refer to appendix A . 1.4 for a description o f the fu l l length longitudinal analysis. Location HORIZONTAL Displacement (centimeters) P r e -Consolidation Vertical Displacement (centimeters) 2-Dimension VERTICAL Displacement (centimeters) TOTAL (3-Dimension) VERTICAL Displacement (centimeters) #8 5.4 3.8 - 2 8 . 2 -20 .3 #4 3.3 - 0 . 4 - 13.6 - 15.9 #3 0.8 - 0 . 6 - 1 1 . 9 - 13.6 Table 8.16 - Northern Half - Longitudinal-direction Displacement Predictions Notej_ - A negative vertical displacement indicates a downward movement 1 4 2 CHAPTER 8 Displacement Remits Location HORIZONTAL Displacement (centimeters) Pre-Consolidation Vertical Displacement (centimeters) 2-Dimension V E R T I C A L Displacement (centimeters) TOTAL (3-Dimension) V E R T I C A L Displacement (centimeters) #3 0.3 0.7 - 10.0 - 11.7 #2 2.0 5.1 -31.2 -22.9 #7 0.03 -7.5 -28.2 -29.0 able 8.17 - Southern - half - Longitudinal-direction Displacement Predictions Notej_ - A negative vertical displacement indicates a downward movement 143 Displacement Results 144 145 146 147 C H A P T E R 8 Displacement Results 8.3.3 LIQDISP — Single-degree-of-freedom Displacements Refer to section 9.4.2 for a discussion o f the results summarized i n this section. Section 9.2 contains a comparison o f the L I Q D I S P results and the other displacement methods. The L I Q D I S P analysis procedure is discussed i n section 4.4.3.3. In a l l analyses, residual strengths and l imi t ing strains were estimated as described i n section 4.3. Location Liquefied Layer Thickness (meters) Residual Strength (Sr) ^ ^ ^ ^ ^ Limiting Strain (%) Factor of i l l l l l l l l Linear Modulus Prediction (meters) Non-Linear Modulus Prediction (meters) #2 10 12 30 1.63 4.18 5.47 #3 2.5 30 27 2.04 0.74 0.96 #4 2.5 30 27 1.90 0.79 1.03 Table 8.18 - LIQDISP: Input Parameters and Displacements at Offshore Locations L I Q D I S P ' s slope failure model was applied at the offshore locations. The slope failure model uses a l imi t equi l ibr ium stability factor o f safety to approximate the static dr iv ing stress, w h i c h is used to estimate the static displacement. A s table 8.18 shows, the estimations are influenced strongly b y l iquefied layer thickness. The so i l inputs i n table 8.18 show that, amongst the offshore locations the main difference is i n the l iquefied layer thickness; consequently, locat ion #2 shows significantly larger displacements than locations #3 and #4. 148 CHAPTER 8 Displacement Remit. Location Liquefied Layer Thickness (meters) Residual Strength (Sr) (kPa) Limiting Strain (%) Slope (%) Linear Modulus Prediction (meters) Non-Linear Modulus Prediction (meters) llllllllllll} 1.18 2.68 #8 16 57 30 l^iliilllil 1.91 3.61 3 2.70 4.36 4 3.52 5.02 Table 8.19 - LIQDISP: Input Parameters and Displacements at Onshore Location Since no liquefaction occurs be low the tunnel invert at locat ion #7, it was not analyzed using L I Q D I S P . L I Q D I S P ' s infrnite slope model was used to analyze the north shore (location #8), since the overlying level dyke material resembles an infinite slope. A s table 8.19 shows, the estimations are influenced strongly by the ground slope input. 149 CHAPTER 9 DISCUSSION OF CHAPTER 8 RESULTS 9.1 Introduction This chapter presents interpretations o f the results that are summarized i n chapter 8. Interpretations focus on h o w the results compare w i t h expectations, and why, i n some cases, they differ. The information i n each upcoming section should be reviewed w i t h the corresponding section i n chapter 8. 9.2 Comparison of Displacement Predictions Section 9.2.1 compares the displacement predictions from the empir ica l and numerica l methods. Section 9.2.2 brief ly provides a review o f past seismic performance o f underground structures (wi thin the context o f this study). 9.2.1 Comparison of Empirical and Numerical Method Predictions Table 9.1 summarizes the empir ical and numerical method horizontal displacement estimates at each location. The detailed inputs and results f rom a l l methods are independently summarized i n chapter 8 (the Bar t le t t -Youd and Hamada results are summarized i n table 8.1, the L I Q D I S P results are i n tables 8.18 and 8.19, and the S O I L S T R E S S results (for Case # 1) are compi led i n table 8.12). The Bar t le t t -Youd method was not applied at locations #3 and #4, where silts are predominant. The empir ical method results i n table 9.1 correspond to ground slopes o f 150 : CHAPTER 9 Discussion of Chapter 8 1% and 3%. B o t h linear and nonlinear modulus L I Q D I S P predictions are provided i n the table. The S O I L S T R E S S results represent the Case #1 (i.e., 2-meter sediment loading ~ refer to section 8.3.2.1.2) horizontal displacement prediction at corner #1 ( ' C I ' ) . Location Bartlett - Youd (meters) Hamada (meters) LIQDISP (meters) SOILSTRESS (Case #7) (meters) 1% 3% 1% 3% Linear !!lfSi!!!!ll Linear #2 0.5 0.8 3.4 4.8 4.2 5.5 0.94 #3 N o t A p p . N o t A p p . 1.1 1.5 0.8 1.0 0.18 #4 N o t A p p . N o t A p p . 1.1 1.5 0.8 1.1 0.19 #8 0.6 1.0 4.0 5.7 1% 1.2 2.7 0.82 3% 2.7 4.4 Table 9.1 - Horizontal Displacement Predictions from Empirical and Numerical M e t h o d s A t locat ion #2, there is approximately 6 meters o f l iquef ied sand and 4 meters o f l iquefied silt be low the tunnel invert. A t locations #3 and #4, there is approximately 3 meters o f l iquefied silt beneath the tunnel. Loca t ion #7 was not analyzed using L I Q D I S P or the empir ical methods because liquefaction was not predicted to occur beneath the tunnel. A t locat ion #8, there is a thick segment (approximately 18 meters) o f l iquef ied sand beneath the tunnel. U p o n comparing the results, it should be noted that, since the empir ical methods 151 CHAPTER 9 Discussion of Chapter 8 are derived from case-histories, they are generally meant to provide a range w i t h i n w h i c h the numerical predictions should l ie . A s table 9.1 shows, the Bar t le t t -Youd results compare w e l l w i th the S O I L S T R E S S results, and the Hamada results are i n good agreement w i t h the L I Q D I S P predictions. In table 9.1, the L I Q D I S P predictions at locat ion #8 correspond to ground slopes o f 1% and 3%. The relatively h igh nonlinear modulus L I Q D I S P predictions compare w e l l w i t h the Hamada values at corresponding ground slopes. A t a l l locations, the Hamada predictions are approximately 5 times as large as the corresponding S O I L S T R E S S estimates. Since the Hamada calculation is strongly dependent upon l iquefied layer thickness, its predictions at locations #2 and #8 are high. S imi la r ly , when using L I Q D I S P ' s slope failure model (i.e., i n wh ich a l imi t equi l ibr ium stability factor o f safety is used to estimate the static shear stress) at the offshore locations, the estimations were strongly dependent upon liquefied layer thickness. A s the so i l inputs i n table 8.18 show, amongst the offshore locations the main difference is i n the l iquefied layer thickness; consequently, locat ion #2 shows significantly larger displacements than locations #3 and #4. Since the surficial deposits are susceptible to l iquefaction (upon the 475-year earthquake), it was important to be able to include the structural influence (i.e., stiffness o f the tunnel) wi th in the deformation analyses. The s impli f ied methods were l imi ted by their inabi l i ty to acknowledge the amount o f submergence o f the tunnel at the offshore 152 CHAPTER 9 Discussion of Chapter 8 locations; consequently, those methods should be cautiously acknowledged i n a study such as this. A s discussed i n section 4.4.3.2, the S O I L S T R E S S analyses were the only ones i n w h i c h the tunnel cou ld be included. Because S O I L S T R E S S properly considers the 2-dimensional nature o f the problem, the results can be considered more accurate than those obtained by the s implif ied methods. L I Q D I S P handles the 2-dimensional nature o f the problem using a "trick" similar to that o f Newmark (1965), whereas S O I L S T R E S S captures the 2-dimensional response, providing a range o f displacements over a discretized (finite element) grid. L I Q D I S P models the embankment as a single-degree-of-freedom system and assumes uniform block movement along a failure surface, thereby provid ing a single max imum displacement. In this study, it was presumed that the m a x i m u m displacement ( L I Q D I S P prediction) occurs i n the l iquefying layers beneath the tunnel. It should be noted that the Hamada equation was derived from data on clean medium-grained sands, and earthquake magnitudes o f M = 7.5 and 7.7. Y o u d et al . (1988) compared historical earthquake-induced displacements from earthquakes i n Japan and the Un i t ed States, and found that displacements i n Japan tended to be larger than those i n the U . S . The generally coarser and cleaner sands i n Japan were hypothesized to be the cause o f the difference. Because the Hamada data base consists o f such sands, the predictions provided for the sandy silt l iquefied layers at the #3 and #4 locations should be acknowledged as being very conservative (low). The estimates have been provided for 153 CHAPTER 9 Discussion of Chapter 8 reference purposes only. 9.2.2 Underground Structure Case-histories In general, underground structures are less severely affected b y seismic motions than surface structures (Owen & Schol l , 1981). The response o f the tunnel w i l l be dependent upon post-earthquake stiffness o f the foundation soils. I f the so i l does not l iquefy, then the tunnel can be expected to deform according to the free-field motions o f the so i l . This phenomenon has been documented i n post-earthquake analyses o f other underground structures (i.e, subways, pipelines, foundation piles). If, on the other hand, the so i l liquefies, then the stiffness o f the tunnel w i l l have more o f an influence on the extent o f deformation. It should be noted that the submerged port ion o f the tunnel comprises six segments, so the connections between those segments w i l l influence the movement. In the event o f an earthquake smaller than the one addressed i n this study, the surface soils may retain their integrity. In that case, the stiffness o f the tunnel may reduce the so i l stiffness-dependent free-field motions. Fo r example, the Trans-Bay subway i n San Francisco was estimated to have reduced the free-field motions w i t h i n the soft surface clays by approximately 15% (Kuesel , 1969). 154 CHAPTER 9 Discussion of Chapter 8 9.3 Empirical Method Displacements This section consists o f two separate discussions. Discuss ion o f the Bar t l e t t -Youd (1992) and Hamada (1987) methods is fo l lowed by the Tokimatsu-Seed method. In this section, the indiv idual points o f discussion are segregated b y headings. i.) Bartlett-Youd and Hamada Methods: The discussion material i n this section corresponds to the results o f section 8.2. Refer to section 4.4.2 for a description o f the empir ical method procedures. The Hamada (1987) predictions (summarized i n table 8.1) are relat ively high. Hamada 's method is commonly acknowledged for overestimating displacements. The Bar t l e t t -Youd (1992) approach proved to be sensitive to fines contents ( F 1 5 ) and distance (R) from the earthquake source. A higher fines content decreases a predict ion significantly. Contained i n appendix D . 1 are estimates using an earthquake distance o f 60 kilometers. These displacements were approximately one-fifth those obtained us ing a distance o f 30 kilometers. The magnitudes i n table 8.1 were estimated using an earthquake distance o f 30 kilometers. This distance was applied to maintain consistency w i t h the attenuation distance input w h i c h was used to develop the target spectrum for the ground response analyses (refer to section 3.2.1). ( A l l empir ical method inputs and calculations are provided i n appendix D ) . T o estimate post-earthquake displacements, both methods require l iquef ied layer 155 CHAPTER 9 Discussion of Chapter 8 data. The l iquefied layers acknowledged i n the analyses were those that were be low the tunnel invert. The Bar t le t t -Youd and Hamada methods a l low each l iquef ied layer to be treated separately, so that the displacements i n each layer can be summed. The loose sand layers adjacent to the excavation (i.e., material #4 i n figure 6.12) and the pump sand f i l l (material #3 i n figure 6.12) were not acknowledged i n the empir ica l analyses because they are presumed to have little effect on the magnitude o f the tunnel displacements. (This is verif ied i n the remediation analyses i n chapter 10). The sand f i l l (material #3) w h i c h was pumped i n beneath the tunnel wasn't included i n the analyses because it is constrained b y the coarse gravel and rockf i l l at the sides o f the excavation. Since the l iquef ied layers at locations #3 and #4 consist o f sandy silts containing fines contents ( F 1 5 ) beyond the recommended l imi t (F 1 5 =50%) o f the Bar t l e t t -Youd data base, these locations were not analyzed using the Bar t le t t -Youd method (1992). Since locat ion #7 doesn't show any liquefaction be low the tunnel invert, i t wasn't possible to analyze it using either o f the empir ical methods. But , the analysis results o f locat ion #8 can be applied to location #7, i n the event l iquefaction does occur at locat ion #7. (The similari t ies between locations #7 and #8 are discussed i n section 9.4.1). The Bar t le t t -Youd and Hamada method predictions were derived using ground slopes o f 1% and 3%. Al though the tunnel is constructed on a level excavation, the under ly ing l iquefied stratum are assumed, to be sloping approximately paral le l to the r iver bed. The river bed slope varies considerably i n both the north-south and east-west 156 CHAPTER 9 Discussion of Chapter 8 directions, but estimates o f downstream stratum slope at the tunnel site were obtainable us ing seismic reflection survey data (Monaghan, 1995). The reflections indicate approximate dips varying between 1% and 3%. The Bar t le t t -Youd model is advantageous over the Hamada mode l i n that it takes into account earthquake and so i l parameters as w e l l as the topographical and geological factors. But , the use o f either o f these methods at the offshore locations is questionable because o f the inabi l i ty to account for the amount o f submergence, and the structural influence w i t h i n the surficial soi l . The ground displacements computed us ing the s impl i f ied formulae are approximate, and are va l id for free-field conditions on ly (i.e. without interaction w i th the tunnel). The Bar t le t t -Youd model was developed from Japanese and western U.S. data, so it is most applicable to regions i n w h i c h significant ground mot ion attenuation can be expected to occur. This may explain the relatively l o w estimates obtained using this method. Furthermore, none o f the case histories used to derive this mode l showed l iquefied layers deeper than 15 meters; there is an 18-meter thick segment o f l iquefied sand at locat ion #8. Addi t iona l ly , the case histories showed a depth to the top o f the l iquefied layer w h i c h was generally wi th in a few meters o f the ground surface. 157 C H A P T E R 9 Discussion of Chapter 8 ii.) Tokimatsu/Seed Post-liquefaction Settlements: A s discussed i n section 4.4.2.4, the method developed by Tokimatsu and Seed (1987) was used to estimate liquefaction-induced settlements. The post-l iquefaction volumetric strain inputs for the S O I L S T R E S S analyses were estimated using the Tokimatsu/Seed correlation. Addi t iona l ly , table 8.2 provides a summary o f the estimated post-liquefaction settlements based on the Tokimatsu/Seed methodology. These values are provided only as a reference, since the pre-consolidation S O I L S T R E S S vert ical displacement estimates indicate that noticeable vertical movements do occur at a l l o f the locations before dissipation o f pore pressures. The use o f these estimates is l imi ted because partial buoyant uplift o f the tunnel (upon generation o f excess pore pressures) can not be accounted for. It should be noted that at locations #3 and #4, the l iquef ied material is silt. Add i t iona l ly , 4 meters o f the l iquefied material at location #2 is silt. The Tokimatsu/Seed method was developed for use i n l iquefied sands, so the tabulated predictions (at those locations) should be acknowledged cautiously. 158 CHAPTER 9 Discussion of Chapter 8 9.4 Numerical Method Displacements The discussion material i n this section corresponds to the results o f section 8.3. Section 9.4.1 discusses the general factors common to a l l (i.e., transverse and longitudinal) S O I L S T R E S S analyses i n this study. Sections 9.4.1.1 and 9.4.1.2 discuss the specific transverse- and longitudinal-direction analyses. Section 9.4.2 discusses the L I Q D I S P analyses. In each section, the indiv idual points o f discussion are segregated by headings. 9.4.1 Finite Element (SOILSTRESS) Displacements i.) Outward Movement of Elements at Left and Right Boundaries: In a l l S O I L S T R E S S analyses there was some degree o f outward movement o f elements at the left and right boundaries. Because a f ixed boundary condi t ion w o u l d not properly represent conditions i n the field, the nodes at those boundaries were, instead, constrained by forces equivalent i n magnitude to i n situ horizontal so i l pressures. (Hor izonta l pressure boundaries are also discussed i n section 4.4.3.2). The displacement vector graphics ( in chapter 8) show the boundary effect, w h i c h is due to insufficient lateral resistance required to confine the vertical seismic coefficient ( k j force applicat ion on the softened soi l . Fo r comparison, the boundary forces were increased at two o f the offshore locations, but the h igh pressure magnitudes caused unacceptable distortions at the mesh boundaries. Al though the displacement estimates at the nodes close to the left 159 C H A P T E R 9 Discussion of Chapter 8 and right boundaries are inaccurate, the cri t ical central port ion o f the mesh is unaffected (i.e., a large enough mesh wid th was chosen to negate boundary effects). ii.) Application of Horizontal (k^) and Vertical (kj Seismic Coefficients: A l l S O I L S T R E S S analyses were done using the method i n w h i c h both vert ical and horizontal seismic coefficients can be applied (i.e., 'opt ion 0'). (The use o f seismic coefficients, k,, and k H , to provide an additional force (AF) to satisfy the energy balance is discussed i n appendix A.3). A s discussed i n section 4.4.3, seismic-induced displacements occur due to: i . ) softening caused by liquefaction, and i i . ) kinet ic energy due to the earthquake-induced veloci ty pulse. Displacements due to stiam-softeriing are best computed using a {AF} based on a vert ical seismic coefficient (k,,), since it is main ly the (vertical) gravitational force that causes the displacements. The displacements due to the veloci ty pulse are best computed using a {AF} based on a horizontal seismic coefficient ( k ^ , since the ve loc i ty is assumed to be i n the hor izontal direction. W h e n using option 0, first, 1^  is used to compute displacements that satisfy the convergence criterion for overall energy balance o f the system due to stram-softening (i.e., due to gravity and boundary forces acting on the softened soil) . Then, the additional displacements due to kinetic energy are accounted for by apply ing k H . The net work ( N . W . ) is estimated as: N . W . = (Wint -Wext ) /Wext . W h e n N . W is 160 CHAPTER 9 Discussion of Chapter 8 negative, the seismic coefficient must be increased to increase the internal work (i.e., increase the strains) to achieve the energy balance o f the system. The longitudinal profile was analyzed as two separate halves. The longi tudinal analyses are described i n section 8.3.2.2. One o f the points o f concern when assessing the va l id i ty o f the S O I L S T R E S S results is i n the horizontal displacement estimates when the profile is analyzed as one continuous section. W h e n the profi le is analyzed as such, the whole system ' lurches ' i n the direction o f the horizontal ve loc i ty impulse. (The results from the full-section longitudinal analyses are presented i n appendix A . 1.4). In reality, this simultaneous movement w i l l not l i ke ly occur. The vert ical movements, on the other hand, can be considered reliable. T o alleviate concerns about the va l id i ty o f using option 0, comparison analyses were performed for a l l S O I L S T R E S S analyses (i.e., longi tudinal and transverse sections) using the options i n w h i c h on ly one o f the two seismic coefficients is applied (i.e., 'opt ion 1' for horizontal seismic coefficient only, and 'op t ion 2' for vert ical coefficient only) . W h e n k H alone (option 1) was applied, the horizontal displacements were very similar to the horizontal magnitudes estimated when using option 0, and, as expected, the vertical displacement estimates o f option 1 were very low. S imi la r ly , when 1^ alone (option 2) was applied, the vert ical displacements were very s imilar to the vert ical magnitudes estimated when using option 0. So, i n effect, the vert ical and horizontal S O I L S T R E S S displacements (estimated i n this study) can be distinguished somewhat, i n the event o f a future comparison o f displacement estimates 161 CHAPTER 9 Discussion of Chapter 8 using a different method such as F L A C (Fast Lagrangian Ana lys i s o f Continua). Add i t iona l ly , the stratigraphy was taken to only 18 meters be low the tunnel (at its lowest point ~ i.e., halfway across the river) i n the S O I L S T R E S S analyses because o f the system mass. I f the so i l depth was increased, the system mass w o u l d increase, and this w o u l d result i n an increase i n the kinetic inertia o f the system (when the earthquake impulse is applied). Since external work is proportional to the displacements, the displacements w i l l have to be too large to achieve an energy balance. The work-energy concepts are discussed i n appendix A . 2 . Hi) Validation of the SOILSTRESS Method: The S O I L S T R E S S finite element analysis has been found to give exact agreement w i th N e w m a r k when a rigid-plastic single-degree-of-freedom system has been assumed (Byrne et a l . , 1992). Addi t iona l ly , i t corroborates w e l l w i t h the observations o f Hamada et al . (1987); furthermore, it predicts the failure o f the L o w e r San Fernando dam, and provides accurate displacement predictions o f pattern and magnitude o f deformations for the Uppe r San Fernando dam (Byrne et a l . , 1992). 162 CHAPTER 9 Discussion of Chapter 8 9.4.1.1 Transverse-direction Analyses i.) Comparison of Soil Response at the South and North Shores (Locations #7 & #8): The liquefaction assessment for location #7 indicated that the sands were 'par t ia l ly- l iquefied ' (i.e., they could experience noticeable softening). Since the stratigraphies at locations #7 and #8 consist o f very similar so i l types and thicknesses, the response at locat ion #7 w o u l d have been nearly identical to that o f locat ion #8; therefore, it was considered most expedient to do t h e S O I L S T R E S S analyses at locat ion #7 w i t h contrasting (i.e., more stiff) post-earthquake shear modulus inputs to those o f location #8. The response could then be considered representative o f h o w both the north and south shores w o u l d perform should a smaller earthquake occur (and the under ly ing sands do not l iquefy to the degree predicted at locat ion #8). O n the other hand, the predicted displacements at locat ion #8 cou ld be considered applicable to those at #7 should an event equal to the design earthquake be considered. Consequently, it is recommended that the displacement predictions for location #8 be conservatively considered applicable to locat ion #7. ii. ) Varying Loading Conditions (Cases #1 to #3): A p p e n d i x A . 1.2 ( 'Case #3') summarizes the results o f analyses at loca t ion #2 w i t h increased sediment loading on top o f the tunnel. These analyses were carried out for two reasons: 163 : CHAPTER 9 Discussion of Chapter 8 i . ) the depth o f sediment on top o f the tunnel varies w i t h season, and i i . ) increasing or decreasing the sediment load was presumed to be a potential remedial measure (to decrease displacements, or prevent buoyant uplift). Displacement Case #1 Case #2 Case #3a HORIZONTAL (m) 1.0 1.0 1.1 VERTICAL (cm) -23 -20 -25 Table 9.2 - Comparison of SOILSTRESS Results - Varying Sediment Loads Table 9.2 summarizes the displacement estimates for Cases #1 to #3. A s outlined i n section 8.3.2, Case #1 represents a 2-meter r iver sediment load, Case #2 represents the removal o f the sediment, and Case #3 a is the 4-meter sediment load. A s expected, when the sediment load increases, the displacement predictions increase slightly, due to the increase i n the gravity-driven deformations o f the l iquefied so i l . Since the differences are small , and buoyant uplift o f the tunnel was not significant, there is no need to perform any such remediation (i.e., removal or addition o f sediment). iii.) Increased Earthquake Ground Velocity (Case #4): Peak ground accelerations estimated by S H A K E varied from 0.32 m/s to 0.38 m/s. A n A / V ratio o f 1 was applied at this site (Byrne, 1994), where ' A ' is the acceleration i n gravity units and ' V is the max imum veloci ty i n m/sec. Consequently, a peak ground veloci ty ( v m a x ) o f 0.30 m/s was applied i n a l l transverse and longitudinal S O I L S T R E S S analyses. The m a x i m u m veloci ty parameter, v m a x , is used to compute the kinet ic inertia 164 CHAPTER 9 Discussion of Chapter 8 o f the system. (Refer to appendix A . 2 for a description o f the incorporat ion o f kinetic inertia w i th in the Extended Newmark procedure). Since a l l the l iquefied so i l cannot be expected to trigger at the same time, it was considered appropriate to sl ight ly decrease the ground veloci ty input from the maximums (i.e., v m a x = 0.32 m/s to 0.38 m/s) inferred from S H A K E . Fo r comparison, location #2 was analyzed wi th an increased ve loc i ty input o f v m a x = 0.38 m/s. The results o f that analysis are summarized i n appendix A . 1.3 ( 'Case 4'). A s expected, horizontal displacements increased proport ionally to the increase i n the veloci ty input. iv. ) Differences in Kinetic Energy Dissipation: Amongs t the offshore locations, locations #3 and #4 show smaller movements than locat ion #2. There is an 8-meter thick segment o f l iquefied soils be low the tunnel at locat ion #2; the other two locations have approximately 3 meters o f l iquef ied so i l beneath the tunnel. The total kinetic energy to be dissipated is approximately the same for a l l three cases, but more energy is dissipated i n the stronger non-l iquefied materials at locations #3 and #4; consequently, the displacements are smaller ~ since less energy is dissipated i n the surrounding and underlying l iquefied soils. v. ) Differential Movements in Transverse Direction: The S O I L S T R E S S results depict differential movements along the w id th o f the tunnel. The displacements at opposite corners ( ' C I ' , and 'C3 ' ) are provided to highlight 165 C H A P T E R 9 Discussion of Chapter 8 the differential movements at the vi ta l concrete roadway section o f the tunnel. These differing levels o f movement are due to variations i n the surficial geology. Locat ions and thicknesses o f l iquefied layers, as w e l l as variations i n residual strength and stiffness are key causes o f the uneven movements. This proves to be a key advantage o f using the S O I L S T R E S S finite element method ~ post-earthquake stress-strain variations can be accounted for, i n addition to the irregularities i n geometry and geology. 9.4.1.2 Longitudinal-direction Analyses Differential Movements: Differential movements are accounted for i n the longitudinal S O I L S T R E S S analyses. These differential movements are very important when considering post-earthquake structural effects, especially at the connections between the subaqueous tunnel elements. W h e n the pre-consolidation vertical movements from the Case #1 analyses are added to the total ( ' 2 - D ' ) longitudinal vertical displacement, the ( ' 3 - D ' ) vert ical displacement predictions are very close i n magnitude. Loca t ion #3 shows a vert ical displacement w h i c h is about 9 centimeters lower than the other locations, yet the closest locat ion (#4) is 140 meters away; consequently, the differential value is o f little significance i n terms o f post-earthquake structural stability. I f the Case #1 transverse analysis vert ical displacement estimates are not added to the longitudinal predict ion, then the largest differential movement occurs between locations #3 and #2. The two locations 166 . CHAPTER 9 Discussion of Chapter 8 are approximately 220 meters apart, and location #2 moves approximately 20 centimeters (8 inches) farther down than location #3. Again, this magnitude of differential displacement is insignificant when considering the distance between locations. 167 ' C H A P T E R 9 Discussion of Chapter 8 9.4.2 LIQDISP - Single-Degree-of-Freedom Analyses i.) General Discussion: L I Q D I S P is a simple method o f analysis that incorporates the Extended N e w m a r k methodology (Byrne, 1990). The closed form L I Q D I S P code is ideal for simple (infinite) slopes. F o r complex geometries, the program is l imi ted to predict ing free-field movements. A l though L I Q D I S P is a simple program, it is advantageous over many procedures since it incorporates a model w h i c h takes into account nonlinear stress-strain behaviour o f so i l . (Refer to appendix A . 2 for a description o f the Extended N e w m a r k model) . Because o f the s implif ied nature o f the program, it is not possible to include the effect o f the tunnel interaction w i th the crust. The properties o f the non-l iquefied surface (crust) layers were averaged and input as one equivalent layer, and the same technique was applied to the underlying l iquefied layers. ii. ) Analysis of Offshore Locations: One o f the difficulties i n using L I Q D I S P at the offshore locations was the inabi l i ty to acknowledge the water table location. Addi t iona l ly , the weight o f the over ly ing water cou ld not be input, unless some form o f averaging (i.e., additional weight) was incorporated i n the crust mass. The offshore locations (#2, #3, #4) were analyzed using the slope failure model , i n 168 C H A P T E R 9 Discussion of Chapter 8 w h i c h a l imi t equi l ibr ium stability factor o f safety is used to approximate the static dr iv ing stress. The displacements are summarized i n table 8.18. O n l y a non-liquefied ('crust') material, and an underlying l iquefied material can be input, regardless o f the interbedded nature o f a soi l column. The method considers the crust (i.e., non-l iquefied layer) and liquefied so i l to be a single-degree-of-freedom elastic-plastic system. A l l l iquefied layer thicknesses are accumulated, and then average so i l parameter inputs must be used to represent the l iquefied layer. Loca t ion #2 comprises 6 meters o f l iquefied sand and 4 meters o f l iquefied silt. W h e n preparing the L I Q D I S P analyses, the sand and silt were combined, and average residual strength (S r ) and b l o w count magnitudes were applied to create one l iquefied layer. The discrepancy between the silt and sand values wasn ' t significant, hence the L I Q D I S P estimates at locat ion #2 can be considered crudely applicable. It should be noted that the loose sand layers adjacent to the excavation (i.e., material #4 i n figure 6.12) and the pump sand f i l l (material #3 i n figure 6.12) were not applied i n the L I Q D I S P analyses because they are presumed to have little effect on the magnitude o f the tunnel displacements. iii.) Analysis of Onshore Locations: Since no l iquefaction occurs be low the tunnel invert at locat ion #7, it was not analyzed us ing L I Q D I S P . The north shore (location #8) was analyzed as an infinite 169 CHAPTER 9 Discussion of Chapter 8 slope, since the over lying dyke material resembles a continuous slope. Slopes varying from 1% to 4 % were analyzed. Results and input parameters are summarized i n table 8.19. Since liquefaction o f the over lying dyke sands w i l l not have m u c h influence on the underlying displacements, the thickness o f that segment o f l iquefied sand was inc luded as part o f the crust (i.e., non-liquefied layer) thickness (t c) to account for the added inertia. 170 CHAPTER 10 REMEDIAL MEASURES 10.1 Introduction This chapter presents the results o f S O I L S T R E S S analyses that simulate the effects o f ground improvement on post-earthquake displacements. The locat ion #2 analyses i n section 8.3.2.1.2 should be referred to when reviewing the information i n this chapter. Potential remedial options were analyzed at location #2 because it is the offshore locat ion determined to be the most susceptible to liquefaction. A s table 8.15 shows, horizontal and vert ical displacements for the original configuration were approximately 1 meter and 0.2 meters, respectively. (Figures 8.10 and 8.11 graphical ly depict the movements). The zones o f improvement represent so i l w h i c h has been densified. A s w i t h other non-l iquefied materials, only a two-fold degradation i n post-earthquake stiffness o f the densified zones was applied i n these analyses. S o i l units #3 to #7 are l iquefied. The pre-and post-earthquake so i l properties o f a l l other materials were not changed from their or iginal values. Tables 8.3, 8.8, and 8.11 summarize the pre- and post-earthquake S O I L S T R E S S inputs, and figure 6.13 provides the material numbering scheme (for locat ion #2). 171 CHAPTER 10 Remedial Measures 10.2 Presentation and Discussion of Results Six densification schemes were analyzed. Table 10.1 summarizes the ground improvements that were simulated. Densification Case Number Description of Densification Scheme D l - densify sides o f tunnel only — not underneath D 2 - densify directly beneath tunnel D 3 - densify outside o f excavation and go to 6 meters be low excavation D 4 - same as D 3 but go to 10 meters be low excavation D 5 - densify closer to roadway and go to 6 meters be low excavation D 6 - same as D 5 but increase wid th o f densified zone; and increase depth by 4 meters Table 10.1 - Descriptions of Densification Schemes Table 10.2 summarizes the results o f the analyses for each case. A s i n previous analyses, C I is the top left corner o f the concrete section and C 3 is the bottom right corner. (Figure 6.13 shows the locations on a transverse-section drawing). The displacements at opposite corners are provided so that differential displacements can be reviewed. i.) Case mi: Figures 10.1 to 10.10 are graphic S O I L S T R E S S output depicting the magnitude 172 CHAPTER 10 Remedial Measures and pattern o f displacement for each densification scheme. In the graphics, the dashed lines represent the in i t ia l state, and the sol id lines show the post-earthquake posi t ion o f the nodes. Densif icat ion Case #D1 shows the effects o f densifying the 16-meter wide (at Remediation Case Number Corner HORIZONTAL i i iB Displacement (meters) VERTICAL Displacement (meters) D l C I 1.0 -0.27 C 3 1.0 -0.29 D 2 C I 0.31 -0.22 C 3 0.31 -0.22 D 3 C I 0.60 -0.10 C 3 0.51 -0.40 D 4 C I 0.37 -0.40 C 3 0.42 -0.26 D 5 C I 0.43 -0.36 C 3 0.43 -0.35 D 6 C I 0.18 -0.65 C 3 0.31 -0.18 Table 10.2 - Modelling of Remediation Schemes — Transverse Displacements at Location #2 base) loose sand segment adjacent to the tunnel. A s table 10.2 shows, there isn't any mit igat ion o f displacements at the concrete section. The displacement pattern is shown i n figure 10.1. The displacements are 173 CHAPTER 10 Remedial Measures magnified b y a factor o f two. Al though the stiffness o f the unit #4 loose sand w i l l degrade significantly upon liquefaction, it has little influence on the horizontal movement o f the concrete section. Figure 10.1 shows that the underlying l iquef ied materials control the movement. A s table 10.2 shows, there is no benefit i n densifying the smal l zones close to the tunnel centerline. Cases #D2 to #D6 show that the extent o f lateral movement is control led b y the under lying unit #5, #6, and #7 soils. ii.) Case#D2: The next option, Case #D2, represents the densification o f a 57-meter segment o f loose sand (units #5 and #6) directly beneath the tunnel. Figure 10.2 shows the displacement pattern. (The displacements are magnified b y a factor o f two). A s expected, this option is very effective i n decreasing horizontal movement. Ve r t i ca l displacements are similar to those i n the original (no ground improvement) analyses, but the horizontal displacements are only one-third the original predictions. It is apparent that the magnitude o f the horizontal movement is controlled b y the underlying l iquefied layers. A l though this densification scheme accomplishes its objective, accessing the zones directly beneath the tunnel may be very difficult. The remaining (more practical) schemes assess the effects o f densifying zones adjacent to the tunnel. 174 CHAPTER 10 Remedial Measures CHAPTER 10 Remedial Measures 176 CHAPTER 10 Remedial Measures iii.) Case #D3: Case #D3 represents the densification o f two 29-meter w id th zones that extend from the ground surface to the base o f the 6-meter segment o f loose sand (unit #6). The densification is initiated 35 meters from the tunnel centerline. Figures 10.3 and 10.4 show the displacement pattern and vectors, respectively. The displacements i n figures 10.3 and 10.4 are to scale (i.e., not magnified). The densified zones cause a slight c lock-wise rotation o f the concrete section. A s summarized i n table 10.2, the horizontal displacements decrease by forty to fifty percent. The decrease i n vert ical displacement at corner #1 is s imply due to the upward rotation. The displacement vectors o f figure 10.4 show the uneven effect o f the transitions from liquefied to non-l iquefied soils. The rotation o f the concrete section is most l ike ly due to the restriction caused b y the second (right side) densified zone. 177 CHAPTER 10 Remedial Measures o (ill) UOI|DA9|3 178 CHAPTER 10 Remedial Measures o tn o i r o o ( i l l ) UOI1DA8J3 179 o o CD O c o Q * Vi es U J -o •<-> u > C 4> E <u w «S a O * CHAPTER 10 Remedial Measures iv. ) Case#D4: Case #D4 is very similar to Case #D3; the only difference is that the zone o f densification extends to a depth 10 meters be low the tunnel invert. The two 29-meter w id th densified zones extend f rom the ground surface to the base o f the l iquef ied silt (unit #7). The densification is initiated 35 meters from the tunnel centerline. A s discussed earlier, the underlying layers control the lateral movement o f the concrete section. Figures 10.5 and 10.6 show the displacement pattern and vectors, respectively. (The displacements i n figures 10.5 and 10.6 are magnif ied by a factor o f two). Compar ing w i th Case #D3, the further restriction o f the movement i n the underlying layers results i n an additional decrease i n the horizontal movement o f the concrete section. A s shown i n table 10.2, this configuration reduces hor izonta l movement b y approximately sixty percent. L i k e Case #D3, the concrete rotates, except i n this configuration, it rotates counter-clockwise. The same amount o f energy has to be dissipated i n a l l the cases. The unit #5 l iquefied sand dissipates more kinetic energy i n Case #D4 (versus Case #D3) because o f the lateral restriction i n the deepest l iquef ied layer (unit #7); consequently, this results i n greater deformation i n the unit #5 sand. v. ) Case#D5: Case #D5 is similar to Case #D3, except the zones o f densification are closer to the tunnel centerline. The densification is initiated 22 meters from the tunnel centerline. 180 CHAPTER 10 Remedial Measures CHAPTER 10 Remedial Measures CHAPTER 10 Remedial Measures Case #D5 represents the densification o f two 36-meter wide segments that extend f rom the ground surface to a depth 6 meters be low the tunnel invert. The densified zones initiate 25 meters from the centerline o f the concrete section (tunnel roadway). Figures 10.7 and 10.8 show the displacement pattern and vectors, respectively. (The displacements i n figures 10.7 and 10.8 are magnified by a factor o f two). The densified zones i n configuration #D5 are effective i n decreasing hor izonta l movement. In this configuration, the horizontal displacements are decreased by approximately sixty percent (refer to table 10.2). Addi t iona l ly , the amount o f rotation is very smal l ; consequently, this densification scheme is m i l d l y preferable to Cases #D3 and #D4. Since accessing the unit #4 sands directly beneath the gravel backf i l l (unit #2) w i l l require temporary removal o f the gravel, this may not be the most economical solution. A s figure 10.7 shows, there is significant outward movement o f the elements at the left and right boundaries. This phenomenon is explained i n section 9.4.1. vi.) CaseW6: Case #D6 is similar to Case #D5. There are two differences: the zone o f densification extends to a depth 10 meters be low the tunnel invert, and an additional 8-meter wide non-l iquefied zone has been added. The two 29-meter w i d t h densified zones extend from the ground surface to the base o f the non-plastic l iquefied silt. E a c h zone is initiated 22 meters from the tunnel centerline. The additional w id th o f the 183 CHAPTER 10 Remedial Measures CHAPTER 10 Remedial Measures in m od o CD O o CO 1 1 1 r IT, a eq U • o u a. > C E U <g " a 5 • oo o o O (UU) UOI|DA8|3 o o OX) 185 CHAPTER 10 Remedial Measures densified zone represents a gradual densification scheme i n w h i c h the addit ional w id th is a part ial ly densified zone (referred to as ' P . D . ' i n figure 10.9). The partial densification is meant to simulate the gradual decrease i n density as the perimeter o f the radius o f influence (of the outer-most timber piles) is approached. This densification scheme is very effective i n l imi t ing horizontal movements, but there is significant rotation o f the concrete section. Figures 10.9 and 10.10 show the displacement pattern and vectors, respectively. (The displacements i n figures 10.9 and 10.10 are to scale). Compar ing the displacement vector graphics o f figures 10.8 and 10.10, the unit #5 sand dissipates more kinetic energy i n Case #D6 (versus Case #D5) because o f the restriction o f f low i n the deepest l iquefied layer (unit #7). The gravel backf i l l (unit #2) restricts the deformations i n the unit #4 l iquefied sand; this leads to a downward movement into the unit #5 loose sand, and the consequent counter-clockwise rotation. 186 CHAPTER 10 Remedial Measures CHAPTER 10 Remedial Measures 10.3 Preliminary Remediation Recommendation The most common method o f foundation remediation is so i l densification. The post-earthquake movements wi th in foundation soils are a function o f pos t -cycl ic strength. Since residual strength is a function o f v o i d ratio, it can be increased b y densifying the soi l . The analyses i n section 10.2 assessed the effects o f densification. S i x densification schemes were analyzed. Al though very effective i n l imi t ing displacements, densification beneath the tunnel w o u l d not be an economical solution due to lack o f accessibil i ty. Cases #D3 to #D6 assessed the effects o f densification o f zones adjacent to the tunnel. This general approach is the most efficient. Location #2 Depth (meters) [(Ni)6olReq'il (bar) 0 - 8 Sand 30 150 8 - 1 1 . 7 Sandy Sil t 20 60 Table 10.3 - Required Densification to Prevent Liquefaction at Location #2 The depth o f l iquefaction along the length o f the tunnel varies. Tables 10.3 to 10.7 summarize the remediation requirements for each location that has been analyzed i n this study. A correlation published by Robertson et al . (1983) was employed to estimate q C / N ratios to convert the [ ( N J ^ R >d estimates to cone bearing ( C P T ) equivalents. In si l ty 189 sands, q0/N = 3.5 was applied, and i n silt q0/N = 3. In sands, a q0/N value o f 5 was used. A p p e n d i x F contains details o f the calculations at each location. Locat ion #3 Depth (meters) Soil Type [(Ni)<jo]Req'd (QcXtcq'd (Blows/ft) (bar) 0 - 6 . 4 Sand 30 150 6 . 4 - 9 . 1 Sandy Sil t 20 60 Table 10.4 - Required Densification to Prevent Liquefaction at Location #3 Location #4 Depth (meters) Soil Type [(Nl)6<)]Req'd (q^Rcq'd (bar) 0 - 6 . 5 Sand 30 150 6.5 - 10.5 Sil t 23 69 Table 10.5 - Required Densification to Prevent Liquefaction at Location #4 Location #7 Depth (meters) Soil Type I(Nl)6o]Req'd (q^Rcq'd (bar) 0 - 3 . 5 Sand 20 100 3 . 5 - 7 . 2 Sand 23 115 7 . 2 - 11.0 Sand 30 150 ' 1 1 . 0 - 1 7 . 0 Sand 28 140 1 7 . 0 - 2 0 . 0 Sand 26 130 Table 10.6 - Required Densification to Prevent Liquefaction at Location #7 190 CHAPTER 10 Remedial Measures Location #8 Depth (meters) Soil Type [(N|)6fl]uei|'d (Qc)Req'd isiiiiiiiiiiiiiiiiiiiiiiiiiiii 0 - 10.0 Sand 20 100 1 0 . 0 - 12.0 Sand 19 95 1 2 . 0 - 1 5 . 8 Sand 18 90 1 5 . 8 - 2 1 . 6 Sand 17 85 2 1 . 6 - 2 3 . 0 Si l ty Sand 26 91 23.0 - 27.2 Sand 16 80 Table 10.7 - Required Densification to Prevent Liquefaction at Location #8 There is no headroom access restriction at this site, but due to the submergence o f the structure, the choice o f methods o f remediation are l imited. T imber piles w o u l d be a log ica l choice for a site such as this. Specifications o f the remediation (i.e., densification method and relevant parameters, construction specifications, etc.) w o u l d be assessed after a confirmational study has been done. 191 CHAPTER 11 CONCLUSIONS The George Massey Tunnel ' s response to the 1:475 year seismic event has been evaluated. The analyzed event is an M 7.0 causing a peak ground acceleration o f 0.24g. Current research papers i n the f ie ld o f seismic response evaluation and geotechnical analysis reports for the Fraser De l t a were reviewed. Structure and foundation geometry were assessed using as-built construction drawings. Ava i l ab le site-specific in-si tu test data was applied to assess engineering parameters and the so i l stratigraphy along the length o f the tunnel. The soils underlying the tunnel comprise loose to medium-dense sands and silts o f l o w plastici ty that could be triggered to l iquefy i n the event o f a major earthquake. Liquefac t ion resistance o f the predominant sands and non-plastic silts was based on the indirect approach using penetration test data together w i t h charts based on performance during past earthquakes. Large zones o f these underlying soils were predicted to l iquefy for the M 7.0 event. The south-most offshore location (#2), and the northern river bank (location #8) are most susceptible to triggering. Post-liquefaction stability analyses indicated that residual strengths were adequate to prevent f l ow slides at a l l locations. Because significant zones were predicted to trigger, empir ical and numerical method deformation analyses were carried out. Analyses were done i n directions transverse to and parallel to the tunnel alignment. Predictions using the most detailed empir ical method (Bart let t /Youd, 1992) corroborated w e l l w i t h 192 , , CHAPTER 11 Conclusions those o f the most rigorous numerical method ( S O I L S T R E S S — pseudo-dynamic finite element). Since the predicted displacements were large i n some cases, remedial measures were analyzed at locat ion #2. A structural analysis should be carried out to determine structural tolerances; i f foundation remediation is required, it is recommended that zones adjacent to the tunnel be densified to the predicted depth o f l iquefaction (at each analyzed location). T o affirm the results o f this study, a seismic response analysis using a method such as F . L . A . C (Fast Lagrangian Ana lys i s o f Continua) should be carried out. Furthermore, i t is recommended that, to decrease uncertainties i n parameter estimates, more in-si tu soi l test data be acquired. 193 R E F E R E N C E S Anderson, D . , and P . M . Byrne (1991). "Earthquake Des ign i n the Fraser Del ta" , Task Force Report, June. Bartlett, S.F., and Y o u d , T . L . (1992). " E m p i r i c a l Predict ion o f Lateral Spread Displacement", Proceedings, 4th US-Japan Workshop on Earthquake Resistant Des ign Li fe l ine Faci l i t ies and Countermeasures for S o i l Liquefact ion, N C E E R , H o n o l u l u , H a w a i i , M a y . B C H y d r o Pr ior i ty Transmission Tower Study (1990). " C y c l i c Test ing o f Sil ts Laboratory Test Results". Unpubl ished data. B i shop , A . W . , (1955). "The Use o f the Sl ip C i rc le i n the Stabil i ty A n a l y s i s o f Slopes". Geotechnique, V o l . 5 , N o . l . B lunden , R . H . , (1975). "Urban Geology o f R i c h m o n d Br i t i sh C o l u m b i a " , Adventures i n Ear th Science Series Number 15, B . C . Science Teachers ' Assoc ia t ion . B y r n e , P . M . , and Wal ter Janzen (1981). " S O I L S T R E S S : A Computer Program for N o n -linear Ana lys i s o f Stresses and Deformations i n S o i l " , Department o f C i v i l Engineering, Univers i ty o f Br i t i sh Co lumbia , S o i l Mechanics Series #52, December. Byrne , P . , D . Anderson, and W . Janzen (1984). "Response o f Pi les and Casings to Hor izon ta l Free F i e l d S o i l Displacements", Canadian Geotechnical Journal, V o l . 21, N o . 4. Byrne , P . M . , and U . Atukora la (1985). " A Report to the M i n i s t r y o f Transportation & Highways on Seismic Stabil i ty o f Br idge Abutments Under la in b y Organic Soi ls , Annac i s Island Project, January. Byrne , P . M . (1990). " A M o d e l for Predicting Liquefact ion Induced Displacements", Department o f C i v i l Engineering, Univers i ty o f Br i t i sh Co lumbia , S o i l Mechan ics Series #147, September. Byrne , Peter M . & Hendra Jitno (1993). "Ana lys i s o f Earthquake Induced Displacements o f the Intake Structures, John Hart D a m " , Department o f C i v i l Engineering, Univers i ty o f Br i t i sh Co lumbia , September. Byrne , P . M . (1993a). C i v i l 413 Class Notes ~ Univers i ty o f B r i t i s h Co lumbia . 194 References Byrne , P . M . (1993b). Personal Communicat ion. Byrne , P . M . , A . S . Imrie, and N . R . Morgenstern (1994). "Results and Implications o f Seismic Performance Studies for Duncan D a m " , 46th A n n u a l Canadian Geotechnical Conference Saskatoon, Saskatchewan, September 27-29, 1993, Canadian Geotechnical Journal V o l . 3 1 , pp.979-988. Byrne , P . M . (1994a). C i v i l 581 Class Notes ~ Univers i ty o f B r i t i s h C o l u m b i a . Byrne , P . M . (1994b). Personal Communicat ion. Byrne , P . M . (1994c). Memorandum to M i n i s t r y o f Transportation & Highways ~ R e v i e w o f Seismic Liquefact ion Ana lys i s o f Second Nar rows Br idge . Report Rev iewed: "Seismic Retrofit o f Second Narrows Br idge" by Golder & Associates. June. Byrne , P . M . (1995). Personal Communicat ion. Clague, John J . , and J . L . Luternauer (1991). "Postglacial Del ta ic Sediments, Southern Fraser R i v e r Del ta , B r i t i sh Co lumbia" Cra ig , R . F . (1989). " S o i l Mechanics" , fourth edition, V a n Nost rand R e i n h o l d (International) C o . L t d . . Duncan , J . M . , Peter Byrne , K a i S. W o n g , and P h i l l i p M a b r y (1980). "Strength, Stress-strain and B u l k M o d u l u s Parameters for Fini te Element Analyses o f Stresses and Movements i n S o i l Masses", Report N o . U C B / G T / 8 0 - 0 1 , Univers i ty o f Ca l i fo rn ia at Berkeley , August . Ersoy , T. (1994). Private Communicat ion. Ersoy, T. (1995). Private Communicat ion. F i n n , W . D . L . , Byrne , P . M . , and Mar t in , G . , (1976). "Seismic Response and Liquefac t ion o f Sands", Journal o f Geotechnical Engineering, V o l . 1 0 2 , N o . 8 , pp.841-856. F i n n , W . D . L . , (1988). "Dynamic Ana lys i s i n Geotechnical Engineer ing", Proceedings o f Earthquake Engineering and S o i l Dynamics II ~ Recent Advances i n G r o u n d - M o t i o n Evalua t ion Specialty Conference, A S C E , Geotechnical Special Publ ica t ion #20, pp. 523-591. 195 References F i n n , W . D . L i a m , and A n d r e w M . N icho l s (1988a). "Seismic Response o f Long-Pe r iod sites: lessons from the September 19, 1985 M e x i c a n Earthquake", Canadian Geotechnical Journal, V o l . 25. F i n n , W . D . L . , R . H . Ledbetter, R . L . F leming Jr., A . E . Templeton, T . W . Forrest and S.T. Stacy (1990). " D a m on Liquefiable Foundation: Safety Assessment and Remediat ion", 17th International Congress on Large Dams, Vienna , June. F i n n , W . D . L . (1993). "Evaluat ion o f Liquefact ion Potential", Proceedings o f the Seminar on Dynamics and Geotechnical Earthquake Engineering, A . A . Ba lkema , pp. 127-157. Department o f C i v i l Engineering -- Univers i ty o f Br i t i sh Co lumbia . ~ Gi l l e sp ie , D . G . (1993). Personal Communicat ion . H a l l , P . (1957). "Deas Island Tunnel" ; Journal o f the Structural D i v i s i o n , A S C E , S T 6 , November . Hamada, M . , S. Yasuda, R . Isoyama, and K . Emoto (1986). "Study on Liquefact ion Induced Permanent Ground Displacements", Published b y the Assoc ia t ion for the Development o f Earthquake Predict ion i n Japan, p.87. Hami l ton , Tark (1994). Personal Communica t ion Hunter, J . A . , J . L . Luternauer, K . G . Neave, S .E. Pul lan , R . L . G o o d , R . A . Burns , and M . D o u m a (1993). "Sha l low Shear W a v e Ve loc i ty -Dep th Data i n the Fraser Del ta from Surface Refract ion Measurements, 1989, 1990, 1991". Geo log ica l Survey o f Canada, Open F i l e 2504. Idriss, I . M . (1991). "Procedures for Selecting Earthquake G r o u n d M o t i o n s at R o c k Sites", Univers i ty o f Cal i forn ia at Davis , September 1991 (Revised M a r c h 1993). Jitno, Hendra, & Peter M . Byrne (1993). " A Procedure for Predict ing Seismic Deformations o f Earth Structures", Department o f C i v i l Engineering, Univers i ty o f B r i t i s h Co lumbia , September. Jitno, H . (1994). Personal Communicat ion . Jong, H . , (1988). " A Cr i t i ca l Investigation o f Post-liquefaction Strength and Steady-State F l o w Behaviour o f Saturated Soi ls" , P h . D . Thesis, Department o f C i v i l Engineering, Stanford Univers i ty , Cal i fornia . 196 References Kuese l , T . H . (1969). "Earthquake Des ign Cri ter ia for Subways". Journal o f the Structural D i v i s i o n , A S C E , ST6, June. Lambe, W . T . , R . V . W h i t m a n (1969). " S o i l Mechanics" , First Ed i t ion , Massachusettes Institute o f Technology. L o , Robert C , A l e x Sy, Pau l W . Henderson, D a v i d Y . S i u , W . D . L i a m F i n n , and Ar thur C . Heidebrecht (1991). "Seismic Site Ampl i f i ca t ion Study for Fraser Del ta , B r i t i s h Co lumbia" , 6th Canadian Conference on Earthquake Engineering Toronto, June 12-14. Lutemauer, J . L . , and W . D . L i a m F i n n (1983). "Stabil i ty o f the Fraser R i v e r De l t a Front", Canadian Geotechnical Journal, V . 2 0 , pp. 603-616. M a r c u s o n , W.F. I I I , M . E . Hynes and A . G . F rank l in (1990). "Earthquake Spectra", V o l . 6, N o . 3, August , pp. 529-572. Monaghan , Pat (1994). Personal Communicat ion . Naumosk i , N . (1985). " S Y N T H Program: Generation o f A r t i f i c i a l Acce le ra t ion T ime His to ry Compat ib le w i t h a Target Spectrum", M c M a s t e r Earthquake Engineer ing Software Libra ry , Department o f C i v i l Engineering and Engineering Mechanics , M c M a s t e r Univers i ty , Hami l ton , Canada. Newmark , N . M . (1965). "Effects o f Earthquakes on Dams & Embankments", F i f th Rankine Lecture, Geotechnique, V o l . 15, N o . 2, June, pp. 139-160. O w e n , G . N . and R . E . Scho l l (1981). "Earthquake Engineering o f Large Underground Structures". Report N o . F H W A / R D - 8 0 / 1 9 5 , Federal H i g h w a y Adminis t ra t ion , January. P i l l a i , V . S . , and R . A . Stewart (1994). "Evaluat ion o f Liquefact ion Potential o f Foundat ion Soi ls at Duncan D a m " , 46th A n n u a l Canadian Geotechnical Conference Saskatoon, Saskatchewan, September 27-29, 1993, Canadian Geotechnical Journal V o l . 3 1 , pp.951-966. P i l l a i , V . S . , and P . M . Byrne (1994). "Effect o f Overburden Stress on Liquefact ion Resistance o f Sand", Canadian Geotechnical Journal, V o l . 3 1 , pp.53-60. 197 References P i l l a i , V . S . , and F . M . Salgado (1994). "Post-Liquefact ion Stabil i ty and Deformat ion Ana lys i s o f Duncan D a m " , 46th A n n u a l Canadian Geotechnical Conference Saskatoon, Saskatchewan, September 27-29, 1993, Canadian Geotechnical Journal V o l . 3 1 , pp.967-978. Plewes, Howard , V . Sitham P i l l a i , M i c h a e l R . Morgan , and B r i a n L . K i l p a t r i c k (1994). "In situ Sampling, Densi ty Measurements, and Testing o f Foundat ion Soi ls at Duncan D a m " , 46th A n n u a l Canadian Geotechnical Conference Saskatoon, Saskatchewan, September 27-29, 1993, Canadian Geotechnical Journal V o l . 3 1 , pp.927-938. Poulos , S.J., G . Castro and J . W . France (1985). "Liquefaction Eva lua t ion Procedures", Journal o f the Geotechnical Engineering D i v i s i o n , A S C E , V o l . I l l , N o . 6, June, pp.772-792. Richter , K . J . (1991). " A Report to the M i n i s t r y o f Transportation & H i g h w a y s on the Seismic Retrofit o f Structures i n the L o w e r Ma in l and" , September. R i p l e y & Associates L t d . (1956). "Reports on Testing Program & Laboratory Test Data for Samples from holes 1 to 10 - Deas Island Tunnel" . Robertson, P . K . , R . G . Campanella, D . Gi l lespie , and A . R i c e (1986). "Se ismic C P T to Measure In Si tu Shear W a v e Ve loc i t y " , Journal o f Geotechnical Engineering, V o l . 112, N o . 8, August . Robertson, P. , & R. Campanel la (1989). "Guidelines for Geotechnical Des ign U s i n g C P T and C P T U " , Department o f C i v i l Engineering, Univers i ty o f B r i t i s h Co lumbia , S o i l Mechan ics Series N o . 120. Robertson, P . K . (1990). "Seismic Cone Penetration Testing for Evaluat ing Liquefact ion Potential", Proceedings, Sympos ium on Recent Advances i n Earthquake Des ign U s i n g Laboratory and In Si tu Tests, ConeTec Investigations L t . , Burnaby, B C , February 5. Schnabel, Per B . , John Lysmer , and H . B o l t o n Seed, (1972). " S H A K E : A Computer Program for Earthquake Response Ana lys i s o f Hor izonta l ly Layered Sites", Report N o . E E R C 72-12, Col lege o f Engineering ~ Univers i ty o f Cal i forn ia at Berkeley , December. Seed, H . , and Idriss, I. (1970). " S o i l M o d u l i and Damping Factors for D y n a m i c Response Analyses" , Earthquake Engineering Research Center Report, N o . E E R C 70-10, December, Col lege o f Engineering, Univers i ty o f Cal i fornia . 198 References Seed, H . B . (1979). "Considerations i n the Earthquake-Resistant Des ign o f Ear th and R o c k f i l l Dams" , 19th Rankine Lecture, Geotechnique, V o l . 29, N o . 3, pp.215-263. Seed, H . B . , I . M . Idriss, & I. Arango (1983a). "Evaluation o f Liquefact ion Potential U s i n g F i e l d Performance Data", Journal o f Geotechnical Engineer ing D i v i s i o n , A S C E , V o l . 109, G T 3 , pp. 458-482. Seed, H . B . (1983b). "Earthquake-Resistant Des ign o f Ear th Dams" , i n Seismic D e s i g n o f Embankments and Caverns, Terry R . Howard , Editor , A S C E , pp. 41-64. Seed, H . B . , Tokimatsu, K . , Harder, L . , and Chung, R . (1984). "The Influence o f S P T Procedures i n S o i l Liquefact ion Resistance Evaluations", Report N o . U C B / E E R C - 8 4 / 1 5 , Col lege o f Engineering, Univers i ty o f Cal i fornia , Berkeley, Cal i forn ia . Seed, H . B . (1987). "Design Problems i n S o i l Liquefact ion", Journal o f Geotechnical Engineering, A S C E , V o l . 113, N o . 7, August, pp. 827-845. Seed, H . B . , R . B . Seed, L . F . Harder and H . L . Jong (1988). "Re-evaluation o f the Slide i n the L o w e r San Fernando D a m i n the Earthquake o f February 9, 1971", Report N o . U C B / E E R C - 8 8 / 0 4 , Univers i ty o f Cal i fornia , Berkeley, A p r i l . Seed, H . B . , & L . F . Harder Jr. (1990). "SPT-Based Ana lys i s o f C y c l i c Pore Pressure Generat ion and Undra ined Residual Strength", Proceedings, H . B o l t o n Seed M e m o r i a l Symposium, J. M i c h a e l Duncan (ed.), V o l . 2, M a y , pp.351-376. Sego, D . C . , P . K . Robertson, S. Sasitharan, B . L . Ki lpa t r ick , and V . S . P i l l a i (1994). " G r o u n d Freezing and Sampling o f Foundation Soils at Duncan D a m " , 46th A n n u a l Canadian Geotechnical Conference Saskatoon, Saskatchewan, September 27-29, 1993, Canadian Geotechnical Journal V o l . 3 1 , pp.939-950. Senneset, K . , and N.Janbu (1981). "Shear Strength Parameters Obtained from Static Cone Penetration Tests", A S T M S T P 883, Symposium, San Diego . Stark, T . D . , and G . M e s r i (1992). "Undrained Shear Strength o f L iquef ied Sands for Stabil i ty Ana ly s i s " , Journal o f Geotechnical Engineering, A S C E , V o l . 118, N o . 11, pp. 1727-1747. 199 References Supplement to the Nat iona l B u i l d i n g Code o f Canada (1990). Na t iona l Research C o u n c i l o f Canada. Sy, A l e x , Pau l W . Henderson, Robert C . L o , D a v i d Y . S iu , W . D . L i a m F i n n , and Ar thur C . Heidebrecht (1991). " G r o u n d M o t i o n Response for Fraser Del ta , B r i t i s h C o l u m b i a " , Fourth International Conference on Seismic Zonat ion, Stanford Cal i forn ia , Augus t 26-29. Tokimatsu, K . , H . B . Seed (1987). "Evaluat ion o f Settlements i n Sands D u e to Earthquake Shaking", Journal o f Geotechnical Engineering, A S C E , V o l . 113, N o . 8 , pp. 861-878. V a i d , Y . P . , and J . C . Chern (1985). " C y c l i c and Monoton ic Undra ined Response o f Saturated Sands. A S C E Na t iona l Convent ion, Session ~ Advances i n the A r t o f Test ing Soi ls Unde r C y c l i c Loading" , Detroit, October 21-25, pp. 120-147. V a i d , Y . P . , E . K . F . C h u n g and R . H . Kuerb i s (1989). "Stress Path and Steady State", S o i l Mechanics Series N o . 128, Dept. o f C i v i l Engineering, Univers i ty o f B r i t i s h Co lumbia , Vancouver , B C , M a r c h . W a n g , W . (1979). "Some Findings i n S o i l Liquefact ion", Water Conservancy and Hydroelectr ic Power Scientific Research Institute, Bei j ing , Ch ina , August . Y o u d , T . L . , and S.F. Bartlett (1988). " U . S . Case Histories o f Liquefac t ion Induced G r o u n d Displacement", Proceedings, First Japan-U.S. Workshop on Liquefact ion, Large G r o u n d Deformation and their Effects on Li fe l ine Faci l i t ies , pp. 22-31. Y o u d , T . L . , and D . M . Perkins (1987). " M a p p i n g o f Liquefact ion Severity Index ~ L S I " , Journal o f Geotechnical Engineering, V o l . 113, N o . 11, pp. 1374-1392. 200 APPENDIX A NUMERICAL METHOD DISPLACEMENTS APPENDIX A.1 - Additional SOILSTRESS Analyses A. 1.1 Displacements at Offshore Locations « Case #2: No Sediment Loading The Case #2 analyses are similar to the Case #1 analyses (refer to section 8.3.2.1.2); the only difference is that there is no sediment loading on top o f the tunnel. The three load cases are compared i n section 9.4.1.1. This case was analyzed to determine what effect the removal o f sediment w o u l d have on the displacements. Location Corner HORIZONTAL Displacement (meters) 2-Dimension VERTICAL Displacement (centimeters) TOTAL (3-Dimension) VERTICAL Displacement (centimeters) #2 C I 0.97 -22.1 -18.2 C 3 0.96 -23.9 -20.0 #3 C I 0.19 -13.4 -14.6 C 3 0.19 -14.3 -15.5 #4 C I 0.17 -11.9 -17.9 C 3 0.16 -13.0 -19.0 Table A. 1.1 - Displacements at Offshore Locations - No Sediment Loading Figures A . 1.1 to A . 1.3 show the displacement patterns for the Case #2 analyses at locations #2, #3, and #4. The dashed lines represent the in i t ia l state, and the so l id lines show the post-earthquake posi t ion o f the nodes. The displacements are magnif ied b y a factor o f two. The displacement estimates for the Case #2 analyses are summarized i n table A . 1.1. A s described earlier, the 'total vertical displacement' ( in the f inal co lumn o f the table) represents the fu l l 3-dimensional effect b y incorporating the undrained vert ical distortion estimates from the longitudinal analyses. The displacement patterns i n the Case #2 analyses are very similar to the corresponding ones i n the Case #1 analyses. 201 202 . , A P P E N D I X A SOILSTRESS A. 1.2 Case #3: Displacements with Increased Sediment Loads These analyses were done only at location #2. The Case #3 analyses were done to determine the effects o f various sediment loads. A s described earlier, Case #1 appl ied 2 meters o f sediment loading on top o f the tunnel, and Case #2 analyses were done using no sediment. Thickness of Sediment Corner H O R I Z O N T A L Displacement (meters) 2-Dimension 1111111111111 Displacement (centimeters) TOTAL (3-Dimension) V E R T I C A L Displacement (centimeters) 4 meters C I 1.12 -30.7 -26.8 C 3 1.13 -27.8 -23.9 6 meters C I 1.11 -32.0 -28.1 C 3 1.12 -28.0 -24.1 Table A . 1.2 - Displacements at Location #2 — Increased Sediment Loading In this analysis, the sediment loading has been increased. A mass equivalent to 4 meters (Case #3 a) o f sediment was first incorporated i n the finite element mesh, and then a sediment load case o f 6 meters (Case #3b) was analyzed. Refer to section 9.4.1.1 for a discussion o f the results using the various sediment loads. Figures A . 1.4 and A . 1.5 show the displacement patterns for the Case #3 analyses at locations #2. The dashed lines represent the ini t ia l state, and the so l id lines show the post-earthquake posi t ion o f the nodes. The displacements are magnif ied b y a factor o f two. Table A . 1.2 summarizes the results o f the analyses. Compar ing the results i n table 8.12 (i.e., Case #1) and table A . 1.2, the horizontal displacements increase b y approximately 2 0 % (from 0.96 m to 1.12 m , at corner #1); the vert ical displacements increase b y 16% at corner #1, and at corner #3, they increase b y 6%. 2 0 5 APPENDIX A SOILSTRESS o o m to CD O o w Q ro •3 en e s s •3 «*-0 BO s~ V *j s es % u 01 PS U .o « s C8 o o i i r i 1 r o O ro O CN O o 111) U0I|DA9|3 o o E. 3 OJD 206 _ _ , APPENDIX A SOILSTRESS A.1.3 Case #4: Displacements with Increased Ground Velocity In this comparison analysis, the max imum veloci ty to compute kinetic inertia has been increased by 2 7 % (from 0.30 m/s to 0.38 m/s) at location #2. S o i l input parameters are summarized i n tables 8.8 and 8.11. (The analysis is discussed i n section 9.4). Location Corner HORIZONTAL Displacement i l l l l l l l i i l l l ^ 2-Dimcnsion VERTICAL Displacement (centimeters) T O T A L (3-Dimension) VERTICAL Displacement (centimeters) #2 C I 1.15 -26.4 -22.5 C 3 1.15 -26.7 -22.8 Table A.1.3 - Displacements at Location #2 — Increased Ground Velocity Table A . 1 . 3 summarizes the case #4 results (of the increased ve loc i ty analyses at locat ion #2). Compar ing the results i n table 8.12 (at location #2) and table A . 1.3, the horizontal displacements increase by approximately 20%, and the vert ical displacement estimates are very s imilar i n both cases. Figure A . 1.6 shows the displacement pattern for the Case #4 analysis at locations #2. The dashed lines represent the in i t i a l state, and the sol id lines show the post-earthquake posi t ion o f the nodes. The displacements are magnif ied by a factor o f two. 208 APPENDIX A SOILSTRESS o o in CO CD O co Q "i r o o I o o o o o (ill) UOI}DA9|3 _ o "cD > a s o J -0 •a « % U in es u e J -u es OH •<-> C s u _es O . n • SO E 3 DO 209 Appendix A.1.4 ~ Full-Section Longitudinal Displacements Refer to section 4.4.3.2 for a description o f the procedures used to do the S O I L S T R E S S analyses. Displacements i n the plane parallel to the axis o f the Longitudinal Section Analysis Material Number Soil Type n Kb m Rr fd02> Ao> (deg) c 1 Concrete 50000 (50000) 0.5 200000 0.25 0.5 50 0 1000 2 Liquefied Silt 444 (1.1) 0.5 (0) 2000 0.25 0.7 (0) 34 (° ) 0 30 3 Liquefied Sand Fill 181 (0.1) 0.5 (0) 2000 0.25 0.9 (° ) 32 ( ° ) 0 0 4 Liquefied Sand 505 (1.8) 0.5 (0) 2000 0.25 0.8 (0) 35 (° ) 0 0 5 Silt 444 (222) 0.5 2000 0.25 0.6 34 0 40 6 Liquefied Sand 505 (0.4) 0.5 (° ) 2000 0.25 0.8 (0) 33 (0) 0 0 7 Partially Liquefied Sand 955 (100) 0.5 2000 0.25 0.7 37 2 0 8 Partially Liquefied Sand 570 (100) 0.5 2000 0.25 0.7 31 0 0 9 Silt 747 (374) 0.5 2000 0.25 0.6 34 0 40 10 Liquefied Sand 200 (0.5) 0.5 (0) 2000 0.25 0.8 (0) 33 (0) 0 0 11 Sandy Sediment 200 0.5 2000 0.25 0.6 33 0 0 12 Dense Sand Lens 1137 (568) 0.5 2000 0.25 0.6 37 7 0 13 Liquefied Sand 570 (2.2) 0.5 2000 0.25 0.8 (0) 35 (0) 0 0 Table A.1.4 - SOILSTRESS Pre- and Post-liquefaction Inputs For Longitudinal Section Analyses at All Locations Notes: - values in parentheses indicate post-liquefaction estimates 210 APPENDIX A SOILSTRESS tunnel ( 'longitudinal-direction') are summarized i n this section. Res idua l strength and l imi t strain estimates are presented i n section 6.3. Figure 6.1 shows the estimated so i l profile along the length o f the tunnel. The material numbering scheme for the longitudinal section is summarized i n figure 6.14. Figures A . 1.7 and A . 1.8 show the displacement pattern and displacement vector graphics for the longitudinal section S O I L S T R E S S analyses. The displacement vector images has been provided to aid i n interpreting the displacement patterns along the fu l l length o f the tunnel. Pre- and post-earthquake inputs for the S O I L S T R E S S analyses are summarized i n table A . 1.4. Fo r post-earthquake conditions, a 5 0 % reduction i n stiffness was applied to a l l non-l iquefied materials due to the severity o f shaking (i.e., v ^ = 0.3 m/s). Res idua l strength and l imi t strain estimates that were applied i n the transverse-direction analyses were also applied i n the longitudinal analyses. (Section 6.3 summarizes the post-l iquefaction parameter magnitudes). Table A . 1.5 summarizes the displacements at the base o f the concrete section i n the v ic in i ty o f each o f the five locations at w h i c h the transverse-section analyses were done. The 'pre-consolidation vertical displacement' is the movement due to undrained distortion, and the '2-Dimension vertical displacement' is the total predicted vert ical movement i n w h i c h settlement due to dissipation o f excess pore-pressures is also included. It should be noted that the Tokimatsu/Seed method (1987) was applied (refer to section 4.4.2.4) to determine the volumetric strain inputs for the settlement calculations i n Location HORIZONTAL Displacement (meters) Pre-Consolidation Vertical Displacement (centimeters) 2-Dimension VERTICAL Displacement (centimeters) T O T A L (3-Dimension) VERTICAL Displacement (centimeters) #8 1.20 2.2 -30.4 -22.5 #4 1.20 -6.0 -19.7 -22.0 #3 1.20 -1.2 -12.5 -14.2 #2 1.19 3.9 -31.1 -22.8 #7 1.21 -0.6 -23.2 -24.0 Table A.1.5 - Longitudinal-direction Analysis Displacement Predictions 2 1 1 A P P E N D I X A SOTLSTRFSS a l l o f the S O I L S T R E S S analyses. The 'total vertical displacement' ( in the final c o l u m n o f each displacement summary table) represents the fu l l 3-dimensional effect b y incorporating the undrained vert ical distortion estimates from the onshore and offshore (i.e., Case #1) transverse-section analyses. (Refer to sections 8.3.2.1.1 and 8.3.2.1.2 for transverse-section pre-consolidation estimates). Figures A . 1.7 and A . 1.8 show the displacement pattern and displacement vector graphics for the longitudinal-direction analyses. The dashed lines represent the in i t i a l state, and the so l id lines show the post-earthquake posi t ion o f the nodes. The displacements are magnif ied b y a factor o f three. Figure 6.1 ( in section 6.2.) shows the zones o f l iquefaction a long the length o f the tunnel, and the distances between the analyzed locations. A s table A . 1.5 shows, the hor izontal movements are predicted to be uniform over the length o f the tunnel. The phenomenon is discussed i n section 9.4. The 3-dimension vertical displacement predictions are very s imi lar along the length o f the tunnel. A s anticipated, location #3 shows the lowest vert ical displacement, since the depth to l iquefaction is least i n the central segment o f the tunnel. The displacement vectors (figure A . 1.8) shows the uneven effect o f the transitions from l iquefied to non-l iquefied soils. Fo r instance, the unit #7 sands were not predicted to l iquefy; consequently, the l iquefied unit #6 and #2 soils experience greater distortion as the horizontal veloci ty impulse is applied. A s figure A . 1.7 shows, there is a noticeable outward movement o f the elements at the left and right boundaries. This phenomenon is discussed i n section 9.4. The vert ical displacement estimates at the five locations show h o w m u c h differential movement occurs along the length o f the tunnel. Different ial movements are discussed i n section 9.4.1.2. 212 213 APPENDTX A SOILSTRFSS 214 APPENDIX A SOILSTRESS APPENDIX A.2 -- Extended Newmark Model Newmark ' s s impl i f ied (i.e., r igid-plastic single-degree-of-freedom) mode l is that o f a b lock o f mass M resting on an inc l ined plane o f slope a. The mass is subjected to a veloci ty pulse ( V ) relative to the base (refer to figure A.2.1). The mass (block) w i l l commence s l id ing along the plane when the base acceleration induces a dr iv ing force w h i c h together w i th the static dr iving force is equal to the s l id ing resistance o f the block. The base acceleration at w h i c h movement is initiated is cal led the ' y i e l d acceleration' or 'resistance coefficient ' (N) . The resistance coefficient (N) can be determined through l im i t equi l ibr ium analysis. The product o f the coefficient and the weight o f the s l id ing mass is used to estimate the resistance to movement. A n y prescribed time-history o f acceleration can be applied at the base and the resulting Figure A.2.1 - Block on an Inclined Plane Subjected to a Velocity Pulse ~ Newmark Model (After Newmark, 1965) displacements can be computed by numerical integration. Alternat ively, N e w m a r k found that the m a x i m u m displacement (d) at the end o f the shaking per iod could be estimated f rom simple formulae by considering the earthquake mot ion to be estimated by a chosen V V = Velocity M = Mass of Block D = Seismic Displacement 215 APPENDIX A SOILSTRESS number o f pulses. The resulting displacement is g iven by: d = 6V^_ (equation A . 2 . 1 ) 2gN where: d - maximum displacement V ~ velocity pulse (taken to be equal to the peak ground velocity) N ~ yield acceleration (as a fraction of 'g') required to initiate yield and sliding g — acceleration of gravity N e w m a r k found agreement w i th the integrated records when s ix pulses o f ve loc i ty ( V ) were considered and the ratio N / A < 0.13 (where 'A ' ' is the peak ground acceleration). (The coefficient '6 ' i n equation [A.2.1] reflects the six pulses). Th i s ratio o f N / A reflects the conditions i n most practical cases. The number o f pulses ( in equation [A.2.1]) depends on the N / A ratio. Unde r ly ing the N e w m a r k procedure is the assumption that the stress-strain behaviour o f so i l is rigid-plastic. Th i s implies that strains (deformations) w i l l not occur unt i l inertia forces exceed the failure loads. Consequently, displacements attributed to strain-softening are not accounted for. The Extended N e w m a r k method (Byrne, 1990), on the other hand, extends the Newmark work-energy model to a general formulation and accounts for N e w m a r k ' s discrepancy. The essential difference between the Newmark (1965) and Byrne (1990) methods is i n the w a y the post-liquefaction undrained shear stress-strain response o f the so i l is model led. The characteristic post-liquefaction stress-strain curve is shown i n figure A . 2 . 2 . The figure illustrates that i f a shear stress is applied to a l iquef ied so i l , it w i l l undergo large deformations unt i l the di lat ion o f the soi l skeleton results i n a gradual gain i n stiffness and strength. The key parameters that control the post l iquefact ion stress-strain response are the residual shear strength (S r) and the l imi t ing shear strain (v l i m ) . (Refer to section 4.3 for a discussion o f these parameters). It has been shown that the shear strains required to trigger l iquefaction are generally smal l (approximately 0 . 1 % to 1%) when compared w i th the ensuing strains caused by the earthquake and gravity loads acting on the l iquefied so i l ; consequently Byrne (1990) proposed that the triggering strains be neglected. Post-liquefaction strains depend on the post-cycl ic stress-strain characteristic behaviour o f the so i l and on the geometry o f the structure (i.e., static bias). Consequently, l iquefaction induced displacements can be adequately predicted by acknowledging both the gravity loads and earthquake-induced inertia forces, i n conjunction w i t h the appropriate post-liquefaction stress-strain relations. The Extended N e w m a r k model fol lows the work-energy theorem that states that the work done by the internal forces ( W ^ ) minus the work done by the external forces ( W E X T ) must be equal to the change i n kinetic energy o f the system: 216 APPENDIX A SOfLSTRFXS W -W yy m T yy, 2 (equation A . 2 . 2 ) where: 1/2MV2 ~ change in kinetic energy of the system V — specified initial velocity of the system Vf -- final resting velocity (equal to zero) M — mass of the system Displacements caused by the gravity forces acting on the softened so i l are depicted between points B and D i n figure A . 2 . 2 . U p o n liquefaction, the pore-pressure rise causes the stress i n the so i l to drop from its static value at point A to point B . A s the so i l dilates Stress Pre-Liquefaction Resistance 1 Driving Stress (rst) Post-Liquefaction Soil Resistance Strain Figure A.2.2 - Work-energy principle — Extended Newmark (Byrne, 1990) 217 APPENDIX A SOILSTRESS and strains increase, it begins to regain stiffness to balance out the dr iv ing force (gravity loading). A s depicted i n figure A . 2 . 1 , the dr iving force from the static stresses remains constant. Because the soil's resistance had decreased upon the triggering o f l iquefact ion (i.e., point A to B ) , the constant static dr iving force causes the system to accelerate as it deforms. W h e n the strain reaches point C , the material has hardened sufficiently to counter the original dr iving force; however, the veloci ty developed b y the system between points B and C causes the stress to increase unt i l the resistance reaches the residual value (point D ) . (It should be noted, that i f the dr iving stress exceeds the residual strength (S r) o f the soi l , a f l ow slide w i l l occur; refer to section 4.3 for a discussion o f post-liquefaction stability). A t point D , the net work, ( W I N T - W E X T ) is equal to zero as depicted b y the equal area portions A B C and C D F . The displacements between points A and D represent the strains that occur due to the static (gravity and boundary) loads acting on the softened soi l . The earthquake-induced in i t i a l ve loc i ty ( V 0 ) (developed at the time liquefaction is triggered) also contributes to the movement; it is represented b y the movement from point D to E . A s mentioned earlier, the N e w m a r k (1965) method is over ly s impl i f ied. N e w m a r k does not account for the displacements from point A to D (figure A . 2 . 2 ) . In cases where l iquefaction occurs, this discrepancy results i n a significant underestimation o f displacements. Strains o f approximately 20-50% are usually required to mobi l i ze the residual strength (S r ); consequently, i f Newmark ' s method is used, these strains aren't accounted for i n the overal l displacements. Fo r verif icat ion purposes, Newmark ' s (1965) method is n o w derived i n terms o f the work-energy principles. A s illustrated i n figure A . 2 . 3 , the external force that does work is the gravity (static) dr iving force (= Mg-sinoc), and is taken to be constant w i t h displacement. Therefore, the work done is represented b y the area beneath the dr iving force l ine. The work done by the internal forces, based on Newmark ' s assumption o f r igid-plastic stress-strain behaviour o f soi l , is shown by the total area beneath the so i l resistance l ine. A s the figure shows, this internal force (resistance) is assumed to be constant w i t h displacement. The net work done must equal the loss i n kinetic energy ( 1 / 2 M V 2 ) , and is illustrated as the difference between the two areas (i.e., the shaded area i n figure A . 2 . 3 ) . The dr iving force is made up o f both the static (gravity load) and earthquake pulse dr iving components. The static dr iving force is the self-weight vector component (= Mg-sinoc) dr iving the b lock down the inc l ined surface. External work ( W E X T ) done b y the gravity force, therefore, is approximated as: WEXT = iMS since)-*/ (equation A . 2 . 3 ) Oppos ing the static dr iving force is the residual strength (S r) o f the l iquefied 218 . ; APPENDIX A SOILSTRESS layer that provides the sliding (internal) resistance; therefore, the work done by the residual shear resistance of the soil is approximated as: WINT = {Sr-L-b)-d (equation A.2.4) where S r is the residual shear strength of the soil, and ' L ' and 'b' are the length and width of the slide block, respectively. Consequently, equation [A.2.4] reduces to: d-(S-L-b - Mg sina) = - MV1 (equation A.2.5) Force A 1 Soil Resistance T Driving Force (Mgsincx) Displacement, D Figure A.2.3 - Work-energy (Rigid-plastic Stress-strain Behaviour) ~ Newmark Method 219 APPENDIX A SOfLSTRFSS Solv ing for the m a x i m u m displacement (d): (equation A . 2 . 6 ) where the y i e l d acceleration (N) is given by: N= Sr-L-b-Mgsma Mg (equation A . 2 . 7 ) Equat ion [A.2.6] represents a single veloci ty pulse; when s ix pulses are considered, a solution identical to Newmark 's (equation [A.2.1]) is obtained. Therefore, i f r ig id -plastic stress-strain behaviour is assumed, Newmark 's model is obtained i n terms o f the work-energy theorem. It should be noted that, though Newmark simulated the effect o f an earthquake by applying as many as s ix veloci ty pulses (depending upon the N / A ratio), Byrne (1990) considered only one pulse when dealing w i th l iquef ied soi l . The displacements due to pulses prior to l iquefaction w i l l be, i n general, smal l (i.e.,strains o f 0.1 to 1%) when compared to those w h i c h occur upon l iquefact ion; once l iquefaction is triggered, it is assumed that no further major pulses w o u l d occur (Byrne, 1992). 220 APPENDIX A SOILSTRESS APPENDIX A.3 - Extended Newmark Model - Integration with SOILSTRESS A.3.1 Extended Newmark: Multi-degree-of-freedom Finite Element Analyses: SOILSTRESS F o r the multi-degree-of-freedom system, the S O I L S T R E S S finite element approach computes displacements from the solution of: where: [K] ~ global stiffness matrix of the system {6} — vector of nodal displacements {F} ~ static load (i.e., gravity and boundary load) vector acting on the system {AF} ~ additional load applied to satisfy the energy balance I f { A F } equals zero, then for a single-degree-of-freedom analysis, a displacement corresponding to point C i n figure A.2.2 w o u l d be predicted; therefore, an addit ional force is required to balance the energy and predict a displacement corresponding to point D or E . The additional force is applied using a seismic coefficient (k): where: {AF} ~ additional force vector (to satisfy the energy balance) k ~ pseudo seismic coefficient w - weight of soil element It should be noted that 'k' is not related to the peak ground acceleration. The seismic coefficient is applied iteratively unti l the system energy is balanced according to equation[A.2.2]. For the multi-degree-of-freedom system, represents the work done by the element stresses, and W E X T represents the work done by the static load vector: [K]{S} = {F+AF} (equation A.3.1) {AT7} = {hw} (equation A . 3.2) (equation A . 3.3) 221 APPENDTX A SOfLSTRF.SS The seismic coefficient acts to adjust { A F } unt i l displacements {5} are obtained that balance the system energy. It should be noted that this additional force is not inc luded as part o f the external work ( W E X T ) done by the static forces. In developing the multi-degree-of-freedom system, the components o f the M o d i f i e d Newmark method analysis are applied to the elements. The internal work ( W I N X ) is represented b y the work done by a l l the element stresses, the external work ( W E X T ) is the work done by the static load vector ( { F } { A } T ) , and the change i n kinet ic energy o f the system is summed element by element: |^max = J £ iK'Kj e = l,ne (equation A . 3 . 4 ) 'e' represents a soi l element and is defined by the l imits e=l ,n , where h ' is the total number o f elements used to define the mesh. Hyperbo l ic stress-strain models are used to represent the shear modulus ( G s ) and bulk modulus ( B s ) i n the S O I L S T R E S S program code. The so i l is treated as equivalent isotropic elastic, and secant estimations are used to approximate the variations o f the shear and bulk modu l i w i t h stress level as fol lows: o j n T-Rf Gs = kgPai-j-Ti1--^-) (equation A . 3 . 5 ) " f a 1 Bs = kbpai-f-r (equation A . 3 . 6 ) a Figures A . 2 . 4 and A . 2 . 5 show the shear and bulk modu l i as secant estimations. Figure A . 2 . 5 shows a graphical description o f the bulk modulus as the ratio o f the mean normal stress ( o ' m ) to the volumetric strain (e v). Figure A.2 .4(c) depicts the shear modulus and other related parameters as they are estimated and applied iteratively i n the S O I L S T R E S S code, ' x ' is the mobi l i zed shear stress, x f is the failure strength, and Rf is the ratio o f the strength at failure to the ultimate strength from the best-fit hyperbola. In figure A . 2 . 4 , ' G m ' represents the in i t ia l max imum shear modulus estimate, and P a is atmospheric pressure (to define the unit system being used). ' k g ' and ' k b ' are shear and bu lk modulus numbers, and V a n d ' m ' are their respective modulus exponents. The determination o f the hyperbolic parameters is described i n detail b y Byrne & Janzen (1989). 222 APPENDIX A SOILSTRESS t Simple Shear (a) Shear Strain, 7 Typical Shear Stress — Strain Curve (b) Shear Strain, f ( C ) Figure A.2.4 - SOILSTRESS - Idealized Nonlinear Shear Modulus Estimation (After Byrne & Janzen, 1989) 223 APPENDIX A SOfLSTRFSS APPENDIX B SHAKE ANALYSES B.1 SHAKE Analysis Method T o solve the nonlinear problem, S H A K E uses an iterative viscoelastic method o f analysis. S o i l , upon deforming, fol lows a hysteretic stress-strain path and the shape o f the stress-strain path is dependent upon the stress-strain amplitude; therefore, to approximate the nonlinear soi l behaviour, the program uses secant shear modu l i and Stress, A L Damping - 4 t t A t / / / • / ' / — h ~ / f l ^ - y/ 1 ) / \ N y 1 f\- Area of Triangle \ = Area of Loop j / / // / \ °2 Strain (l) 1 / G = Shear Modulus ^ = Shear Strain Figure B. 1.1 - Estimation of Shear Modulus and Damping --. SHAKE damping ratios. Figure B . 1.1 shows the hysteretic stress-strain path. A straight l ine through the ends o f the stress loop defines the shear modulus; the damping ratio is the ratio o f the area o f the hysteresis loop to the area o f the triangle defined b y the shear modulus and the endpoint o f the loop. The iterative procedure is based on the assessment o f this equivalent linear modulus and a viscous damping ratio to detennine the strain amplitude, w h i c h w i l l then lead to the designation o f a new modulus and damping ratio. The iteration continues unt i l stabilization o f the solution is achieved. 2 2 5 APPENDIX B Liquefaction Analysis B.2 Comparison of CSR's Using Various Bedrock Velocities Figure B.2.1 shows the variation in the computed cyclic stress ratios if the firm ground velocity input is adjusted. Throughout this study, a firm ground velocity of 3500 m/s was applied. As expected, the CSR's increase as the firm ground input increases. Cyclic Stress Ratio (CSR) i 1 i 1 r 1 — i — 1 — i — 1 — i 0.00 0.10 0.20 0.30 0.40 0.50 0.00 20.00 40.00 60.00 F 80.00 <u & 100.00 -•—« QL a 120.00 140.00 160.00 — 180.00 — 200.00 Figure B.2.1 - Comparison of C.S.R.'s Using Varying Bedrock Velocities at Location #2 226 APPENDIX B Liquefaction Analysis B.3 Surface Spectral Response — SHAKE The computer program S H A K E was used to compute the response of the soil deposits to the selected input motions. The analysis procedure is discussed in section 4.2.3.2. The surface spectral response for the 475-year return period earthquake are graphically illustrated in figures B.3.1 to B.3.5. The results summaries (in this section) should be reviewed in conjunction with the discussion in section 7.2.2. The surficial soils at the north and south river banks consist mostly of loose sands, whereas the offshore locations show more silt content. The peak surface response spectral values at the onshore and offshore locations are very similar. As figures B.3.1 to B.3.5 show, the surface response spectra for all locations show considerable amplfications at periods close to about 0.5 seconds. : Surface Spectral Response - Location #2 0.00 1.00 2.00 3.00 4.00 5.00 6.00 Period (sec) Figure B.3.1 - Ground Surface Spectral Response ~ Location #2 227 APPENDIX B Liquefaction Analysis 3 e 0 2 & 0.70 1 o TJ <D (A Q_ 0.10 —| o.oo -Surface Spectral Response - Location #3 (5% Damping) Caltech K Hughes Lake Griffith Park 0.00 1.00 2.00 3.00 4.00 Period (sec) 5.00 6.00 Figure B.3.2 - Ground Surface Spectral Response — Location #3 228 APPENDIX B Liquefaction Analysis 1.20 1.10 1.00 Surface Spectral Response - Location #4 (5% Damping) 0.00 1.00 2.00 3.00 4.00 Period (sec) 5.00 6.00 Figure B.3.3 - Ground Surface Spectral Response ~ Location #4 229 APPENDIX B Liquefaction Analysis Surface Spectral Response - Location #7 (5% Damping) Caltech V Hughes Lake Griffith Park 0.00 1.00 2.00 3.00 4.00 Period (sec) 5.00 6.00 Figure B.3.4 - Ground Surface Spectral Response ~ Location #7 230 APPENDIX B Liquefaction Analysis 1.00 0.00 Surface Spectral Response - Location #8 (5% Damping) Caltech v y ^ *% Hughes Lake Griffith Park 0.00 1.00 2.00 3.00 4.00 Period (sec) 5.00 6.00 Figure B.3.5 - Ground Surface Spectral Response ~ Location #8 231 APPENDIX B Liquefaction Analysis B.4 Ground Motion Amplification Summary G r o u n d mot ion amplifications o f peak horizontal f i rm ground accelerations (a ,^) were computed at each location. Ampl i f ica t ions varied between 4 0 % (i.e., a s u r f a c e /a f i n n = 1.4) and 60%. A f i rm ground design acceleration o f 0.24g was applied i n this study (as described i n chapter 3). Peak surface accelerations varied between 0.33 g and 0.38 g. Location # Amplification Factor (^ •surface/ ^ firm) (%) Peak Surface Acceleration (g) 2 40 0.33 3 45 0.35 4 45 0.35 7 60 0.38 8 50 0.36 Table B.4.1 - Ground Motion Amplification at Each Location 232 APPENDIX B Liquefaction Analysis Appendix B.5 — Liquefaction Analysis Summaries Appendix B.5.1 — Location #2 L I Q U E F A C T I O N A N A L Y S I S for CPT 91-2 Profile 1 1 Depth Cyclic Stress Ratio (CSR) Layer # (ft) Caltech Griffith Hughes Average Standard M ean + S ( C R R ) l Park Lake CSR D eviation + Sigma 1 12.35 0.458 0.46 0.42 0.446 0.022539 0.468539 0.24 2 17.25 0.453 0.455 0.414 0.440667 0.0231 16 0.463782 0.24 3 22.05 0.444 0.447 0.405 0.432 0.023431 0.455431 0.148 4 26.75 0.433 0.433 0.392 0.419333 0.023671 0.443005 0.148 5 29.75 0.424 0.42 0.383 0.409 0.022605 0.43 1605 0.148 6 31.05 0.418 0.414 0.377 0.403 0.022605 0.425605 0.2 7 35.15 0.401 0.391 0.36 0.384 0.021378 0.405378 0.21 8 41.75 0.373 0.357 0.335 0.355 0.019079 0.374079 0.21 9 48.15 0.344 0.324 0.314 0.327333 0.01 5275 0.342609 Not App. 10 54.35 0.316 0.294 0.296 0.302 0.012166 0.314166 Not App. 1 1 59.75 0.29 0.267 0.281 0.279333 0.01 159 0.290924 0.3 12 65.45 0.267 0.241 0.268 0.258667 0.01 5308 0.273975 0.3 13 1 16.45 0.139 0.125 0.178 0.147333 0.027465 0.174798 0.152 14 205.05 0.105 0.069 0.097 0.090333 0.01 8903 0.109237 0.152 15 287.05 0.072 0.061 0.092 0.075 0.01 5716 0.090716 0.152 16 369.15 0.065 0.058 0.06 0.061 0.003606 0.064606 0.152 17 451.15 0.043 0.052 0.054 0.049667 0.005859 0.055526 0.152 18 533.15 0.044 0.04 0.047 0.043667 0.003512 0.047179 0.152 19 615.15 0.041 0.043 0.043 0.042333 0.001155 0.043488 0.152 Table B . 5 . 1 — Triggering Summary for Loca t ion #2 Layer # Cm (K)sigma C R R F.o.S. Depth Soil Criteria Status C R R / C S R (meters) Type 1 1.1 1 0.24 0.5381 17 3.76428 Sand Robertson Liquefy 2 1.1 1 0.24 0.544629 5.2578 Sand Robertson Liquefy 3 1.1 1 0.1628 0.376852 6.72084 Sand Seed Liquefy 4 1.1 1 0.1628 0.388235 8.1534 Sand Seed Liquefy 5 1.1 1 0.1628 0.398044 9.0678 Silt Seed Liquefy 6 1.1 1 0.22 0.545906 9.46404 Sand Seed Liquefy 7 1.1 1 0.21 0.546875 10.71372 Silt Robertson Liquefy 8 1.1 1 0.21 0.591549 12.7254 Silt Robertson Liquefy 9 1.1 0.98 0.360067 1.1 14.67612 Silt Chinese No L i q . 10 1.1 0.97 0.3322 1.1 16.56588 Silt Chinese No L i q . 11 1.1 0.96 0.3168 1.134129 18.2118 Sand Seed No L i q . 12 1.1 0.92 0.3036 1.173711 19.94916 Sand Seed No L i q . 13 1.1 0.8 0.152 1.031674 35.49396 Ext. Silt Robertson No L i q . 14 1.1 0.69 0.152 1.682657 62.49924 Ext. Silt Robertson No Liq . 15 1.1 0.62 0.152 2.026667 87.49284 Ext. Silt Robertson No L i q . 16 1.1 0.56 0.152 2.491803 112.51692 Ext. Silt Robertson No L i q . 17 1.1 0.52 0.152 3.060403 137.51052 Ext. Silt Robertson No L i q . 18 1.1 0.5 0.152 3.480916 162.50412 Ext. Silt Robertson No Liq . 19 1.1 0.5 0.152 3.590551 187.49772 Ext. Silt Robertson No L i q . Table B .5 .2 ~ Triggering Summary for Loca t ion #2 ~ Cont inued 233 APPENDIX B Liquefaction Analysis Appendix B.5.2 — Location #3 LIQUEFACT ION ANALYSIS for CPT 91-3 Profile I Cyclic Stress Ratio (CSR) Layer # Depth Caltech Griffith Hughes Average Standard Mean + S (CRR)l (ft) Park Lake CSR Deviation + Sigma 1 13.75 0.495 0.488 0.46 0.481 0.01852026 0.49952026 0.17 2 18.65 0.489 0.481 0.453 0.47433333 0.01890326 0.4932366 0.16 3 23.25 0.473 0.464 0.432 0.45633333 0.0215484 0.47788173 0.168 4 27.65 0.45 0.44 0.406 0.432 0.02306513 0.45506513 0.168 5 32.25 0.427 0.415 0.385 0.409 0.02163331 0.43063331 Not App. 6 36.95 0.402 0.389 0.365 0.38533333 0.01877054 0.40410388 Not App. 7 42.55 0.372 0.36 0.344 0.35866667 0.00404754 0.37271421 Not App. 8 49.05 0.334 0.323 0.32 0.32566667 0.00737111 0.33303778 Not App. 9 55.45 0.297 0.285 0.3 0.294 0.00793725 0.30193725 Not App. 10 65.45 0.253 0.232 0.272 0.25233333 0.02000833 0.27234166 0.3 11 86.95 0.196 0.164 0.228 0.196 0.032 0.228 0.152 12 116.45 0.144 0.127 0.178 0.14966667 0.02596793 0.1756346 0.152 13 147.65 0.125 0.094 0.137 6. i 1866667 0.02218859 0.14085525 0.152 14 180.45 0.117 0.079 0.11 0.102 0.02022375 0.12222375 0.152 15 229.65 0.097 0.066 0.1 0.08766667 0.01882374 0.10649041 0:152 16 295.25 0.071 0.062 0.086 0.073 0.01212436 0.08512436 0.152 17 369.05 0.065 0.06 0.064 0.063 0.00264575 0.06564575 0.152 18 467.55 0.039 0.051 0.05 0.04666667 0.00665833 0.05332499 0.152 19 590.55 0.042 0.042 0.045 0.043 0.00173205 0.04473205 0.152 Table B.5.3 — Triggering Summary for Location #3 Layer # Cm (K)sigma CRR F.o.S. Depth Soil Criteria Status CRR/CSR (meters) Type 1 1.1 1 0.187 0.38877339 4.191 Sand Seed Liquefy 2 1.1 1 0.176 0.37104708 5.68452 Sand Seed Liquefy 3 1.1 1 0.1848 0.40496713 7.0866 Silt Seed Liquefy 4 1.1 1 0.1848 0.42777778 8.42772 Silt Seed Liquefy 5 1.1 1 Not App. 1.1 9.8298 Silt Chinese No Liq. 6 1.1 1 Not App. 1.1 11.26236 Silt Chinese No Liq. 7 1.1 1 Not App. 1.1 12.96924 Silt Chinese No Liq. 8 1.1 0.99 Not App. 1.1 14.95044 Silt Chinese No Liq. 9 1.1 0.95 Not App. 1.1 16.90116 Silt Chinese No Liq. 10 1.1 0.93 0.3069 1.21624835 19.94916 Sand Seed No Liq. 11 1.1 0.86 0.152 1 26.50236 ext. silt Robertson No Liq. 12 1.1 0.8 0.152 1.0155902 35.49396 ext. silt Robertson No Liq. 13 1.1 0.75 0.152 1.28089888 45.00372 ext. silt Robertson No Liq. 14 1.1 0.72 0.152 1.49019608 55.00116 ext. silt Robertson No Liq. 15 1.1 0.66 0.152 i.7338403 69.99732 ext. silt Robertson No Liq. 16 1.1 0.61 0.152 2.08219178 89.9922 ext. silt Robertson No Liq. 17 1.1 0.56 0.152 2.41269841 112.48644 ext. silt Robertson No Liq. 18 1.1 0.51 0.152 3.25714286 142.50924 ext. silt Robertson No Liq. 19 1.1 0.5 0.152 3.53488372 179.99964 ext. silt Robertson No Liq. Table B.5.4 ~ Triggering Summary for Location #3 ~ Continued 234 APPENDIX B Liquefaction Analysis Appendix B.5.3 — Location #4 LIQUEFACTION ANALYSIS for CPT 91-4 Profile j Depth Cyclic Stress Ratio (CSR) Layer # (ft) Caltech Griffith Hughes Average Standard Mean + S (CRR)l Park Lake CSR Deviation + Sigma 1 12.3 0.451 0.47 0.426 0.449 0.02206808 6.47106808 0.26 2 15 0.455 0.472 0.429 0.452 0.02165641 0.47365641 0.245 3 • 18.7 0.451 0.467 0.424 0.44733333 0.02173323 0.46906656 0.245 4 23.5 0.446 0.46 0.419 0.44166667 0.02084067 0.46250733 0.21 5 29.5 0.429 0.436 0.401 0.422 0.01852026 0.44052026 0.21 6 33.4 0.412 0.414 0.384 0.40333333 0.01677299 0.42010633 0.292 7 37.2 0.396 0.395 0.37 0.387 0.01473092 0.40173092 Not App. 8 43 0.372 0.365 0.351 0.36266667 0.01069268 0.37335934 Not App. 9 50.9 0.337 0.327 0.327 0.33033333 0.0057735 0.33610684 Not App. 10 59.9 0.292 0.276 0.303 0.29033333 0.01357694 0.30391027 Not App. 11 68.1 0.26 0.237 0.282 0.25966667 0.02250185 0.28216852 Not App. 12 75.6 0.238 0.209 0.263 0.23666667 0.02702468 0.26369135 Not App. 13 121.6 0.145 0.123 0.174 0.14733333 0.02557994 0.17291327 0.152 14 205.1 0.107 0.068 0.1 0.09166667 0.02079263 0.11245929 0.152 15 287.1 0.073 0.064 0.09 0.07566667 0.01320353 0.0888702 0.152 16 369.1 0.065 0.061 0.064 0.06333333 0.00208167 0.065415 0.152 17 451.2 0.042 0.054 0.053 0.04966667 0.00665833 0.05632499 0.152 18 533.2 0.044 0.04 0.051 0.045 0.00556776 0.05056776 0.152 19 615.2 0.04 0.042 0.04 0.04066667 0.0011547 0.04182137 0.152 Table B.5.5 ~ Triggering Summary for Loca t ion #4 Layer # Cm (K)sigma CRR F.o.S. Depth Soil Criteria Status CRR/CSR (meters) Type 1 1.1 1 0.286 0.636971047 3.74904 Sand Seed Liquefy 2 1.1 1 0.2695 0.596238938 4.572 Sand Seed Liquefy 3 1.1 1 0.2695 0.602459016 5.69976 Sand Seed Liquefy 4 1.1 1 0.231 0.523018868 7.1628 Sandy Silt Seed Liquefy 5 1.1 1 0.231 0.547393365 8.9916 Sandy Silt Seed Liquefy 6 1.1 1 0.3212 0.796363636 10.18032 Silty Sand Seed Liquefy 7 1.1 1 Not App. 1.1 11.33856 Silt Chinese Not Liq. 8 1.1 Not App. 1.1 13.1064 Silt Chinese Not Liq. 9 1.1 0.98 Not App. 1.1 15.51432 Silt Chinese Not Liq. 10 1.1 0.97 Not App. 1.1 18.25752 Silt Chinese Not Liq. 11 1.1 0.94 Not App. 1.1 20.75688 Silt Chinese Not Liq. 12 1.1 0.91 Not App. 1.1 23.04288 Silt Chinese Not Liq. 13 1.1 0.8 0.152 1.031674208 37.06368 Ext. Silt Robertson Not Liq. 14 1.1 0.69 0.152 1.658181818 62.51448 Ext. Silt Robertson Not Liq. 15 1.1 0.62 0.152 2.008810573 87.50808 Ext. Silt Robertson Not Liq. 16 1.1 0.56 0.152 2.4 112.50168 Ext. Silt Robertson Not Liq. 17 1.1 0.52 0.152 3.060402685 137.52576 Ext. Silt Robertson Not Liq. 18 1.1 0.5 0.152 3.377777778 162.51936 Ext. Silt Robertson Not Liq. 19 1.1 0.5| 0.152 3.737704918 187.51296 Ext. Silt Robertson Not Liq. Table B.5.6 — Triggering Summary for Loca t ion #4 ~ Cont inued 235 APPENDIX B Liquefaction Analysis Appendix B.5.4 — Location #7 LIQUEFACTION ANALYSIS for CPT 91-7 Profile I Depth Cyclic Stress Ratio (CSR) Layer # (ft) Caltech Griffith Hughes Average Standard Mean + S , (CRR)l Park Lake CSR Deviation + Sigma 1 9.86 0.252 0.277 0.247 0.25866667 0.01607275 0.27473942 0.182 2 14.06 0.378 0.414 0.369 0.387 0.02381176 0.41081176 0.182 3 20.56 0.427 0.461 0.413 0.43366667 0.02468468 0.45835134 0.182 4 26.56 0.414 0.439 0.402 0.41833333 0.01887679 0.43721013 0.182 5 32.06 0.375 0.394 0.377 0.382 0.01044031 0.39244031 0.182 6 40.36 0.309 0.308 0.34 0.319 0.01819341 0.33719341 0.1 7 50.86 0.262 6.236 0.307 0.26833333 0.03592121 6.36425454 0.1 8 60.66 0.228 0.2 0.278 0.23533333 0.03951371 0.27484704 0.1 9 72.16 0.196 0.183 0.248 0.209 0.03439477 0.24339477 0.223 10 83.66 0.176 0.166 0.223 0.18833333 0.03043572 0.21876906 0.23 11 100.06 0.16 0.145 0.193 0.166 0.02455606 0.19055606 0.22 12 118.66 0.141 0.128 0.162 0.14366667 0.01715615 0.16082281 0.18 13 144.86 0.131 0.097 0.13 0.11933333 0.0193477 0.13868103 0.152 14 205.06 0.103 0.071 0.107 0.09366667 0.01973153 0.1133982 0.152 15 287.06 0.075 0.066 0.088 0.07633333 0.01106044 0.08739377 0.152 16 369.16 0.067 0.064 0.07 0.067 0.003 0.07 0.152 17 451.16 0.043 0.051 0.057 0.05033333 0.00702377 0.0573571 0.152 18 533.16 0.04 0.041 0.052 0.04433333 0.00665833 0.05099166 0.152 19 615.16 0.04 0.038 0.045 0.041 0.00360555 0.04460555 0.152 Table B.5.7 - - Triggering Summary for Loca t ion #7 Layer # Cm j(K)sigma CRR i F.o.S. i Depth Soil Criteria Status iCRR/CSR i (meters) Type 1 1.1! 1 0.2002i 0.77396907! 3.005328 Sand Seed Liquefy 2 1.1! 1 0.2002i 0.51731266! 4.285488 Sand Seed ..Liquefy, „ 3 l.li 1 0.2002 i 0.461644891 6.266688 Sand Seed Liquefy 4 i.il i 6.2002! 6.47856574 ! 8.095488 Sand Seed Liquefy 5 i.'ii i 6.2002J 6.52468377 ! 9.771888 Sand Seed Liquefy 6 i.il i O.lii 6.34482759! 12.301728 Sand Seed Liquefy 7 i.il 0.98 6.1078! 6.40i739i3] i5.502i28 Sand Seed Liquefy 8 l.li 0.96 0.1056! 0.4487252H 18.489168 Sand Seed Liquefy 9 l.lj 0.9 0.22077i 1.05631579! 21.994368 Sand Seed Partial Liq 10 1.1! 0.89 0.22517! 1.19559292; 25.499568 Sand Rob/Seed Partial Liq 11 1.1! 0.85 0.2057! 1.23915663 i 30.498288 Sand Rob/Seed Partial Liq 12 l.li 0.8 0.1584! 1.10255221 36.167568 Sand Seed Partial Liq 13 i.i! 0.75 0.152 j 1.27374302 i 44.153328 Ext. Silt Robertson Not Liq. 14 i.il 6.69 0.152! 1.6227758! 62.502288 Ext. Silt Robertson Not Liq. 15 i.il 6.62 6.152! i.99126638; 87.495888 Ext. Silt Robertson Not Liq. 16 i.i! 0.56 6.152! 2.26865672! 112.519968 Ext. Silt Robertson Not Liq. 17 l.lj 0.52 6.152! 3 6i986755j i37.513568 Ext. Silt Robertson Not Liq. 18 l.li 0.5 0.152! 3.42857143! 162.507168 Ext. Silt Robertson Not Liq. 19 l.li 0.5 0.152! 3.70731707= 187566768 Ext. Silt Robertson Not Liq. Table B.5.8 -- Triggering Summary for Loca t ion #7 ~ Cont inued 236 APPENDIX B Liquefaction Analysis Appendix B.5.5 — Location #8 LIQUEFACTION ANALYSIS for CPT 91-8 Profile 1 Depth Cyclic Stress Ratio (CSR) Layer # (ft) Caltech Griffith Hughes Average Standard Mean + S (CRR)l Park Lake CSR Deviation + Sigma 1 19.6 0.489 0.547 0.483 0.50633333 0.03534591 0.54167924 0.134 2 26 0.474 0.523 0.465 0.48733333 0.03121431 0.51854765 0.134 3 32.4 0.448 0.485 0.438 0.457 0.02475884 0.48175884 0.134 4 38.8 0.418 0.44 0.409 0.42233333 0.01594783 0.43828116 0.134 5 43.3 0.393 0.408 0.389 0.39666667 0.01001665 0.40668332 0.23 6 48.6 0.363 0.372 0.368 0.36766667 0.00450925 0.37217592 0.125 7 55.8 0.323 0.325 0.343 0.33033333 0.01101514 0.34134847 0.125 8 66.4 0.272 0.261 0.308 0.28033333 0.02458319 0.30491653 0.1 9 75 0.237 0.214 0.277 0.24266667 0.03187998 0.27454665 0.168 10 79.5 0.217 0.194 0.261 0.224 0.03404409 0.25804409 0.125 11 84.3 0.2 0.176 0.245 0.207 0.03502856 0.24202856 0.125 12 87.5 0.187 0.167 0.235 0.19633333 0.03494758 0.23128091 0.134 13 126.6 0.144 0.116 0.164 0.14133333 0.02411086 0.16544419 0.152 14 205.1 0.109 0.072 0.106 0.09566667 0.02055075 0.11621742 0.152 15 287.1 0.079 0.067 0.088 0.078 0.01053565 0.08853565 0.152 16 369.1 0.064 0.063 0.068 0.065 0.00264575 0.06764575 0.152 17 451.1 0.044 0.053 0.054 0.05033333 0.00550757 0.0558409 0.152 18 533.2 0.044 0.041 0.053 0.046 0.006245 0.052245 0.152 19 615.2 0.039 0.039 0.041 0.03966667 0.0011547 0.04082137 0.152 Table B .5 .9 — Triggering Summary for Loca t ion #8 Layer # Cm (K)sigma CRR F.o.S. Depth Soil Criteria Status CRR/CSR (meters) Type 1 1.1 1 0.1474 0.29111257 5.97408 Sand Seed Liquefy 2 1.1 1 0.1474 0.30246238 7.9248 Sand Seed Liquefy 3 1.1 1 0.1474 0.32253829 9.87552 Sand Seed Liquefy 4 1.1 1 0.1474 0.34901342 11.82624 Sand Seed Liquefy 5 1.1 1 0.253 0.63781513 13.19784 Sand Seed Liquefy 6 1.1 1 0.1375 0.37398005 14.81328 Sand Seed Liquefy 7 1.1 0.98 0.13475 0.40792129 17.00784 Sand Seed Liquefy 8 1.1 0.96 0.1056 0.37669441 20.23872 Sand Seed Liquefy 9 1.1 0.92 0.170016 0.70061538 22.86 Sand Seed Liquefy 10 1.1 0.91 0.125125 0.55859375 24.2316 Sand Seed Liquefy 11 1.1 0.9 0.12375 0.59782609 25.69464 Sand Seed Liquefy 12 1.1 0.89 0.131186 0.66817997 26.67 Sand Seed Liquefy 13 1.1 0.8 0.152 1.1 38.58768 Ext. Silt Robertson Not Liq. 14 1.1 0.69 0.152 1.1 62.51448 Ext. Silt Robertson Not Liq. 15 1.1 0.63 0.152 1.1 87.50808 Ext. Silt Robertson Not Liq. 16 1.1 0.57 0.152 1.1 112.50168 Ext. Silt Robertson Not Liq. 17 1.1 0.52 0.152 1.1 137.49528 Ext. Silt Robertson Not Liq. 18 1.1 0.5 0.152 1.1 162.51936 Ext. Silt Robertson Not Liq. 19 1.1 0.5 0.152 1.1 187.51296 Ext. Silt Robertson Not Liq. Table B .5 .10 ~ Triggering Summary for Loca t ion #8 ~ Cont inued 237 A P P E N D I X B Liquefaction Analysis Appendix B.6 — Idriss (1991) Ground Motion Attenuation IDRISS A T T E N U A T I O N RELATIONSHIP (1991) : BETA1 = 2.475 BETA2 = -0.286 M = 7 R = 31.4 PERIOD A L P H A 0 ALPHA1 A L P H A 2 BETA0 ERROR L N ( P G A ) (m+1) SEC M<7.25 (PGA) 0 -0.05 3.477 -0.284 0 0.41 -1.4296761 0.23938644 0.03 -0.05 3.477 -0.284 0 0.41 -1.4296761 0.23938644 0.05 -0.278 3.426 -0.269 0.066 0.41 -1.1517156 0.31609401 0.075 -0.308 3.359 -0.252 0.07 0.41 -0.916233 0.40002308 0.1 -0.318 3.327 -0.243 0.072 0.44 -0.7331828 0.48037759 0.11 -0.328 3.289 -0.236 0.073 0.44 -0.683016 0.50509133 0.13 -0.338 3.233 -0.225 0.075 0.44 -0.5760612 0.56210805 0.15 -0.348 3.185 -0.216 0.076 0.44 -0.5027961 0.60483713 0.2 -0.358 3.1 -0.201 0.078 0.44 -0.3972815 0.67214482 0.25 -0.429 3.034 -0.19 0.08 0.44 -0.4002787 0.67013325 0.3 -0.486 2.982 -0.182 0.082 0.46 -0.4073719 0.66539672 0.35 -0.535 2.943 -0.177 0.087 0.46 -0.4587012 0.63210407 0.4 -0.577 2.906 -0.173 0.092 0.46 -0.5302441 0.5884613 0.5 -0.648 2.85 -0.169 0.099 0.48 -0.7040574 0.49457452 0.6 -0.705 2.803 -0.166 0.105 0.48 -0.8733475 0.41755145 0.7 -0.754 2.765 -0.165 0.111 0.5 -1.0362257 0.35479124 0.8 -0.796 2.728 -0.164 0.115 0.5 -1.2103226 0.2981011 0.9 -0.834 2.694 -0.163 0.119 0.5 -1.361894 0.25617511 1 -0.867 2.662 -0.162 0.123 0.5 -1.4958116 0.22406668 1.5 -0.97 2.536 -0.16 0.136 0.5 -2.035941 0.13055757 2 -1.046 2.447 -0.16 0.146 0.54 -2.3834324 0.09223346 3 -1.143 2.295 -0.159 0.16 0.54 -2.9341052 0.05317828 4 -1.177 2.169 -0.159 0.169 0.54 -3.3186894 0.03620025 5 -1.214 2.042 -0.157 0.177 0.54 -3.6313479 0.02648047 Table B.6 .1 - Est imat ion o f Target Spectrum for Mod i f i ca t i on o f Time-Histor ies Refer to Idriss (1991) for a general summary o f spectral ordinates used i n the calculation. 238 APPENDIX C POST-EARTHQUAKE STABILITY C.l Flowslide Analyses: Limit Equilibrium Stability at Offshore Locations Since the tunnel is founded on a level excavation, the tunnel is very stable i n the direction transverse to the roadway (i.e., cross-section view). In each analysis at the offshore locations, the failure surface intersects an underlying l iquefied layer. A t locat ion #2 (figure C . 1.1), the failure surface intersects the unit #5 loose sand. (Refer to figure 6.13 for the material numbering scheme). A factor o f safety o f about 2.3 was estimated i n the l imi t equl ibr ium analysis, i n w h i c h a residual strength o f 12 k P a was applied i n the underlying l iquefied sands. A f lowslide type movement at locat ion #2 is un l ike ly . A t locat ion #3 (figure C.1.2) , the failure surface intersects the unit #5 l iquefied silt. A residual strength o f 30 k P a was applied i n the silt. A factor o f safety against f lowsl ide ( F . S F L ) o f about 1.3 was estimated i n the stability analysis. A f lowsl ide at locat ion #3 is unl ikely . S imi la r ly to locat ion #3, the failure surface at locat ion #4 intersects the unit #5 l iquefied silt. A residual strength o f 30 k P a was applied i n the silt. A factor o f safety against f lowsl ide ( F . S F L ) o f about 1.6 was estimated i n the stability analysis. Based on these analyses, it can be concluded that residual strengths are sufficient to prevent post-earthquake instabil i ty at a l l o f the locations analyzed. 239 240 o o •«T cn I I I O J — . I I I I I I I I I I I m > to c >» cn 1—1 4-* c c •#4 o t— o o l-H a. 4-1 •»H 4-J C J u J 3 CD cn ID 4-» *J > c ID cu on o 1 cn CL CO CO in i o cr CO :>• i 't~t ID >• IO 4-1 c-X C J 1— ID xz cu ID C . . O l m <z 4— 4— ID •r4 =». c OJ o 1 X Z J C cn e= h— cr >. o C D »*— o c_ • i io >* _ i 4» -L-l m 4J cu i cn C J ID cn cu C D U CO CO c to tn O ID o 1 <t 3 : 1 z Q . z X z 3 az o M CU C_ o o o o o o o o o o cn cu TJ o o c z o o o to o n « nj ni o o o •»•« o co « n in m n m in CM in i n i n CO CT) 03 CT) CD CTl 1—1 c 1—1 * J *J T 3 o C • ^ 4 ~ H L L . • * 4 I D 4 - » • — 1 •o cn cn cn cn I D T 3 cz 4 - J * J L L . cu C I D .—1 . c -L-l I D CO • • H CT C T cr C T cu cu C O cn • 1 4 [= CU c_ CU _ i —i i—i 1—1 :> C J cn i i 1 "O 1 0 cz •a o C T c e c c CU c o o o o o o C O U J C_] X _ l —i z z z z . o in i i 242 APPENDIX C Post-Earthquake Stability C.2 Flowslide Analyses with Higher Residual Strengths This appendix summarizes the graphic output using a more realistic residual strength estimate of 26 kPa in the overlying dyke material at the north and south river banks (locations #8 and #7). Based on the results of these analyses, it can be concluded that the north and south shores are stable. 243 244 245 APPENDIX C Post-Earthquake Stability C.3 Residual Strength (Sr) in Sands - Stark & Mesri (1992) The Stark & M e s r i method o f estimating residual strength is discussed i n section 4.3.2. This appendix provides more details about their procedure. Stark & M e s r i based their findings on comparison of: - back-calculations o f shear strength from case-histories ( including the Seed & Harder (1990) data base) o f l iquefaction failure. (Back-calculated shear strengths were normal ized w i th respect to pre-liquefaction overburden pressure.), and - the cyc l i c shear stress, at 15 equivalent cycles (earthquake magnitude, M=7.5) causing liquefaction (i.e., y i e l d strength o f the soil) ~ estimated using Seed's chart (Seed et al . 1984 — refer to section 4.2.1). (1.6 # Measured SPT and Critical Strength Data 0 Estimated SPT and Critical Strength Data 0.S- 0 Construction-induced - Estimated Data 0.4-0.3 -S,(Yield, Moll) 0.2- • / a. « / M=7.5 0.1- • r / / • V * 0 10 20 30 40 50 60 Equivalent Clean Sand SPT Blowcount Figure C.3.1 - Comparison of Undrained Critical Strength Ratios and Yield Strength Ratios Back--calculated from Field Case-histories (after Stark & Mesri, 1992) Some data used by Seed & Harder (1990) i n developing their S r - correlation (figure C.3.1) is based on soils that had enough time to drain after the post-liquefaction f l o w ; therefore, b y the time the l iquefied mass came to rest, the residual strength was actually greater than the in i t ia l y i e l d strength that controls the triggering o f l iquefaction (Stark & M e s r i , 1992). 246 A P P E N D I X C Post-Earthquake Stability C A Residual Strength -- Seed & Harder (1990) This approach is based on back-analysis o f liquefaction case histories where values o f the residual strength were calculated for so i l zones i n w h i c h Standard penetration test (SPT) results were available. The Seed & Harder (1990) residual strength magnitudes were back-calculated using l imi t equi l ibr ium analyses, the f inal geometry o f the slide mass, and varying failure surfaces to approximate the lower bound estimate o f the residual strength. It shows a large scatter i n its data, so at l o w b l o w count magnitudes (i.e., less than ( N , ) 6 0 = 10) there is significant uncertainty i n the S r estimates. Overa l l , b l o w count magnitudes at the tunnel site are low, so results f rom the Seed correlation can be considered unreliable and, therefore, have only been provided as a reference. The Seed & Harder (1990) (figure C.4.1) predictions for l iquefied layers at larger depths w i l l be conservative because the large confining stresses (at greater depths) w i l l not be acknowledged. 2000 1600 1200-/ 800-Residual Strength (^psf) jf/J 400-Tfr^m f Lower San Fernando Dam 0 4 8 12 16 20 24 28 Equivalent Clean Sand (!\)60 Figure C.4.1 - Relationship Between Residual Strength and (N , ) 6 0 (after Seed & Harder, 1990) 247 APPENDIX C Post-Earthquake Stability S r - estimates using the Seed & Harder (1990) correlation are summarized i n table C .4 .1 . (NO*, Corrected (lSt)60 S r (psf) S r (kPa) 5% 15% 35% 5% 15% 35% 5% 15% 35% 5 • 5 6 8 50 (0-225) 100 (25-325) 200 (75-425) 2.39 (0-10.8) 4.79 (1.2-15.6) 9.58 (3.6-20.4) 10 10 11 13 325 (175-525) 400 (200-625) 530 (325-800) 15.56 (8.4-25.2) 19.2 (9.6-30.0) 25.38 (15.6-38.4) 15 15 16 18 700 (450-925) 790 (500-1050) 1010 (760-1200) 33.52 (21.6-44.4) 37.83 (24.0-50.4) 48.36 (36.5-57.6) Table C.4.1 - Res idua l Strength Estimates -- Seed & Harder (1990) Note: Values in brackets indicate the upper and lower bounds of the Sr predictions 248 APPENDIX D EMPIRICAL METHOD DISPLACEMENT PREDICTIONS Appendix D.l — Bartlett/Youd: Analysis Details Appendix D.l.l - Displacement Predictions with Epicentral Distance of 30km and Varying Ground Slopes Layer # M R (km) S (%) T 1 5 (m) F 1 5 (%) (D 5,) 1 5 (mm) Log(Dh+0.01) (m) (m) 1 7 30 1 6 5 0.3 -0.353 0.434 2 7 30 1 4 45 0.1 -1.304 0.0497 Total 0.484 1 7 30 3 6 5 0.3 -0.148 0.711 2 7 30 3 4 45 0.1 -1.100 0.0794 Total 0.790 Table D. l . l - Bartlett/Youc Parameters and Displacement Predictions at Location #2 ~ R=30km Layer # M R (km) S (%) T 1 5 (m) F 1 5 (%) (D S 0) 1 5 (mm) Log(Dh+0.01) (m) (m) 1 7 30 1 2 80 0.3 -3.582 2.6e-4 1 7 30 3 2 80 0.3 -3.378 4.2e-4 D. 1.2 - Bartlett/Youd Parameters and Displacement Predictions at Location #3 ~ R=30km Layer # M R (km) S (%) T 1 5 (m) F 1 5 (%) (D5o),5 (mm) Log(Dh+0.01) (m) (m) 1 7 30 1 2 80 0.3 -1.781 0.0166 1 7 30 3 2 80 0.3 -1.576 0.0265 D. 1.3 - Bartlett/Youd Parameters and Displacement Predictions at Location #4 — R=30km 249 APPENDIX D Empirical Methods Layer # M R S T 1 5 F 1 5 (Dso),5 Log(Dh+0.01) (km) (%) (m) (%) (mm) (m) (m) 1 7 30 1 6 5 0.3 -0.353 0.434 2 7 30 1 8 30 0.17 -0.790 0.162 Total 0.596 1 7 30 3 6 5 0.3 -0.148 0.711 2 7 30 3 8 30 0.17 -0.585 0.250 Total 0.961 Table D.1.4 - Bartlett/Youd Parameters and Displacement Predictions at Location #8 - R=30km Appendix D.1.2 — Disnlacement Predictions with Enicentral Distance of 60km and Varying Ground Slopes Layer # M R (km) S (%) T 1 5 (m) F 1 5 (%) (D5o)i5 (mm) Log(D„+0.01) (m) D h (m) 1 7 60 1 6 5 0.3 -1.022 0095 2 7 60 1 4 45 0.1 -1.973 0.0106 Total 0.1056 1 7 60 3 6 5 0.3 -0.817 0.152 2 7 60 3 4 45 0.1 -1.769 0.017 Total 0.169 Table D. 1.5 - BartlettA^ ouc Parameters and Displacement Predictions at Location #2 — R=60km Layer # M R (km) S (%) T 1 5 (m) F 1 5 (%) (D5o)i5 (mm) Log(Dh+0.01) (m) D h (m) 1 7 60 1 2 80 0.3 -4.251 6e-5 1 7 60 3 2 80 0.3 -4.047 9e-5 D.1.6 - Bartlett/Youd Parameters and Displacement Predictions at Location #3 ~ R=60km 250 APPENDIX D Empirical Methods* Layer # M R S T 1 S F 1 5 (Ds«)i5 Log(Dh+0.01) (km) (%) (m) (%) (mm) (m) (m) 1 7 60 1 2 80 0.3 -2.450 0.0035 1 7 60 3 2 80 0.3 -2.245 0.0057 D. 1.7 - Bartlett/Youd Parameters and Displacement Predictions at Location #4 - R=60km Layer # M R (km) S (%) T 1 5 (m) F 1 5 (%) (D5«)l5 (mm) Log(Dh+0.01) (m) (m) 1 7 60 1 6 5 0.3 -1.022 0.095 2 7 60 1 8 30 0.17 -1.459 0.035 Total 0.13 1 7 60 3 6 5 0.3 -0.817 0.152 2 7 60 3 8 30 0.17 -1.254 0.056 Total 0.21 Table D.1.8 - Bartlett/Youd Parameters and Displacement Predictions at Location #8 ~ R=60km 251 APPENDIX D Empirical Methods Appendix D.2 — Hamada: Analysis Details Appendix D.2.1 — Model Details The Hamada model was developed using pre- and post-earthquake aerial photographs. From these, vector maps of liquefaction induced ground displacements were developed based on ground deformation patterns witliin areas of similar surface topography. Figure D.2.1 shows a displacement vector map for part of the Niigata site. Legend Displacement Vector • SPT Borehole 100 m I 1 Figure D.2.1- Displacement Vectors and SPT Boreholes for Part of Niigata, Japan Analysis by Hamada (From Bartlett & Youd, 1992) 252 APPENDIX D Empirical Methods Appendix D.2.2 — Hamada Displacement Predictions with Varying Ground Slopes Layer # Thickness (m) Slope (%) Displacement (m) 1 6 1 1.84 2 8 1 1.5 Total 3.34 1 6 3 2.64 2 8 3 2.16 Total 4.80 Table D.2.1 — Hamada Parameters and Predictions at Location #2 Layer # Thickness (m) Slope (%) Displacement (m) 1 2 1 1.06 1 2 3 1.52 Table D.2.2 ~ Hamada Parameters and Predictions at Location #3 Layer # Thickness (m) Slope (%) Displacement (m) 1 . 2 1 1.06 1 2 3 1.52 Table D.2.3 ~ Hamada Parameters and Predictions at Location #4 253 APPENDIX D Empirical Methods Layer # Thickness Slope Displacement (m) (%) (m) 1 6 1 1.84 2 8 1 2.12 Total 3.96 1 6 3 2.64 2 8 3 3.05 Total 5.69 Table D.2.4 — Hamada Parameters and Predictions at Location #8 254 APPENDIX D Empirical Methods Appendix D.3 « Tokimatsu/Seed: Analysis Details Layer # Thickness (m) CSR Volumetric Strain (%) Settlement (m) 1 2 0.403 13 2.1 0.042 2 4 0.327 13 2.1 0.084 3 4 0.259 5 4 0.16 Total 0.286 Table D.3.1 - Tokimatsu/Seed Parameters and Predictions at Location #2 Layer # Thickness (m) CSR (N, )„ Volumetric Strain (%) Settlement (m) 1 2 0.41 8 2.9 0.058 Table D.3.2 - Tokimatsu/Seed Parameters and Predictions at Location #3 Layer # Thickness (m) CSR (N,)«o Volumetric Strain (%) Settlement (m) 1 2 0.42 12 2.2 0.044 Table D.3.3 - Tokimatsu/Seed Parameters and Predictions at Location #4 Layer # Thickness (m) CSR (NO* Volumetric Strain (%) Settlement (m) 1 2 0.33 11 2.4 0.048 2 4 0.28 9 2.7 0.108 3 4 0.22 11 2.4 0.096 4 4 0.2 11 2.4 0.096 Total 0.348 Table D.3.4 - Tokimatsu/Seed Predictions at Location #8 255 A P P E N D I X E Additional Figures & Charts Used in Analyses (70 ^ 3 tsf -I 1 1 1 1 0.1 0.2 0.3 0.4 0.5 rx Figure E. 1: Ranges in K„ Factors (Seed & Harder, 1990) 256 2.0 1.5-Volumetric Strain Ratio c 1.0-X 0.5 J i i i i i i 0.0 5.25 6.0 6.75 7.5 8.0 8.5 Magnitude of Earthquake (M) Figure E.2: Relationship between Volumetric Strain Ratio and Number of Cycles (Earthquake Magnitude) (After Tokimatsu and Seed, 1984) 257 APPENDIX E Additional Figures & Charts Earthquake Magnitude (M) No. of Representative Cycles at 0.65 T 0 Magnitude or Duration Correction Factor: K M 8.5 26 0.89 7.5 15 1.0 6.75 10 1.13 6.0 5 - 6 1.32 5.25 2 - 3 1.5 Table E . l : Correction Factors for Magnitude (after Seed et al., 1984) 258 APPENDIX F Remediation Estimations For Each Location Depth SOIL CSR Ksig Km (CRR)req [(Nl)60]req (meters) TYPE 5% 15% 35% 0.8 Sand 0.47 1 1.1 0.51272727 30 23 20 2.3 Sand 0.46 1 1.1 0.50181818 30 23 20 3.7 Sand 0.46 1 1.1 0.50181818 30 23 20 5.2 Sand 0.44 1 1.1 0.48 30 23 20 6.1 Silt 0.43 1 1.1 0.46909091 30 23 20 6.5 Sand 0.43 1 1.1 0.46909091 30 23 20 7.7 Silt 0.41 1 1.1 0.44727273 30 23 20 9.7 Silt 0.38 1 1.1 0.41454545 29 22 19 Table F.l — Estimation of Blowcounts Required to Prevent Liquefaction at Location #2 Depth SOIL CSR Ksig Km (CRR)req [(Nl)60]req (meters) TYPE 5% 15% 35% 0.8 Sand 0.5 1 1.1 0.54545455 30 23 20 2.3 Sand 0.49 1 1.1 0.53454545 30 23 20 3.7 Silt 0.48 1 1.1 0.52363636 30 23 20 5.0 Silt 0.46 1 1.1 0.50181818 30 23 20 Table F.2 ~ Estimation of Blowcounts Required to Prevent Liquefaction at Location #3 Depth SOIL CSR Ksig Km (CRR)req [(Nl)60]req (meters) TYPE 5% 15% 35% 0.3 Sand 0.47 1 1.1 0.51272727 30 23 20 1.1 Sand 0.47 1 1.1 0.51272727 30 23 20 2.2 Sand 0.47 1 1.1 0.51272727 30 23 20 3.7 Silt 0.46 1 1.1 0.50181818 30 23 20 5.5 Silt 0.44 1 1.1 0.48 30 23 20 6.7 Silt 0.42 1 1.1 0.45818182 30 23 20 Table F.3 ~ Estimation of Blowcounts Required to Prevent Liquefaction at Location #4 259 APPENDIX F Remediation Estimations Depth SOIL CSR Ksig Km (CRR)req [(Nl)60]req (meters) TYPE 5% 15% 35% 0.3 Sand 0.28 1 1.1 0.30545455 26 20 17 1.6 Sand 0.41 1 1.1 0.44727273 30 23 20 3.6 Sand 0.46 1 1.1 0.50181818 30 23 20 5.4 Sand 0.44 1 1.1 0.48 30 23 20 7.1 Sand 0.39 1 1.1 0.42545455 30 23 20 9.6 Sand 0.34 1 1.1 0.37090909 28 21 18 12.8 Sand 0.31 0.98 1.1 0.34508349 28 21 18 15.8 Sand 0.28 0.96 1.1 0.31818182 26 20 17 19.3 Sand 0.25 0.9 1.1 0.3030303 26 20 17 22.8 Sand 0.22 0.89 1.1 0.26966292 24 19 16 27.8 Sand 0.19 0.85 1.1 0.24385027 22 17 14 33.5 Sand 0.16 0.8 1.1 0.21818182 20 15 13 Table F.4 ~ Estimation of Blowcounts Required to Prevent Liquefaction at Location #7 Depth SOIL CSR Ksig Km (CRR)req [(Nl)60]req (meters) TYPE 5% 15% 35% 1.0 Sand 0.54 1 1.1 0.58909091 30 23 20 2.9 Sand 0.52 1 1.1 0.56727273 30 23 20 4.9 Sand 0.48 1 1.1 0.52363636 30 23 20 6.8 Sand 0.44 1 1.1 0.48 30 23 20 8.2 Sand 0.41 1 1.1 0.44727273 30 23 20 9.8 Sand 0.37 1 1.1 0.40363636 29 22 19 12.0 Sand 0.34 0.98 1.1 0.37847866 29 22 19 15.2 Sand 0.31 0.96 1.1 0.35227273 28 21 18 17.9 Sand 0.28 0.92 1.1 0.33201581 28 21 18 19.2 Sand 0.26 0.91 1.1 0.31168831 26 20 17 20.7 Sand 0.24 0.9 i . i 0.29090909 26 20 17 21.7 Sand 0.23 0.89 1.1 0.28192033 25 19 18 Table F.5 - Estimation of Blowcounts Required to Prevent Liquefaction at Location #8 260 APPENDIX G Available Soil Data APPENDIX G . l CPT & Borehole Soil Profiles and Summaries M a s e y O p e r a t o r i G I L L E S P I E . S o u n d i n g . 1 6 P9 1 / 1 Cone U s e d . NS10D S E I S M I C T i 11—1 r—1 e 1 CPT D a t a . 9 1 / 0 3 / 1 4 13i 30 L o c a t i o n t MASSEY 9 1 - 2 J o b N o . 1 Z TIP RESISTANCE ' Dc (HN/n-2) 0 .V 0 j — • , 1—,—L 20 3)1 '• 1 ' FRICTION RATIO fs/0 CO 0 S PORE PRESSURE P», Cbor) .0 ' 10 D e p t h I n c r e m e n t 1 . 0 5 tn Max D e p t h t 18 .90 m 263 Klo bATA -Z, - 3 I /f\ \%S •» t i " ills /cufw. ft is u*eu-UiK uht~ fa far , M _H L ,\ y L^Q? ! t 0 ^ - - T 2 « 13 *15 3r M.A. N.« 5oo J i5_ 3 t.lo f " Ifeo (wuur) 7 if 7 \~?% ti €y r~ T-t>* SariA) s;/f .: i ? g . i s o*».ttfl 7 * 7 \7# t1 i o * ~ ~ sw>_. t* : s t ^ 1 ^ :$;Ha-Sa.rx<liY^^s-%) n,0 34.77 tW..w) g g nT 1 3 M N-A. MA ~ iso w.s [••'"O - i d u.n» 97 - „ CLju, Si If /SgkA _ »»•»• 5 _ _ , ^ M A *>-A- I-1 « Ha? M.o [„.«tj fed*). \\^) ' - u -f\J.«-P/aif-;c (B.H.#7t - 13 1 5 5 5 ow.™; i N . A . J N.A. W-A- o.!5 4.1 [coe \<\.o • 15 [u.?Cf] W5 -16 n 2H° Ozw^  ^ 1 17 180 - n '°"> H-5 ^ ^ JSS _ _ .J**''r"^ 264 CPr 7/-2 V/s ( • A C * ) Y AM. »J.1 115-2 2 -•2\ 1M . t 1 — 5o [S2.02S] - 7 ? 100 — I3< — — -|7f ~ / , s f , r . , 7 7 - 7 s •• / s s 7 - 7 s / ^ 7 " i w fieisnczAie 2 6 5 M CD S 3 S 3 £ O p e r a t o r t G I L L E S P I E S o u n d i n s • 16 P g 1 / 1 Cona U s e d • NS10D SEISMIC T I I I 1 I I €==3 1 CPT D a t a t 9 1 / 0 3 / 1 5 14tDO L o c a t i o n i MASSEY 9 1 - 3 J o b No . i Z TIP RESISTANCE Cc CMN/n-2 IS' t7,<? 20' • • 20 FRICTION RATIO Fs/0 a) 0 S PORE PRESSURE Pw (bor) 0 10 10' 20-D e p t h I n c r a m o n t i . 0 5 n Max D e p t h i 18. 20 ra V* 1^ H Wi 4' Vr i OC.rV. _ * No DATA - 3 — 3." 5" . • y i trains ' • AO -— - - - i 1 " 3<u £-7 6 e N.A. IM- I? IJ no W-<>_% C tt9t} bvJ) 7 7 : M : — — - — • : • e/yeyJ,// . ... ..^SVi,- £wof.o) — 1 . 3 7-5 110 I1.<? ^ /jet- B.H.*S ' • leu - V .•'/ Sub® 6 W-A- 5 N A . . _ — ' 6 5 ) - 13 n% -- I H  v / /-Jj — A/on - /iful/ia ti£ » /[/ex/ ^ / £ -Sxfrttfc/afo/ layer JT^pct/s 2 / « 2 6 7 5ltT I St. J| *7.ff xi.il 310 -101.11 tb.Xt Sec -(zm.i/s) 2/t¥ noD-//S.71 iWf <f oo -3*1.18 I31-H3 IZ50-StCT - l *MJ-S.eUt [VltfSTOttHt] MS-P.i*t U3.7/ x 2*4.10 C*ii*Ki) l/.W pin CSXUT] /32e.ee 771.60 \ l/o.3i (ztsoo.) If0.19 M i l * , / A J ¥ (330S9tr> H25-Itoo-2 500 35<x> 5SOO-t>5oo -Bedrock f>p}£. c^fWxfe.;^ 700^ bo \<*J> -to ll.o |«0 -—i-lto n.c (nun) - - loo 5.1-6 -(32tJ>(--3oO 2 1 ? -HoO X I - 5 -5oo 2 1 . 5 21.5 268 M • s s G y Operator i GILLESPIE Sounding t 19 P g . l / 2 Cone Uced « NS10D SEISMIC ~ T ~ i 11—i m ^  1 CPT Data • 91/03/16 12:30 L o c a t i o n i Mascey 91-4 Job No. t 2 TIP RESISTANCE 0c <HN/n*2) FRICTION RATIO Fs/0 CO 0 S PORE PRESSURE P* (bar) 0 ]0 10' Depth Increment i . 05 m Max Depth i 24.55 m 269 M a s s e y Oparator t GILLESPIE Sounding i 19 Pg 2 / 2 Cone Used t NS10D SEISMIC T U I — I I — I 1 CPT Data i 91/03/16 12i 30 L o c a t i o n i Massey 91-4 Job No. i Z CO L, ai +» QI E CL LU a TIP RESISTANCE 0c QWn'2) FRICTION RATIO Fs/0 CO 0 S 20 PORE PRESSURE Pw (bar) 0 10 3 25 30 35' 40-Depth Increment • . 05 m Max Depth i 24. 55 m 270 Sr, GM« I Rf M, |v o-fcii W 4 3 ~"3'y ft, — <jl •7TA.~IO7A t < V ~ ^ y Ho -8 4JL 55 Ho' =£» 2-5" /? IS"*-** . ! • » ' " ' 151 " „ . c-THfo-102-6 »36 - -13. If-W5 -175 -|.| b.5 1500 K.5 - - J t D.%5 %5 I3S0 • - i i - * - 2.0 2»y 2". - 23-AJ.A MA iso igc. (b.ntf) - 24 - 2 ? - 30 271 Coyvfr 7^ .7 2'-iW $75 -62? 875 -4 24.1 fj«q.«17e] • J o ~7S • loo HZ5" 1*75 1 5 0 / /• s r ' /67./ Uxs-175 ~7~7r~7~~7~7 /" ^ s s s s s s ? 7~ 272 M C 3 £=3 s <=> x Operator i GILLESPIE Sounding i 41 Pg 1 / 2 Cona Used i NS10D SEISMIC T u i — I I — i e • 1 CPT Data . S l / 0 4 / 1 0 lOi 45 L o c a t i o n . MASSEY 91-07 Job No. i Z TIP RESISTANCE Oc CNN/n'2) FRICTION RATIO Fs/D CO 0 S 1«H 15H PORE PRESSURE P* (bar) 0 10 KH J5H 2TJJC3L 1 tm Depth Increment i . 05 m Max Depth t 38.30 m 273 M a s s s >/ Operator i GILLESPIE Sounding i 41 Pg 2 / 2 Cone Used t NS10D SEISMIC T u i — I I — i e 1 CPT Data . 91/04/10 lOi 45 L o c a t i o n i MASSEY 91-07 " Job No. i Z TIP RESISTANCE Oc (HN/m'2) FRICTION RATIO Fs/O CO 0 5 (li L. OJ +> a) E CL LU a 20-PORE PRESSURE P» (bar) 0 10 25H 3CH 35) 40-4 l i Depth Increment t . 05 m Max Depth i 36.30 m 274 STUTURAPHY BASU **L QrT 11-7 (u.7i<e 3.5m) ^ '' ' -_ * Kj.*f7~SZ :•— _D _u» _ _ * _ _ ' ft — -.110 {SHMI) I'8 Z** • * - - « - - 45 35-_ l i o (so*.iti) K^-Ji tt _ 7-t 7-2 III -25" 35 1 2 8 -17 • » » £"J.4r) _____ .,70 ^L^-\ g IIS" - • / f » -V 170- 33 rto-S;/6%(*,S<7.) " ***** 7 MJ. 7 zee - £ 0 3/ s ' S<*"<S (loost. -fttJi'utri ) jm*Sju&.£fiL \%o _ _ ;_ ___ _. _: -_ z*° _ ISO II < ,/^7 ./*7;-i*- i'» f , ,' k 218 - 5"0 37 28J- 35 no 210 kx=33.-7<> 3oo - \ , 3 __J1»_55- 3 7 1 1 3 2 o -13 W 37 -12 55P /8-0-\<\r~ -16 . S P o ( , . ( « ) ----- ::.}~ 20 (13. ISV) -32-5 - 2 1 (l.r«j) - 2 ? JIOP /g.o -30.-30 l\ MI ^ _ -32 • — 3J.«5 -3^, 0 275 Crrii-7 Sjrir Ms (Vn= iso~ys) Ml 2.7S.9 (Wt,.27<>) 96 (2DCT.799) (Z37*9i7j /V'3.1 (z<ito.?</<>) 0*104*3) 31.1 Wo -t7S-U2S-/37S-UXS-/f.o [n-m] --T5 .ICS •ISO U7.f (3^p.9e7) /87S- /7.e 276 X M a s s s Operator • GILLESPIE Sounding i 13 Pg 1 / 2 Cona Uced > NS10D SEISMIC 1~ L_J I I H Q 1 CPT Data . 91/05/15 11.45 L o c a t i o n . MASSEY 91-8 Job No. . Z TIP RESISTANCE FRICTION RATIO Fe/0 CO 0 5 10H 15 20 PORE PRESSURE Pw (bar) _0 10 0-5- • ID- • T { I. IS-• 2nl Depth Increment • .OS n Max Depth i 27.15 m 277 M a s s e y Operator i GILLESPIE Sounding i 13 Pg 2 / 2 Cone Uced . NS10D SEISMIC X L_J m r ~ i e 1 CPT Data . 91/05/15 11. 4S L o c a t i o n . MASSEY 91-8 Job No. . Z m L Ql +> Ql E 0_ LU a 30H 35H TIP RESISTANCE Oc WN/n'2) 20 F R I C T I O N R A T I O FE/0 C O 0 5 4QJ——t-30' 35' 40 Depth Increment > . 05 m PORE PRESSURE Pw (bar) 0 10 30H 3SH 40 Max Depth . 27 .15m J5 TXATlSfiArHY DAseo B  V, W ! *» >t Ni 4' Pr * Y -2m la . .170 0*»«"7)-S1.*»«* ^ 7 Uj«r<D _ /C=J£.V/_ • ~ • lr- (137 5')-'W&*^*"' -s.e / /s\ - ' 7 ° f'2'*"7; \t\0-158 -l .JS I - ) L 6) - P o Os"t-tti) ^ j W ^ . g / r g ; no Oi^jny^tssti g <*» fci ts J7~ 720 - I - I 2 . E -It 7/57V./TTT' 1 1 " ^07_50 37 F J X t 1 , 0 0 no J*l™*__. (v.t>c) -lb (k.Sbl) . Oou.u7ysh&M -no 18 *3<» -234 -JS 35 -22 _ , _ & (noijn) 85« •fr/fr J W hoy'f5%% 7-1 " 8 7 " .27, J o ; : ^ ! . . ; __7oo _^ -2? - 50 279 0%) m.57 2/7.6 %Stjtt 1U.7.S Z7B.1 75-7 721. f /vs./ (zlse.lfO) (fife, (,2s) U7./ G»taxx . ' :(&)*** : y \ 325 -(,35. 275 1/15-1375-'4—1. WAS •so • 75 lS7f- 2.l.t,H /l.p — too -I as — ISO — I7S -2JOO r*\ PieZS7VtZHE 280 m IOCON | O F F I C E R E P O R T ON SOIL EXPLORATION -4" PIPE GT H C O N T R A C T . Jt2?7.TUNNEt B O R I N G 1 J^*_?._i«J_}oATlJ« GE8QEBC C A S I N O . . . . S O R I N G D A T * M A R C H < 6 - 2 2 > K ? 5 6 w g P O » T D A T E APJU_.^»"5» , L » * f c C O M P I L E D BY b _ H _ , C H E C K E D BY f A W U R H A M M E R W T . . _ J 4 _ . L « . D R O P 3Q_ 1 N C H C 8 ( P E N E T R A T I O N R E S I S T A N C E S C O M V - R T - O T O B L O W S O f 4>OQ IN . L B S . E N E R O Y I SAMPUE CONDITION S A M P L E TYPES . ABBREVIATIONS . . A * . • A U O E R S A M P L E P J . . F O I L S A M P L E V • IN-SITU V A W E T E S T * - WET U N I T W E I G H T S . T . • S L O T T E D T U B E W . S . . W A S H E D S A M P L E C O . - D R I V E - O P E N O.P . D R I V E - T O O T V A L V E C S . ' C H U N K S A M P L E SJO. . S L E E V E 4 > P K N S . F . - S L E C V C - P O O T V A L V E 1 O . . T H I N W A L L E O O P E N R . C - R O C K C O R E M  M E C H A N I C A L A N A L Y S I S U < U N C O H F I N E O C O M P R E S S I O N OC. . T R I A X I A L C O N S O L I D A T E D Q U I C K O . T R I A X I A L Q U I C K S - T R I A X I A L S L O W SOIL PROFILE O E S C R I P T I O N «*V On. S A N D } m*<tium\ tome fine Bond f sitt it ceasing *lt£ content Htih increos ing dmf>tt> SANO« fine* (tome, airo' .iifi&atian) S> I — T , vtrcct#ied\ • o c o o ' f'ooArf P^ostitUty 281 I G E O C O N I O F F I C E R E P O R T O N ' S O I L EXPLORATION C O N T R A C T i i T 7 _ _ 3 r " - * i 5 ! S t _ BOHINO s X '. • O R I M O DATE _A«SlV_fi^ lOjS>«. REPORT DATE JtV« IR..ia*< " W W I " * » " " W T . J « S L B S . O R O P _ a . ? _ I N C H C S SAMPLE CONDITION DISTURBED ' ESJFAIR P l o o o o ••LOST O A T U M G g O O g p c ; C A S I N O " V " •f- C O M P I L E D B Y _ . V > ~ C H E C K E D B r j S f / J / . . . . « P E N E T R A T I O N R E S I S T A N C E S C O N V E R T E D T O B L O W S O P 4 2 0 0 I N . L B S . E M * * / SAMPLE TYPES ' If "l.U™1^,11i t * . S . - F O I L S A M P L E 8 .T. • SLOTTED T U B E S.O. • B L E E V E * O P E N oi-S ,^l»5S.?iM"-e - S L E E V E T O O " V A L V E D.O. - O R I V C O P E N T.O. -TH IN W A L L E D O P E N O.F. .DRIVE-FOOT V A L V E R-JC. - R O C K C O R E C * - C H U N K S A M P L E • ABBREVIATIONS V • IN-SITU*VANE TEST 1. WET UNIT Wcrcnr M • MECHANICAL ANALYSIS K • PERMEABILITY U .UNCONPINEO COMPRESSION C . CONSOLIDATION OC..TRIAXIAL CONSOLIOATEO OUICK O .TRIAXIAL.OUICK WL • WATER LCVEt S ..TRIAXIAL SLOW WT . WATER TABLE SOIL PROFILE D E S C R I P T I O N ^l»_T f i n * fc» nXujMI^ i.»\lj S'.ll J Pin. S«LTT -*1«-«ltft«^  f f i o 3 < ^ ' s a t . WATER CONTENT W* M ML-L. 282 .[GEOCON.I * O F F I C E R E P O R T O N S O I L EXPLORATION C O N T R A C T . !«n.T>*!!W«Ar_J B O R I N G S 3 ^ O A T U M _.5*P^ «*«=.T.*£r . S O * ( N O O A T * 3 r * L Af9ft>J|%J>U R E P O R T O A T C . M t * V T — A . . . A ^ J I W * . . C O M P I L E D B Y . . . C H E C K E D B Y S A M P L E R H A M M E R W T . . . L B S . O R O P . ..-59™ I N C H E S ( P E N E T R A T I O N R E S I S T A N C E S C O N V E R T E D T O B L O W S O F 4 2 0 0 I N • I B S . E N E R G Y ) S A M P L E C O N D I T I O N S A M P L E T Y P E S . A B B R E V I A T I O N S A . S . - A U G E R S A M P L E WA... F O I L S A M P L E V . I N V I T O V A N E T E S T - » . W E T U N I T W E I G H T J D I S T U R B E D J F A I R loooo • L O S T S . T . • S L O T T E D T U B E W J . * W A S H E D S A M P L E O . O . - D R I V E - O P E N O . F , - D R I V E - F O O T V A L V E C A . C H U N K S A M P L E V J O . * S L E E V E 4 > P C N S . F . . S L E E V E - F O O T V A L V E T X > . . T H I N W A L L E D O P E N njC. ' R O C K C O R E M . M E C H A N I C A L A N A L Y S I S U > U N C O N F I N E O C O M P R E S S I O N O C . T R I A X I A L C O N S O L I D A T E D Q U I C K O - T R I A X I A L O U I C K 8 . T R I A X I A L S L O W W L • W A T E R L E V E L IN C * S I M C W T - W A T E R T A B L E IN S O I L S O I L P R O F I L E D E S C R I P T I O N f M : i . l i l ! t n i . l l l l l l l l l l l l ! I I I I I M I . I H I i ! l , W A T E R C O N T E N T XI L w Pw i!: i . i n i •. 111111 .TTTT O T H E R T E S T S O V N A M I C . P E N E T R A T I O N T E S T B L O W S P E R F O O T H Z * S A K O - tint (, medio***. u < p r « oo m.—«Ilj h> * « — ) " S e a l t t r e d 5 Q - S E H O O f I t 13 19 j 13 20 .283 O F F I C E : R E P O R T O N S O I L E X P L O R A T I O N - D A T U M * GjLOJO£._rj.C_. _ C O M . P U . E O B Y _J&JA'-S*. B O R I N G D A T S __&___._A.PJt_9_i-_ R E P O R T D A T E LifeftL. J£ft__J„S«,_ S A M P L E R H A M M E R W T . L4f i_-_VBS. D R O P ! _ & _ „ t N C H E S l t P E N E T R A T I O N ' R E S I S T A N C E S C O N V E R T E O . T O B L O W S O P A 2 0 0 IN - L B S . E N E R G Y I _ 1 . . C A S I N O . . & .\..A-C H E C K E D B Y jGTpJ-. SAMPLE CONDITION . O f S T U R B E O I •'AIR '. G O O O L O S T SAMPLE TYPES A S . • A U O E R S A M P L E S.T. • S L O T T E O T U B E W J . > W A S H E D S A M P L E . O A V t O R I V E - O P E N P J S . ' * P O I L S A M P L E S . O . i S L E E V E - O P E N %JT. - S L E E V E - T O O T V A L V E TJO. - T H I N W A L L E D O P E N * R X . • R O C K C O R E A B B R E V I A T I O N S V . IN-SITU V A N E T E S T * - W E T U N I T W E I G H T ti *- M E C H A N I C A L A N A L Y S I S K • P E R M E A B I L I T Y U . U N C O N F I N E O C O M P R E S S I O N • C • C O N S O L I D A T I O N O C • T R I A X I A L C O N S O L I D A T E D O U I C K * * K*T*.»«ttO »•* T U O C O • T R I A X I A L O U I C K - W L • W A T E R L E V E L IN C A S I N G 8 * T R I A X I A L S L O W W T • W A T E R T A B L E IN S O I L L - ATTCROtJtCt U** i*T»>-284 OFFICE REPORT ©N SOIL EXPLORATION CONTRACT . 11.77 : T » « « V . . . B O R I N G c ... . 6 OATUM ...&«•?_<>«:T?.«% CASING C " i *»•" «>•••«. BORING OATE _<^ P•i,te-?.:?_-!?3*_ REPORT OATC .. .*>rS.»T...5«5! . A?** . .COMPILED DY_ti»« CHECKCO BT ...CT/-// SAMPLER HAMMER WT. ,|*?>P_„. LBS. OROP. 3 0 , .INCHES I PENETRATION RESISTANCES CONVERTED TO BLOWS O F ' I Z O O IN - LBS. ENERGVI SAMPLE CONDITION O I S T U R B E D P A I R eooo L O S T " SAMPLE TYPES A . S . - A U G E R S A M P L E F . S - . F O I L S A M P L E S.T. - S L O T T E D T U B E S O . . S L E E V E O P E N W . S . • W A S H E D S A M P L E S . F . - S L E E V E - F O O T V A L V E OX}.-ORIVf-OPCN TJO. - T H I N W A L L E O O P E N O . F . - D R I V E - F O O T V A L V E R . C . . R O C K C O R E C S . ' C H U N K S A M P L E ABBREVIATIONS V . iN-arru V A N E T E S T • • 1. W E T U N I T W E I G H T M . M E C H A N I C A L A N A L V 8 I S K . P E R M E A B I L I T Y U • U N C O N F I N E D C O M P R E S S I O N C - C O N S O L I D A T I O N O C . T R 1 A X I A L C O N S O L I D A T E D Q U I C K O - T R I A X I A L Q U I C K W L . W A T E R L E V E L I N C A S I N G S - T R I A X I A L S L O W W T . W A T E R T A B L E I N S O I L SOIL PROFILE D E S C R I P T I O N If f I S A N O - f i r t C T o KO 31 Z o j 3 + i 285 IOEOCOMI OFFICE REPORT ON SOIL EXPLORATION . DATUM G U B b T C O N T R A C T l£T7 B O R I N G D A T E J t f e J ^ B X i f c j a f e f t _ R E P O R T D A T E AWIIL. - S o t3*A C O M P I U C O B Y B ) K , S A M P L E R H A M M E R W T . _M0 L B S . D R O P S O I N C H E S « P E N E T R A T I O N R E S I S T A N C E S C O N V E R T E D T O S L O W S o r « O 0 IH . L B S . E N E R G Y } S A M P L E C O N D I T I O N S A M P L E T Y P E S A . S . • A U G E R S A M P L E TS. . T O I L S A M P L E S.T. . 8 L O T T E O T U B E I X ) . . S L E E V C ^ O P E N W A . W A S H E D S A M P L E S.P. . S L E E V E - F O O T V A L V E • C O . . O R I V E O P E N T A . . T H I N W A L L E O O P E N C P . . O R I V E - P O O T V A L V E B . C . • R O C K C O R E C S . . C H U N K S A M P L E A B B R E V I A T I O N S V . I N . S t T U V A N E T E S T I . WET UNIT WEIGHT M . MECHANICAL ANALYSIS K . PERMEABILITY U - UNCONPINEO COMPRESSION C - CONSOLIDATION OC. • TRIAXIAL CONSOLIDATED QUICK O -TRIAXIAL OUICK WL - W A I ER LKVEL IN C A S i M S . TRIAXIAL SLOW WT . WATER TABLE IN SOIL S O I L P R O F I L E R -Mr DESCRIPTION I I I I I I I ; I I I I H I I : I I I I I I I : I I ; H ' ! | | — r O N A T . "I-" i*! 8 ! i m i ; • 19°, 1111 • i ; i j i mmm m m t t h u j ' fin* */-_>. ^ i V w / . ' r V V ^ MP/*** M-•*itt m*ict »<>W **+n ?+actr cJby ftm.tr tfaimk 4 n k m l . m i J l c i ^ x ' r ; ; - - , i i t * OTHER TESTS O i i h f i l e M _3 inn !_3 -*r» 13 ••/ft3 ISA J ' , 34 23 19 30 13 10 10 12. to 286 OFFICE REPORT ON SOIL EXPLORATION C O N T R A C T \U7 TIINNEL. B O R I N G *T . " & ; ..1 (<28*P?-f2^ATUM GEQPE.UC__ C A S I N G f . P l P E O O R t N O O A T E ^ l ^ C M . i ^ 7 J 5 S 6 R C I K ) R T O A T C ^ O - J 4 l ^ ^ . W 5 * > C O M P U X O B Y ..WW C H E C K E D B Y . &*TiW-... S A M P L E R H A M M E R W T . JiBi.. L B S . D R O P $0. I N C H E S | P E N E T R A T I O N R E S I S T A N C E S C O N V E R T E D T O B L O W S O P 4 » O Q IN * L B S . E H E R G i I S A M P L E C O N D I T I O N S A M P L E T Y P E S AM. *AUGER SAMPLE PA. . FOIL SAMPLE DISTURBED S.T. .SLOTTED TUBE WA.> WASHED SAMPLE O.O. * ORIVE>OPEN o.r. .ORivE-rooT VALVE CA. .CHUNK SAMPLE S O . - S L E E V E O P E N S.P. - S L E E V E - T O O T V A L V E T.O. . T H I N W A L L E O O P E N R . C • R O C K C O R E A B B R E V I A T I O N S V • IN-SITU V A N E T E S T W E T U N I T W E I G H T M . M E C H A N I C A L A N A L Y S I S K • P E R M E A B I L W Y U • U N C O N P I N C O C O M P R E S S I O N C • C O N S O L I D A T I O N QC. • TRI A X I A L C O N S O L I D A T E D O U t C K O - TRI A X I A L O U I C K W L - W A T E R L C V f c L IN C A S I N G 8 . T R 1 A X I A L S L O W W T . W A T E R T A B L E IN S O I L 287 OFFICE REPORT ON SOIL EXPLORATION C O N T R A C T ' 1 1 * 1 , 7 T * " ~ ~ C > - . B O R I N G » . . . . 9 O A T U M Q C o O E p C C A S I N G « o ^ S O R I N G D A T E A < - a a . O A T«"SO R E P O R T D A T E . . H A X . . A S , . i S S C . . C O M P I L E D B Y . . . . W l C H E C K E D B Y < 3 7 / V S A M P L E R H A M M E R W T . .\4.a . . _ L B S . D R O P S o . . I N C H E S I P E N E T R A T I O N R E S I S T A N C E S C O N V E R T E D T O B L O W S O r 4 2 0 0 I N . L B S . E N E R G Y 288 C O N T R A C T •?.7* 7...TW~ , S 4.?:V_._ B O R I N G tt . » 0 . BATUM..Gcpper<C ' ' C A S I N G J**"" B O R I N G D A T E Af_R<k W - H .ItjS*. R E P O R T O A T C . . f c K y . ] q , " , ! j * a 3 < A . . . . C O M P I L C O B Y H j N _11 . C H E C K E D B Y C _ 7"*V S A M P L E R H A M M E R W T . . . - * Q L B S . D R O P 3-.0... I N C H E S * P E N E T R A T I O N R E S I S T A N C E S C O N V E R T E O T O B L O W S O F A 2 0 0 IN - L B S E N E R G Y S A M P L E CONDITION Lj_J D I S T U R B E D [- f IFAIR r~~)0O;3O • • L O S T S A M P L E T Y P E S A . S . • A U G E R S A M P L E P _ . - F O I L S A M P L E S.T. . S L O T T E D T U B E B.O. • G L C E V E - O P E N W . S . - W A S H E D S A M P L E 9JT. . S L E C V E - T O O T V A L V E OJO. . D R I V E - O P E N . T . O . - T H I N W A L L E D O P E N O . F . . O R I V E - P O O T V A L V E . R . C - R O C K C O R E C . S . - C H U N K S A M P L E A B B R E V I A T I O N S V - • IN-SITU V A N E T E S T 1 . W E T U N I T W E I G H T M • M E C H A N I C A L A N A L Y S I S K - P E R M E A B I L I T Y U • U N C O N P I N E O C O M P R E S S I O N C - C O N S O L I D A T I O N QC. . T R I A X I A L C O N S O L I D A T E D O U I C K Q . T R I A X I A L Q U I C K W L - W A T E R L E V E L IN C A S E S . T R I A X I A L S L O W W T . W A T E R T A B L E IN S O I L SOIL PROFILE D E S C R I P T I O N ..I I • StlrolifxtJ w'.tti with a«p*h • H . . l l . l i ! . M . I I I I I . ! l l . i . I W A T E R C O N T E N T W j t r o n I I • • ? • ' I M M . . : D Y N A M I C P E N E T R A T I O N T E S T B L O W S P E R F O O T . — O N A T . t I L w A Pw 8 0 . t o o • 1:11114.11 • j 111 •. fi^ flilSSJpffla ifi*J with 5 A N O - Cl mm 'If+rnftt I i i H i , - ; . a * ! L+tuuH ,1 Hirt 1111 i m m m , l l , = 1 ! rjrrtrmr t r K u r i i J . i i i ni:;jj.|j..t 1: : T-p^i-uIT It} : f -I ! J 'If i -••'•••rlht f UiJ.lli.1,, •it i i l f O T H E R T E S T S S A M P L E S 9" V y t-T.O T . o . I " I " T o . 3 " £ 0 I - n ! _* \ * * ! 1 I z o j j G O 4 0 . «1 I I S sz t o Z S 21 Z S 2-J Z 4 1 fa Z S IB z«. Z 9 23. 28 0 Z S O 3 c 289 APPENDIX G.2 Fine-Grained Soils Data — — BC Hydro Transmission Tower Study (1991) Table Test Information and Summary of Test R e s u l t s C y c l i c Test Data Post C y c l i c Monotonic Test No. Water Content % °vo or ° d c y / 2 o 3 c or N £ i cl-V ° 3 c or e ^ e a k or ° 3 c kPa T c y / o v o imax % s u / o vo fpeak % HSS2 41.5 43.6 34.2 37.0 80^ 80 0.185 0.210 20 12 6.5 5.0 0.335 27.1 20.5 HCT1 - 37.6 so' 0.213 20 4.0 0.456 16.0 Remarks C y c l i c SimpJU^_hear C y c l i c Simple Shear C y c l i c T r i a x i a l ° 3 c T = i n i t i a l water content = water content at end of c o n s o l i d a t i o n = v e r t i c a l e f f e c t i v e c o n s o l i d a t i o n s t r e s s i n s imple shear t e s t = h y d r o s t a t i c c o n s o l i d a t i o n s t r e s s i n t r i a x i a l t e s t = c y c l i c shear s t r e s s i n simple shear ° d c y ^ 2 o 3 c = c 7 c l i c s h e a r s t r e s s i n t r i a x i a l N = n o . o f c y c l e s t o . f a i l u r e (very la rge s t r a i n ) or c y c l e s p r i o r to post c y c l i c monotonic load ing 'max = maximum amplitude of shear s t r a i n dur ing c y c l i c load ing i n s imple shear e1 = maximum amplitude of a x i a l s t r a i n dur ing c y c l i c load ing i n t r i a x i a l '''peak = P e a ^ shear s t r a i n dur ing monotonic load ing i n simple shear e 1 peak = peak a x i a l s t r a i n dur ing monotonic load ing i n t r i a x i a l S.. = peak undrained shear s t r e s s (undrained strength) T . „ „ v i n s imple shear i n t r i a x i a l max = 1/2 o dmax 291 CD 292 >-• ZD r — co CC LU CO LO X CO I CO ZD CO LU CC CD 1—1 Q < o »—1 CO CO LO z a § a. >-x c S . > m 0 co o Q_ cc < LU X CO LU _I Q_ CO U PQ m m X +^ <L> H 60 G c-3 O b O -*-» CO O PM C°dH) S53yiS UV3HS CN 0 I 0 293 (V co co W # E-» CO CO CO w CC « LU . ° ^  1 i—i CO ^ 2 o >-o co co z CC o O a O cc S o CD o o I H CO o w w CO CO _I 1 I L a ro CDdH) SS3cUS dV3HS 3 00 V 

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.831.1-0050338/manifest

Comment

Related Items